Classifications

• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
  • G06V10/00—Arrangements for image or video recognition or understanding
  • G06V10/40—Extraction of image or video features
  • G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
  • G06V10/443—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering
  • G06V10/449—Biologically inspired filters, e.g. difference of Gaussians [DoG] or Gabor filters
  • G06V10/451—Biologically inspired filters, e.g. difference of Gaussians [DoG] or Gabor filters with interaction between the filter responses, e.g. cortical complex cells
  • G06V10/454—Integrating the filters into a hierarchical structure, e.g. convolutional neural networks [CNN]
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06F—ELECTRIC DIGITAL DATA PROCESSING
  • G06F18/00—Pattern recognition
  • G06F18/20—Analysing
  • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
  • G06F18/217—Validation; Performance evaluation; Active pattern learning techniques
  • G06F18/2193—Validation; Performance evaluation; Active pattern learning techniques based on specific statistical tests
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06F—ELECTRIC DIGITAL DATA PROCESSING
  • G06F18/00—Pattern recognition
  • G06F18/20—Analysing
  • G06F18/24—Classification techniques
  • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
  • G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
  • G06F18/24133—Distances to prototypes
  • G06F18/24137—Distances to cluster centroïds
  • G06F18/2414—Smoothing the distance, e.g. radial basis function networks [RBFN]
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
  • G06N3/00—Computing arrangements based on biological models
  • G06N3/02—Neural networks
  • G06N3/08—Learning methods
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T7/00—Image analysis
  • G06T7/0002—Inspection of images, e.g. flaw detection
  • G06T7/0012—Biomedical image inspection
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T7/00—Image analysis
  • G06T7/10—Segmentation; Edge detection
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T7/00—Image analysis
  • G06T7/60—Analysis of geometric attributes
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
  • G06V10/00—Arrangements for image or video recognition or understanding
  • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
  • G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T2207/00—Indexing scheme for image analysis or image enhancement
  • G06T2207/10—Image acquisition modality
  • G06T2207/10072—Tomographic images
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T2207/00—Indexing scheme for image analysis or image enhancement
  • G06T2207/10—Image acquisition modality
  • G06T2207/10116—X-ray image
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T2207/00—Indexing scheme for image analysis or image enhancement
  • G06T2207/20—Special algorithmic details
  • G06T2207/20081—Training; Learning
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T2207/00—Indexing scheme for image analysis or image enhancement
  • G06T2207/20—Special algorithmic details
  • G06T2207/20084—Artificial neural networks [ANN]
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T2207/00—Indexing scheme for image analysis or image enhancement
  • G06T2207/30—Subject of image; Context of image processing
  • G06T2207/30004—Biomedical image processing
  • G06T2207/30096—Tumor; Lesion
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T2207/00—Indexing scheme for image analysis or image enhancement
  • G06T2207/30—Subject of image; Context of image processing
  • G06T2207/30004—Biomedical image processing
  • G06T2207/30101—Blood vessel; Artery; Vein; Vascular
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
  • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
  • G06V2201/03—Recognition of patterns in medical or anatomical images
  • G06V2201/032—Recognition of patterns in medical or anatomical images of protuberances, polyps nodules, etc.

Definitions

• the present disclosure related to quantitative imaging and analytics. More
• the present disclosure relates to systems and methods for analyzing pathologies utilizing quantitative imaging.
• Imaging particularly with safe and non-invasive methods, represents the most powerful methods for locating the disease origin, capturing its detailed pathology, directing therapy, and monitoring progression to health. Imaging is also an extremely valuable and low- cost method to mitigate these human and financial costs by allowing for appropriate early interventions that are both less expensive and disruptive.
• Imaging is especially valuable because it provides spatially- and temporally- localized anatomic and functional information, using non- or minimally invasive methods.
• techniques to effectively utilize increasing spatial and temporal resolution are needed, both to exploit patterns or signatures in the data not readily assessed with the human eye as well as to manage the large magnitude of data in such a way as to efficiently integrate it into the clinical workflow.
• the clinician has neither the time nor often the ability to effectively extract the information content which is available, and in any case generally interprets the information subjectively and qualitatively. Integrating quantitative imaging for individual patient management as well as clinical trials for therapy development requires a new class of decision support informatics tools to enable the medical community to fully exploit the capabilities possible with evolving and growing imaging modalities within the realities of existing work flows and reimbursement constraints.
• Quantitative results from imaging methods have the potential to be used as biomarkers in both routine clinical care and in clinical trials, for example, in accordance with the widely accepted NIH Consensus Conference definition of a biomarker.
• quantitative imaging is intended to (a) detect and characterize disease, before, during or after a course of therapy, and (b) predict the course of disease, with or without therapy.
• imaging biomarkers may be used in defining endpoints of clinical trials. Quantification builds on imaging physics developments which have resulted in improvements of spatial, temporal, and contrast resolution as well as the ability to excite tissues with multiple energies/sequences, yielding diverse tissue-specific responses.
• CT computed tomography
• DECT dual energy computed tomography
• spectral CT computed tomography
• CTA computed tomography angiography
• CCTA cardiac computed tomography angiography
• MRI magnetic resonance imaging
• multi-contrast magnetic resonance imaging multi-contrast MRI
• US ultrasound
• targeted or general contrast agent approaches with various imaging modalities.
• a scanner combined with image processing of the formed images has the ability to measure
• Imaging in the early phases of clinical testing of novel therapeutics contributes to the understanding of underlying biological pathways and pharmacological effects. It may also reduce the cost and time needed to develop novel pharmaceuticals and therapeutics.
• imaging biomarkers may serve as important endpoints for clinical benefit and/or as companion diagnostics to help prescribe and/or follow specific patient conditions for personalized therapy.
• imaging biomarkers may be used to select or stratify patients based on disease status, in order to better demonstrate therapeutic effect.
• Imaging is especially valuable because it provides spatially- and temporally- localized anatomic and functional information, using non- or minimally invasive methods.
• techniques to deal with increasing resolution are needed, both to exploit patterns or signatures in the data not readily assessed with the human eye as well as to manage the large magnitude of data in such a way as to efficiently integrate it into the clinical workflow.
• the radiologist would“drown” in data. Integrating quantitative imaging for individual patient management will require a new class of decision support informatics tools to enable the community to fully exploit the capabilities of these new tools within the realities of existing work flows and reimbursement constraints.
• Imaging in the early phases of clinical testing of novel therapeutics contributes to the understanding of underlying biological pathways and pharmacological effects. It may also reduce the cost and time needed to develop novel pharmaceuticals and therapeutics. In later phases of
• imaging biomarkers may serve as important endpoints for clinical benefit. In all phases, imaging biomarkers may be used to select or stratify patients based on disease status, in order to better demonstrate therapeutic effect.
• Improved patient selection through use of quantitative imaging could reduce the required sample size for a given trial (by increasing the fraction of evaluable patients and/or decreasing the impact of nuisance variables) and help to identify the sub-population that could benefit most from the proposed treatment. This should reduce development time and cost for new drugs, but might also result in reducing the size of the‘target’ population accordingly.
• Radiological imaging is generally interpreted subjectively and qualitatively for medical conditions.
• the medical literature uses the term phenotype as the set of observable characteristics of an individual resulting from the interaction of its genotype with the environment. Phenotype generally implies objectivity, namely that the phenotype may be said as being true rather than subjective. Radiology is well known for its ability to visualize characteristics, and increasingly it may be validated to objective truth standards (U.S.
• phenotyping has a truth basis and therefore can be independently and objectively evaluated. Furthermore, phenotyping is already an accepted formalism in the healthcare industry for managing patient therapeutic decisions. Thus, it has a high degree of clinical relevance. Finally, phenotyping is consumer relatable. This allows for both self- advocacy as well as serving as motivator for lifestyle changes.
• Imaging phenotypes may be correlated gene-expression patterns in association studies. This may have a clinical impact as imaging is routinely used in clinical practice, providing an unprecedented opportunity to improve decision-support in personalized treatments at low cost. Correlating specific imaging phenotypes with large-scale genomic, proteomic, transcriptomic, and/or metabolomic analyses has potential to impact therapy strategies by creating more deterministic and patient-specific prognostics as well as measurements of response to drug or other therapy. Methods for extracting imaging phenotypes to date, however, are mostly empirical in nature, and primarily based on human, albeit expert, observations. These methods have embedded human variability, and are clearly not scalable to underpin high-throughput analysis.
• phenotype classification included clinical relevance is provided below:
• Active disease may have high response to intensified lipid lowering and/or anti-inflammatory regime
• Such plaques have been reported to have 3x-4x the adverse event rate compared to Stable phenotypes. These two examples may be assessed at a single patient encounter, but other phenotypes such as“Rapid Progressors” may be determined by comparing the rate of change in characteristics over time, i.e., phenotypes informed by not only what is statically present at one point in time but rather which are determined based on kinetics and/or how things are changing n time.
• Deep Learning (DL) methods have been applied, with large success, to a number of difficult Machine Learning (ML) and classification tasks stemming from complex real-life problems.
• Notable recent applications include computer vision (e.g., optical character recognition, facial recognition, interpretation of satellite imagery, etc.), speech recognition, natural language processing, medical images analysis (image segmentation, feature extraction, and classification), clinical and molecular data biomarkers discovery and verification.
• An attractive feature of the approach is its ability to be applied to both unsupervised and supervised learning tasks.
• Neural Networks and the Deep NN approach, broadly including convolutional neural networks (CNNs), recurrent convolutional neural networks (RCNNs), etc., have been shown to be based on a sound theoretical foundation, and are broadly modelled after principles believed to represent the high-level cognition functions of the human brain.
• CNNs convolutional neural networks
• RCNNs recurrent convolutional neural networks
• the sensory signals propagate thru a complex local, modular hierarchy which learns to represent observations over time - a design principle that has led to the general definition and construction of CNNs used for i.e. image classification and feature extraction.
• it has been somewhat of an undecided question what are the more fundamental reasons for the superior performance of the DL networks and approach compared to frameworks with the same number of fitting parameters but without the deep layered architecture.
• FFR plaque reserve
• HRFl plaque stability or hgh-risk plaque
• ACS acute coronary syndrome
• Heart health can be significantly impacted by the degradation of arteries surrounding the heart.
• a variety of factors tissue characteristics such as angiogenesis, neovascularization, inflammation, calcification, lipid-deposits, necrosis, hemorrhage, rigidity, density, stenosis, dilation, remodeling ratio, ulceration, flow (e.g., of blood in channel), pressure (e.g., of blood in channel or one tissue pressing against another), cell types (e.g., macrophages), cell alignment (e.g., of smooth muscle cells), or shear stress (e.g., of blood in channel), cap thickness, and/or tortuosity (e.g., entrance and exit angles) by way of example) may cause these arteries to reduce their effectiveness in transmitting oxygen filled blood to the surrounding tissue (Figure 35).
• tissue characteristics such as angiogenesis, neovascularization, inflammation, calcification, lipid-deposits, necrosis, hemorrhage, rigidity, density, stenosis
• Functional testing of the coronary arteries mainly stress-echocardiography and single photon emission computed tomography myocardial perfusion testing (SPECT MPI) is currently the dominant noninvasive method for diagnosing obstructive coronary artery disease.
• SPECT MPI single photon emission computed tomography myocardial perfusion testing
• FFR fractional flow reserve
• flow reserve applies to perfusion of brain tissues as well (e.g., as related to hyperemia in the brain).
• Functional testing has known issues with sensitivity and specificity. It is estimated that some 30-50% of cardiovascular disease patients are misclassified and are over-treated or under-treated with significant monetary and quality of life costs. Functional testing is also expensive, time consuming and of limited use with patients that have pre-obstructive disease. False positives from non-invasive function tests are a large factor in the overuse of both invasive coronary angiography and percutaneous coronary interventions in stable patients that are a major policy issue in the U.S., Great Britain and China. Studies of the impact of false negatives estimate that of 3.8M annual MPI tests given to U.S.
• Ischemia is a lagging indicator that follows the anatomical changes brought on by disease progression. Patients at high risk for ACS would also be better served if future culprit lesions can be detected and reduced with intensive drug therapy prior to the onset of ischemia.
• Coronary computed tomography angiography is evolving into an ideal testing modality to fill this gap in understanding the extent and rate of progression coronary artery disease.
• CT scanner fleet in most countries has been upgraded to higher-speed, higher detector count machines capable of excellent spatial resolution without slowing of the heart or extensive breath-holds.
• Radiation dose has been greatly lowered, to the point where it is equivalent or lower than SPECT MPI and invasive angiography.
• the recent favorable determination from NICE positioning CCTA as a front-line diagnostic is based on a significant reduction in CV events in the CCTA arm of the SCOT-HEART study that was attributed to drug treatment initiation or changes on discovery of plaques with CCTA.
• An important target patient group are those with stable chest pain and no prior history of CAD with typical or atypical anginal symptoms (based on SCOT-HEART data), and those with those with non-obstructive disease ( ⁇ 70% stenosis) and in younger patients (e.g., 50-65 years group), based on the PROMISE findings that suggest that assessment of plaque is most needed.
• CAD found with high-risk plaque profile based on CCTA analysis can be assigned to most appropriate high intensity statin therapy (particularly when a decision on new lipid-lowing therapies that are very expensive such as PCSK9 inhibitors, or anti-inflammatory drugs such as canakinumab are considered), or add a new antiplatelet agent to mitigate the risk of coronary thrombosis, and/or longitudinal follow-up for possible intensification or downgrading of the therapies.
• CCTA is an ideal diagnostic tool as it is noninvasive and requires less radiation than cardiac catheterization.
• IVUS intra-vascular ultrasound
• NIRS near-infra red spectroscopy
• OCT optical coherence tomography
• CCTA cerebral spastic characterization and assessment of the entire coronary tree, instead of just the culprit lesion, to allow more accurate risk stratification, which suitably analyzed CCTA can do.
• CCTA had good diagnostic accuracy to detect coronary plaques compared to IVUS with small differences in assessment of plaque area and volume, percent area stenosis, and slight overestimation of lumen area. Adding rate of change of lipid rich necrotic core and its distance from the lumen was also found to have high prognostic value.
• results from the ROMICAT II Trial show that identifying high-risk plaque on CCTA for stable CAD patients with acute chest pain but negative initial ECG and troponin increases the likelihood of ACS independent of significant CAD and clinical risk assessment.
• Atherosclerosis develops progressively through evolution of arterial wall lesions resulting from the accumulation of cholesterol-rich lipids and inflammatory response. These changes similar (even if not identical) in the coronary arteries, the carotid arteries, aorta, and peripheral arteries. Certain plaque characteristics such as large atheromatous core with lipid- rich content, thin cap, outward remodeling, infiltration of the plaque with macrophages and lymphocytes and thinning of the media are predisposing to vulnerability and rupture.
• Non-invasive assessment of the functional significance of stenosis using CCTA is of clinical and financial interest.
• the combination of a lesion or vessel’s geometry or anatomic structure together with the characteristics or composition of the tissue comprising the walls and/or plaque in the walls, collectively referred to as plaque morphology, may explain outcomes in lesions with higher or lower risk plaque (HRP) and or the orthogonal
• FFR normal and abnormal flow reserve
• plaque morphology increases accuracy by evaluating complex factors such as LRNC, calcification, hemorrhage, and ulceration with an objective truth that may be used to validate the underlying information, in a manner that other approaches cannot due to the lack of an intermediate measurement objective validation.
• plaque morphology also identifies and quantifies these high-risk plaques. For example, plaque morphology may be used to determine how close the necrotic core is to the lumen: a key determinant of infarction risk. Knowing whether a lesion limits flow under stress doesn’t indicate the risk of rupture or vice versa. Other methods such as computational fluid dynamics (CFD), without objectively validated plaque morphology, can simulate flow limitation but not infarction risk.
• CFD computational fluid dynamics
• CFD-based flow reserve considers the lumen only, or at best, how the luminal surface changes at different parts of the cardiac cycle. Considering only the luminal surface at best requires processing both systolic and diastolic to get motion vector (which isn’t even done by most available methods), but even that does not consider what occurs at stress, because these analyses are done with computer simulations of what may happen under stress rather than measuring actual stress, and not based on the actual characteristics that originate vasodilatory capacity but rather just the blood channel. Some methods attempt to simulate forces and apply biomechanical models, but with assumptions rather than validated measurements of wall tissues.
• Systems and methods are provided herein which utilize a hierarchical analytics framework to identify and quantify biological properties/analytes from imaging data and then identify and characterize one or more medical conditions based on the quantified biological properties/analytes.
• the systems and methods incorporate
• the systems and methods of the present disclosure may advantageously implement biological and clinical insights in advanced computational models. These models may then interface with sophisticated image processing through rich ontologies that specify factors associated with the growing understanding of pathogenesis and takes the form of rigorous definitions of what is being measured, how it is measured and assessed, and how it is relates to clinically-relevant subtypes and stages of disease that may be validated.
• the hierarchical inference scheme ensures the clinical inferences can be understood, validated, and explained at each level in the hierarchy.
• systems and methods of the present disclosure utilize a hierarchical analytics framework comprised of a first level of algorithms which measure biological properties capable of being objectively validated against a truth standard independent of imaging, followed by a second set of algorithms to determine medical or clinical conditions based on the measured biological properties.
• This framework is applicable to a number of distinct biological properties in an“and/or” fashion, i.e., singly or in combination, such as angiogenesis, neovascularization, inflammation, calcification, lipid- deposits, necrosis, hemorrhage, rigidity, density, stenosis, dilation, remodeling ratio, ulceration, flow (e.g., of blood in channel), pressure (e.g., of blood in channel or one tissue pressing against another), cell types (e.g., macrophages), cell alignment (e.g., of smooth muscle cells), or shear stress (e.g., of blood in channel), cap thickness, and/or tortuosity (e.g., entrance and exit angles) by way of examples.
• angiogenesis e.vascularization
• inflammation calcification
• lipid- deposits e.g., necrosis, hemorrhage
• rigidity e.g., density
• stenosis e.g., dilation
• remodeling ratio e
• Measurands for each of these may be measured, such as quantity and/or degree and/or character, of the property.
• Example conditions include perfusion/ischemia (e.g., as limited) (e.g., of brain or heart tissue), perfusion/infarction (as cut off completely) (e.g., of brain or heart tissue), oxygenation, metabolism, flow reserve (ability to perfuse), malignancy, encroachment, and/or risk stratification (whether as probability of event, or time to event (TTE)) e.g., major adverse cardio- or cerebrovascular events (MACCE).
• perfusion/ischemia e.g., as limited
• perfusion/infarction as cut off completely
• oxygenation e.g., of brain or heart tissue
• oxygenation e.g., oxygenation, metabolism, flow reserve (ability to perfuse)
• malignancy e.g., major adverse cardio- or cerebrovascular events (MACCE).
• MACCE major adverse cardio- or cerebrovascular events
• Truth bases may include, for example, biopsy, expert tissue annotations form excised tissue (e.g., endarterectomy or autopsy), expert phenotype annotations on excised tissue (e.g., endarterectomy or autopsy), physical pressure wire, other imaging modalities, physiological monitoring (e.g., ECG, Sa02, etc.), genomic and/or proteomic and/or metabolomics and/or transcriptomic assay, and/or clinical outcomes. These properties and/or conditions may be assessed at a given point in time and/or change across time (longitudinal).
• the systems and methods of the subject application advantageously relate to computer-aided phenotyping (CAP) of disease.
• CAP is a new and exciting complement to the field of computer-aided diagnosis (CAD).
• CAD computer-aided diagnosis
• CAP may apply a hierarchical inference incorporating computerized image analysis and data fusion algorithms to patient clinical chemistry and blood biomarker data to provide a multi factorial panel or“profile” of measurements that may be used to distinguish between different subtypes of disease that would be treated differently.
• CAP implements new approaches to robust feature extraction, data integration, and scalable computational strategies to implement clinical decision support. For example, spatio-temporal texture (SpTeT) method captures relevant statistical features for characterizing tissue spatially as well as kinetically.
• SpTeT spatio-temporal texture
• Spatial features map, for example, to characteristic patterns of lipid intermixed with extracellular matrix fibers, necrotic tissue, and/or inflammatory cells.
• Kinetic features map, for example, to endothelial permeability, neovascularization, necrosis, and/or collagen breakdown.
• the systems and methods of the subject application may advantageously utilize a hierarchical inference scheme may be applied beginning with not only spatially-resolved image features but also time-resolved kinetics at multiple levels of biologically-objective components of morphology and structural composition in the middle, and then clinical inference at the end.
• a hierarchical inference scheme may be applied beginning with not only spatially-resolved image features but also time-resolved kinetics at multiple levels of biologically-objective components of morphology and structural composition in the middle, and then clinical inference at the end.
• Phenotype classification may occur at two levels, individual anatomic locations on the one hand and more generally described body sites on the other.
• the input data for the former may be 2D data sets, and the input data for the latter may be 3D data sets.
• phenotype classification objective truth may be at either level
• outcome prediction / risk stratification generally occurs at the patient level, but can be more specific in certain instances (e.g., which side did stroke symptoms manifest on).
• the implication here is that the same input data may be used for both purposes, but the models will differ substantially because of the level at which the input data will be used as well as the basis of the truth annotations.
• DL Deep Learning
• the systems and methods of the subject application may also advantageously utilize an unsupervised learning application, thereby providing for better scalability across data domains (a very highly desirable feature having in mind the pace at which new biomedical data is generated).
• CNNs Convolutional neural networks
• CNNs may be utilized for building a classifier in an approach that can be characterized as transfer learning with fine-tuning approach.
• CNNs trained on a large compendium of imaging data on a powerful computational platform can be used, with a good success, to classify images that have not been annotated in the network training. This is intuitively understandable, as many common classes of features help identify images of vastly different objects (i.e. shapes, boundaries, orientation in space, etc.).
• CNNs trained to recognize thousands of different objects using pre-annotated datasets of tens of millions of images would perform basic image recognition tasks much better than chance, and would have a comparable performance to CNNs trained from scratch after a relatively minor tweaking of the last classification layer, sometimes referred to as the‘softmax layer. Since these models are very large and have been trained on a huge number of pre-annotated images, they tend‘to learn’ very distinctive, discriminative imaging features.
• the convolutional layers can be used as a feature extractor or the already trained convolutional layers can be tweaked to suit the problem at hand.
• the first approach is referred to as transfer learning and the latter as fine- tuning.
• CNNs are excellent at performing many different computer vision tasks. CNNs have a few drawbacks however. Two important drawbacks of importance to medical systems are 1) the need for vast training and validation datasets, and 2) intermediate CNN computations are not representative of any measurable property (sometimes criticized as being a“black box” with undescribed rationale).
• the approaches disclosed herein may advantageously utilize a pipeline consisting of one or more stages which are separately biologically measurable and capable of independent validation, followed by a convolutional neural network starting from these rather than only raw imagery.
• certain transforms may be applied to reduce variation that does not relate to the problem at hand, such as for example unwrapping a donut- shaped vessel cross section to become a rectangular representation with a normalized coordinate system prior to feeding the network.
• CNNs for medical imaging.
• the early convolutional layers act as feature extractor of increasing specificity
• the fully connected one or two last layers act as cclassifiers (e.g.,“softmax layers”).
• cclassifiers e.g.,“softmax layers”.
• the systems and methods of the present disclosure utilize enriched data sets to enable non-invasive phenotyping of tissues assayed by radiological data sets.
• One type of enrichment is to pre-process the data to perform tissue classification and use“false color” overlays to provide a data set that can be objective validated (as opposed to only using raw imagery, which does not have this possibility).
• Another type of enrichment is to use transformations on the coordinate system, to accentuate biologically-plausible spatial context while removing noise variation to either improve the classification accuracy, allow for smaller training sets, or both.
• the systems and methods of the subject application may employ a multi-stage pipeline: (i) semantic segmentation to identify and classify regions of interest (e.g., which may be representative of quantitative biological analytes) (ii) spatial unwrapping to convert cross-sections of a tubular structure (e.g., a vein/artery cross section) into rectangles, and (iii) application of a trained CNN to read the annotated rectangles and identify which phenotype (e.g., stable or unstable plaque and/or normal or abnormal FFR) it pertains to, and/or predicted time to event (TTE).
• semantic segmentation to identify and classify regions of interest (e.g., which may be representative of quantitative biological analytes)
• spatial unwrapping to convert cross-sections of a tubular structure (e.g., a vein/artery cross section) into rectangles
• a trained CNN to read the annotated rectangles and identify which phenotype (e.g., stable or unstable plaque and/or normal or abnormal
• untransformed datasets e.g., datasets that are not spatially unwrapped, for example
• phenotype e.g., either in conjunction with or independent of untransformed datasets
• semantic segmentation and spatial transformation may involve the following:
• the image volume may be preprocessed including target initialization, normalization, and any other desired pre-pressing such as deblurring or restoring to form a region of interest containing a physiological target that is to be phenotyped.
• said region of interest may be a volume composed of cross sections through that volume is the body site may be either automatically determined or is provided explicitly by user.
• Targets for body sites that are tubular in nature may be accompanied with a centerline. Centerlines, when present, can branch. Branches can be labelled either automatically or by user. Note that generalizations on the centerline concept may be represented for anatomy that is not tubular but which benefit by some structural directionality, e.g., regions of a tumor.
• centroid may be determined for each cross section in the volume.
• this may be the center of the channel, e.g., the lumen of a vessel.
• lesions this may be the center of mass of the tumor.
• the (optionally deblurred or restored) image may be represented in a Cartesian data set where x is used to represent how far from centroid, y represents a rotational theta, and z represents the cross section.
• Cartesian set will be formed per branch or region.
• a“null” value may be used for overlapping regions, that is, each physical voxel may be represented only once across the sets, in such a way as to geometrically fit together.
• Each data set can be paired with an additional data set with sub-regions labelled by objectively verifiable tissue composition.
• Example labels for vascular tissue can be lumen, calcification, LRNC, P ⁇ , etc.
• Example labels for lesions could be necrotic, neovascularized, etc. These labels can be validated objectively, e.g. by histology. Paired data sets may can used as input to a training step to build a convolutional neural network. In example embodiments, two levels of analysis can be supported, one at an individual cross-section level, and a second at the volume level. Output labels represent phenotype or risk stratification.
• Exemplary image pre-processing may include deblurring or restoring using, for example, a patient-specific point spread determination algorithm to mitigate artifacts or image limitations that result from the image formation process. These artifacts and image limitations may decrease the ability to determine characteristics predictive of the phenotype. Deblurring or restoring may be achieved as a result of, for example, iteratively fitting a physical model of the scanner point spread function with regularizing assumptions about the true latent density of different regions of the image.
• the CNN may be AlexNet, Inception, CaffeNet, or other networks.
• refactoring may be done to the CNN, e.g., where a same number and type of layers are used, but the input and output dimensions are changed (such as to change the aspect ratio).
• Example implementations of various example CNNs are provided as open source on, for example, TensorFlow, and/or in other frameworks, available as open source and/or licensed configurations.
• the dataset may be augmented.
• 2D or 3D rotations may be applied to the dataset.
• augmentation may involve, e.g., randomly horizontally flipping the dataset in conjunction with randomly rotating the data set (such as by a random angle between 0 and 360 degrees).
• augmentation may involve e.g., randomly horizontally flipping in conjunction with a random“scrolling” of the image such as by a random number of pixels in the range from 0 to the width of the image (where scrolling akin to rotating around theta).
• the dataset may be enriched by using different colors to represent different tissue analyte types. These colors may be selected to visually contrast relative to each other, as well as relative to a non-analyte surface (e.g., normal wall). In some embodiments, a non-analyte surface may be depicted in grey.
• dataset enrichment may result in ground truth annotation of tissue characteristics (e.g., tissue characteristics that are indicative of plaque phenotype) as well as provide a spatial context of how such tissue characteristics present in cross section (e.g., such as taken orthogonal to an axis of the vessel).
• Such spatial context may include a coordinate system (e.g., based on polar coordinates relative to a centroid of the cross-section) which provides a common basis for analysis of dataset relative to histological cross sections.
• a coordinate system e.g., based on polar coordinates relative to a centroid of the cross-section
• enriched datasets may be advantageously overlaid on top of color-coded pathologist annotations (or vice versa).
• a histology based annotated dataset may then be used for training (e.g., training of the CNN) in conjunction with or independent of image feature analysis of a radiological dataset.
• a histology based annotated dataset may improve efficiency in DL approaches since the histology based annotated dataset uses a relatively simpler false color image in place of a higher-resolution full image without losing spatial context.
• coordinate directions may be internally represented using unit phasors and phasor angle.
• the coordinate system may be normalized, e.g., by normalizing the radial coordinate with respect to wall thickness (such as to provide a common basis for comparing tubular structures/cross-sections of different diameters/thicknesses).
• a normalized radial distance may have a value of 0 at an inner (inner wall luminal boundary) and value of 1 at an outer boundary (outer wall boundary).
• this may be applied to tubular structures relevant to vascular or other pathophysiology (e.g., the gastro- intestinal tract).
• the enriched datasets of the subject application provide for in vivo non-invasive image-based classification (e.g., where a tissue classification scheme can be used to determine phenotype non-invasively) which is based on a known ground truth.
• the known ground truth may be non-radiological (such as histology or another ex vivo based tissue analysis).
• radiology datasets annotated to include ex vivo ground truth data may be advantageously used as input data for the classifier.
• a plurality of different known ground truths may be used in conjunction with one another or independent of one another in annotating an enriched dataset.
• an enriched dataset may utilize a normalized coordinate system to avoid non-relevant variation associated with, for example, the wall thickness and radial presentation.
• a “donut” shaped dataset may be“unwrapped,” e.g., prior to classification training (e.g., using a CNN) and/or prior to running a training classifier on the dataset.
• analyte annotation of the training dataset may be prior to transformation, e.g., after unwrapping or a combination of both.
• an untransformed dataset may be annotated (e.g., using ex vivo classification data such as histology information) and then transformed for classifier training.
• ex vivo classification data such as histology information
• a finer granularity of ex vivo based classification may be collapsed to match a lower intended granularity for in vivo radiology analysis to not only decrease computational complexity but simulataneously address what would otherwise be open to criticism of being a“black box”.
• colors and/or axes for visualizing the annotated radiological dataset may be selected to correspond to the same colors/axes as typically presented in ex vivo ground truth-based classifications (e.g., same colors/axes as used in histology).
• a trasnformed enriched dataset e.g., which may be normalized for wall thickness
• each analyte is visually represented by a different contrasting color and relative to a background region (e.g., black or grey) for all non analyte regions.
• the common background may or may not be annotated and therefore may or may not visually differentiate between non-analyte regions in and out of the vessel wall or between background features (such as luminal surface irregularity, varying wall thickness, etc.).
• annotated analyte regions e.g., color coded and normalized for wall thickness
• a uniform background e.g., completely black, completely gray, completely white, etc.
• annotated analyte regions may be visually depicted relative to an annotated background (e.g., where different shades (grey, black and/or white) may be used to distinguish between (i) a center lumen region inside the inner lumen of the tubular structure, (ii) non-analyte regions inside the wall and/or (iii) a region outside the outer wall. This may enable analysis of variations of wall thickness (e.g., due to ulceration or thrombus).
• the annotated dataset may include, e.g., identification of (and visualization of) regions of intra plaque hemorrhage and/or other morphology aspects. For example, regions of intra-plaque hemorrhage may be visualized in red, LRNC in yellow, etc.
• One specific implementation of the systems and methods of the subject application may be in directing vascular therapy.
• Classifications may be established according to a likely dynamic behavior of a plaque lesion (based on its physical characteristics or specific mechanisms e.g. inflammatory or cholesterol metabolism based) and/or based on a progression of the disease (e.g., early vs late in its natural history). Such classifications may be used for directing patient treatment.
• the Stary plaque typing system adopted by the AHA may be utilized as an underlay with in vivo determined types shown in color overlays .
• the systems and methods of the present disclosure are not, however, tied to Stary. as another example, the Virmani system ['Calcified nodule', 'CTO', 'FA', 'FCP', 'Healed
• the systems and methods of the present disclosure may provide for phenotype classification of a plaque based on an enriched radiological data set.
• the phenotype classification(s) may include distinguishing stable plaque from unstable plaque, e.g., where the ground truth basis for the classification is based on factors such as (i) luminal narrowing (possibly augmented by additional measures such as tortuosity and/or ulceration), (ii) calcium content (possibly augmented by depth, shape, and/or other complex presentations), (iii) lipid content (possibly augmented by measures of cap thickness and/or other complex presentations), (iv) anatomic structure or geometry, and/or (v) IPH or other content.
• this classification has been demonstrated to have high overall accuracy, sensitivity and specificity as well as a high degree of clinical relevance (with potential to change existing standard of care of patients who are undergoing catheterization and cardiovascular care).
• lung cancer where the subtypes of masses may be determined so as to direct the most likely beneficial treatment for the patient based on the manifest phenotype.
• pre-processing and dataset enrichment may be used to separate out into solid vs. semi-solid (“ground glass”) sub regions, which differ both in degree of malignancy as well as suggesting differing optimal treatments.
• the systems and methods of the present disclosure may provide for image pre-processing, image de-noise and novel geometric representation (e.g., an affine transformation) of CT angiography (CTA) diagnostic images to facilitate and maximize performance of Deep Learning Algorithms based on CNNs for developing best-of- class classifier and a marker of risk of adverse cardiovascular effects during representation procedures.
• image deblurring or restoring may be used to identify lesions of interest and extract plaque composition quantitatively.
• transformation of cross-sectional (along the main axis blood vessel) segmented images into, for example, an ‘unwrapped’ rectangular reference frame which follows the established lumen along the X axis may be applied to provide a normalized frame to allow DL approaches to best learn representative features.
• example embodiments herein utilize 2-D annotated cross-sections for analysis and phenotyping
• the subject application is not limited to such embodiments.
• some embodiments may utilize enriched 3-D datasets, e.g., instead of or in addition to processing 2-D cross-sections separately.
• video interpretation from computer vision may be applied for the classifier input data set.
• processing multiple cross-sections sequentially, as if in a“movie” sequence along a centerline can generalize these methods for tubular structures, e.g., moving up and down a center-line, and/or other 3D manifestations depending on the aspects most suited to the anatomy.
• false color representations in the enriched data set may have continuous values across pixel or voxel locations. This can be used for,
• each pixel/voxel can have values for any number of different features (e.g., can be represented in any number of different overlays for different analytes, sometimes referred to as“multiple occupancy”).
• each pixel/voxel may only be assigned to one analyte (e.g., assigned to only a single analyte overlay).
• the pixel/voxel value for a given analyte can be based on an all or nothing classification scheme (e.g., either the pixel is calcium or it isn’t).
• the pixel/voxel value for a given analyte can be a relative value, e.g., a probability score).
• the relative values for a pixel/voxel are normalized across a set of analytes (e.g., so that the total probability adds up to 100%).
• classification models may be trained in whole or in part by application of multi-scale modeling techniques, such as for example partial differential equations, e.g., to represent likely cell signaling pathways or plausible biologically-motivated presentations.
• change data for example as collected from multiple timepoints, rather than (only) data from a single timepoint.
• change data for example as collected from multiple timepoints, rather than (only) data from a single timepoint.
• change data for example as collected from multiple timepoints, rather than (only) data from a single timepoint.
• a“progressor” phenotype vs. a“regressor” phenotype for decreases.
• the regressor might be, for example, due to response to a drug.
• the rate of change for, say, LRNC is rapid, this may imply a different phenotype, e.g., a“rapid progressor”.
• non-spatial information such as which are derived from other assays (e.g., lab results), or demographics/risk factors, or other measurements taken from the radiological image, may be fed into the final layers of the CNN to combine the spatial information with non-spatial information.
• classifications may be based on ground truth historical outcomes assigned to training data sets. For example, life expectancy, quality of life, treatment efficacy (including comparing different treatment methods), and other outcome predictions can be determined using the systems and methods of the subject application.
• the systems and methods of the present disclosure provide for the determination of fractional flow reserve in myocardial and/or brain tissue by measurement of plaque morphology.
• Systems and methods of the present disclosure may use sophisticated methods to characterize the vasodilatory capacity of vessels via objectively validated determination of tissue type and character which impact its distensibility.
• plaque morphology may be used as input to analysis of dynamic behavior of the vasculature from a flow reserve point of view (training the models with flow reserve truth data).
• Stenosis itself is well known as being of low predictive power in that it only provides a static description; addition of accurate plaque morphology is necessary, for the highest accuracy imaging-based assessment of dynamic function.
• the present disclosure provides systems and methods which determine accurate plaque morphology and then processes such to determine the dynamic function.
• deep learning is utilized to retain the spatial context of tissue characteristics and vessel anatomy (collectively referred to as plaque morphology) at an optimal level of granularity, avoiding excessive non-material variability in the training sets while retaining that which is needed to exceed other more simplistic use of machine learning.
• tissue characteristics do so without objective validation as to their accuracy and/or without the data enrichment methods needed to retain spatial context optimally for medical image deep learning (e.g., transformation such as unwrapping and the validated false color tissue type overlays) necessary to provide the effectiveness of deep learning methods.
• Some methods try to increase training set size by use of synthetic data, but this is ultimately limited to the limited data on which the synthetic generation as based and amounts more to a data augmentation scheme rather than a real expansion of the input training set.
• the systems and methods of the present disclosure are able to create continuous assessments across vessel lengths. The systems and methods of the present disclosure effectively leverage objective tissue characterization validated by histology across multiple arterial beds.
• plaque composition is similar in coronary and carotid arteries, irrespective of its age, and this will largely determine relative stability, suggesting similar presentation at CCTA as at CTA. Minor differences in the extent of the various plaque features may include a thicker cap and a higher prevalence of intraplaque hemorrhage and calcified nodules in the carotid arteries, however, without difference in the nature of plaque components.
• the carotid and coronary arteries have many similarities in the physiology of vascular tone regulation that has effect on plaque evolution.
• Myocardial blood perfusion is regulated by the vasodilation of epicardial coronary arteries in response to a variety of stimuli such as NO, causing dynamic changes in coronary arterial tone that can lead to multifold changes in blood flow.
• carotid arteries are more than simple conduits supporting the brain circulation; they demonstrate vasoreactive properties in response to stimuli, including shear stress changes.
• Endothelial shear stress contributes to endothelial health and a favorable vascular wall transcriptomic profile.
• Clinical studies have demonstrated that areas of low endothelial shear stress are associated with atherosclerosis development and high-risk plaque features.
• lower wall shear stress is associated with plaque development and localization.
• IVUS VH is a method of spectral backscatter analysis that enables plaque composition analysis (and thus
• CCTA scans have been described to be in the range of 300-400 pm resolution. While this already puts CCTA resolution very close to that of IVUS VH, there are additional factors specific to the present invention analysis to consider.
• the systems and methods of the present disclosure perform an iterative deblurring or restoring modeling step with sub-voxel precision, e.g., using a tessellated surface of triangles to represent the true surface of lipid core.
• Lipid core areas are in the range of 6.5-l4.3mm2 in 393 patients (corresponding radius of curvature of 1.4-2.
• the present disclosure is also innovative in dealing with fundamental limitations of the application of artificial intelligence and deep learning to the analysis of atherosclerosis imaging data.
• Conventional competitive approaches that lack a validated and objective basis such as are fundamentally limited in multiple ways.
• Second, this lack of objectivity increases the demands for large volumes of data on which to circumstantially base correlations. This imposes infeasible demands for manual annotation of the radiological images which are themselves the subject of analysis (that is, as opposed to being validated from an independent modality).
• the present invention overcomes these drawbacks by using a pipeline consisting of one or more stages which are biologically measurable (that is, capable of being objectively validated), followed by smaller scope convolutional neural network processing on these validated biological properties to output the desired output conditions not based on subjective or qualitative“imaging features” but rather.
• These architectural capabilities mitigate drawbacks by increasing the efficiency of available training data and make the intermediate steps capable of being objectively validated.
• the systems and methods of the present disclosure simultaneously alleviate drawbacks of using CNNs for medical imaging by 1) reducing the complexity of the vision task to within levels that are acceptable when training a CNN with a moderately sized dataset, and 2) producing intermediate outputs which are both objectively validated and easily interpretable by users or regulating bodies.
• Figure 1 depicts a schematic of an exemplary system for determining
• Figure 2 outlines a re-sampling based model building approach, according to the present disclosure which may be implemented by the systems and methods described herein.
• Figure 3 depicts a sample patient report, according to the present disclosure which may be outputted by the systems and methods described herein.
• Figure 4 depicts example segmentation levels for a multi-scale vessel wall analyte map, according to the present disclosure.
• Figure 5 depicts an exemplary pixel-level probability mass function as a set of analyte probability vectors, according to the present disclosure.
• Figure 6 illustrates a technique for computing putative analyte blobs, according to the present disclosure.
• Figure 7 depicts normalized vessel wall coordinates for an exemplary vessel wall composition model, according to the present disclosure.
• Figure 8 depicts an example margin between plaque removed for a histology specimen and the outer vessel wall, according to the present disclosure.
• Figure 9 illustrates some complex vessel topologies which can be accounted for using the techniques described herein, according to the present disclosure.
• Figure 10 depicts representing an exemplary analyte blob with a distribution of normalized vessel wall coordinates, according to the present disclosure.
• Figure 11 depicts an exemplary distribution of blog descriptors, according to the present disclosure.
• Figure 12 depicts an exemplary model for imaging data correlating between a hidden ground truth state and an observed state, according to the present disclosure.
• Figure 13 depicts a diagram of an example Markov model/Viterbi algorithm for relating an observed state to a hidden state in an image model, according to the present disclosure.
• Figure 14 depicts an example frequency distribution of total number of blobs per histological slide for a plurality of histological slides, according to the present disclosure.
• Figure 15 depicts exemplary implantation of a 1D Markov chain, according to the present disclosure.
• Figure 16 depicts an example first order Markov chain for a text probability table, according to the present disclosure.
• Figure 17 depicts conditional dependence of a first pixel based on its neighboring pixels, according to the present disclosure.
• Figure 18 depicts a further exemplary hierarchical analytics framework according to the present disclosure.
• Figure 19 depicts an example application of phenotyping purpose in directing vascular therapy, according to the present disclosure.
• the depicted example uses the Stary plaque typing system adopted by the AHA as an underlay with in vivo determined types shown in color overlays.
• Figure 20 depicts an example application of phenotyping for lung cancer, according to the present disclosure.
• Figure 21 illustrates an exemplary image pre-processing step of deblurring or restoring, according to the present disclosure.
• the deblurring or restoring uses a patient- specific point spread determination algorithm to mitigate artifacts or image limitations that result from the image formation process that may decrease the ability to determine characteristics predictive of the phenotype.
• the deblurred or restored (processed) images depicted (bottom) derive from CT imaging of a plaque (top) and are a result of iteratively fitting a physical model of the scanner point spread function with regularizing assumptions about the true latent density of different regions of the image.
• Figure 22 depicts an exemplary application demonstrating an enriched dataset for atherosclerotic plaque phenotype, according to the present disclosure.
• Figure 23 illustrates tangential and radial direction variable internally represented using unit phasors and phasor angle (shown coded in grey scale) which exemplifies the use of normalized axes for tubular structures relevant to the vascular and other pathophysiology associated with such structures (e.g., the gastro-intestinal tract), according to the present disclosure.
• Figure 24 illustrates an exemplary overlay of radiology analysis application generated annotations from CTA (outlined regions) on top of pathologist generated annotations from histology (solid regions), according to the present disclosure.
• Figure 25 demonstrates a further step of data enrichment, specifically, utilizing the normalized coordinate system to avoid non-relevant variation associated with the wall thickness and radial presentation, according to the present disclosure. Specifically, the “donut” is“unwrapped” while retaining the pathologist annotations.
• Figure 26 represents a further step of data enrichment relevant to plaque phenotyping, according to the present disclosure.
• luminal irregularity for example from ulceration or thrombus
• local varying wall thickening is represented.
• Figure 27 represents data enriched imaging (in both wrapped and unwrapped forms) including intra-plaque hemorrhage and/or other morphology, according to the present disclosure.
• Figure 28 represents validation results for algorithms trained and validated with different variations of data enriched imaging, according to the present disclosure.
• Figure 29 provides an example application of phenotyping lung lesions, according to the present disclosure.
• Figure 30 provides an example of biological properties (including e.g., tissue characteristics, morphology, etc.) for phenotyping lung lesions, according to the present disclosure. Note that in example embodiments, false colors may be represented as continuous values rather than discrete values with respect to one or more of these biological properties.
• Figure 31 illustrates a high-level view of one example approach to user interaction with a computer-aided phenotyping system, according to the present disclosure.
• Figure 32 illustrates example system architectures, according to the present disclosure.
• Figure 33 illustrates components of example system architectures, according to the present disclosure.
• Figure 34 is a block diagram of example data analysis according to the present disclosure. Images are collected of the patient, the raw slice data is used in a set of algorithms to measure biological properties that may be objectively validated, these are in turn formed as enriched data sets to feed one of more CNNs, in this example where results are forward and back-propagated using recurrent CNNs to implement constraints or creates continuous conditions (such as a monotonically decreasing fractional flow reserve from proximal to distal throughout the vessel tree, or constant HRP value in a focal lesion, or other constraints).
• Figure 35 illustrates causality of coronary ischemia and available diagnostics, according to the present disclosure.
• Figure 36 depicts exemplary 3D segmentations of lumen, vessel wall, and plaque components (LRNC and calcification) for two patients presenting with chest pain and similar risk factors and degree of stenosis, according to the present disclosure.
• LRNC and calcification exemplary 3D segmentations of lumen, vessel wall, and plaque components
• Figure 37 depicts example histology processing steps for objective validation of tissue composition, according to the present disclosure.
• Figure 38 is a schematic depiction of the multiple cross-sections that may be processed to provide a dynamic analysis across the entire scanned vessel tree, illustrating the
• systems and methods for analyzing pathologies utilizing quantitative imaging are presented herein.
• the systems and methods of the present disclosure utilize a hierarchical analytics framework that identifies and quantify biological properties/analytes from imaging data and then identifies and characterizes one or more pathologies based on the quantified biological properties/analytes.
• This hierarchical approach of using imaging to examine underlying biology as an intermediary to assessing pathology provides many analytic and processing advantages over systems and methods that are configured to directly determine and characterize pathology from raw imaging data without the validation steps and/or without advantageous processing described herein.
• One advantage is the ability to utilize training sets from non-radiological sources, e.g., from tissue sample sources such as histological information, in conjunction with or independent of training sets for radiological sources, to correlate radiological imaging features to biological properties/analytes to pathologies.
• histology information may be used in training algorithms for identifying and characterizing one or more pathologies based on quantified biological properties/analytes. More
• biological properties/analytes which are identifiable/quantifiable in non- radiological data may also be identified and quantified in radiological data (which is advantageously non-invasive).
• These biological properties/analytes may then be correlated to clinical findings on pathology using information the from non-radiological sources, for example, utilizing histological information, gene expression profiling, or other clinically rich data sets.
• This set of clinically correlated data may then serve as a training set or part of a training set for determining/tuning (e.g., utilizing machine learning) algorithms correlating biological properties/analytes to pathologies with a known relationship to clinical outcome.
• These algorithms correlating biological properties/analytes to pathologies derived utilizing non-radiological source training sets may then be applied in evaluating biological
• the systems and methods of the present disclosure may advantageously enable utilizing radiological imaging (which may advantageously be cost-effective and non-invasive) to provide surrogate measures for predicting clinical outcome or guiding treatment.
• training data for non-radiological sources may be more accurate/reliable than training data for radiological sources.
• training data from non-radiological sources may be used to augment training data from radiological sources.
• the hierarchical analytics framework disclosed advantageously improves the trainability and resulting reliability of the algorithms disclosed herein.
• one key advantage is that, once trained the systems and methods of the present disclosure may enable deriving comparable clinical information to existing histological and other non-radiological diagnostic-type testing without the need not undergo invasive and/or costly procedures.
• training sets for non-radiological sources may be utilized in conjunction with or independent of training sets for radiological sources, e.g., in correlating image features to biological properties/analytes.
• non-radiological imaging sources e.g., histological sources, and/or non-imaging sources
• training sets for radiological sources may be utilized in conjunction with or independent of training sets for radiological sources, e.g., in correlating image features to biological properties/analytes.
• one or more biological models may be extrapolated and fitted to correlate radiological and non-radiological data.
• histology information may be correlated with radiological information based on an underlying biological model. This correlation may advantageously enable training recognition of biological properties/analytes in radiological data utilizing non-radiological, e.g., histological information.
• data drawn from complementary modalities may be used, e.g., in correlating image features to biological properties/analytes from blood panels, physical FFR, and/or other sources of data.
• one or more biological models may be extrapolated and fitted utilizing imaging data drawn from one imaging modality either correlated with and/or fused with another imaging modality or non-imaging source such as bloodwork. These biological models may advantageously correlate across and between imaging and non imaging data sets based on the biological models. Thus, these biological models may enable the hierarchical analytics framework to utilize data from one imaging modality with another imaging modality or with a non-imaging source in identifying/quantifying one or more biological properties/analytes or identifying/characterizing one or more medical conditions.
• Another advantage to the hierarchical analytics framework disclosed herein is the ability to incorporate data from multiple same or different type data sources into the process of identifying and characterizing pathology based on imaging data.
• one or more non-imaging data sources may be used in conjunction with one or more imaging data sources in identifying and quantifying a set of biological
• the set of biological properties/analytes may include one or more biological properties/analytes identified and/or quantified based on one or more imaging data sources, one or more biological properties/analytes identified and/or quantified based on one or more non-imaging data sources, and/or one or more biological
• the set of biological properties/analytes may generally include at least one or more biological properties/ anal ytes identified and/or quantified based at least in part on an imaging data).
• the ability to augment information from an imaging data source with information from other imaging and/or non-imaging data sources in identifying and quantifying a set of biological properties/analytes adds to the robustness of the systems and methods presented herein and enables utilization of any and all relevant information in identifying and characterizing pathology.
• Yet another advantage of the hierarchical analytics framework involves the ability to adjust / fine-tune data at each level, e.g., prior or subsequent to utilizing that data to assess the subsequent level (note that in some embodiments this may be an iterative process).
• information related to a set of identified and quantified biological properties/analytes may be adjusted in an a posteriori manner (e.g., after an initial identification and/or quantification thereof).
• information related to a set of identified and characterized pathologies may be adjusted in an a posteriori manner (e.g., after an initial identification and/or characterization thereof).
• These adjustments may be automatic or user based and may objective or subjective.
• the ability to adjust / fine- tune data at each level may advantageously improve data accountability and reliability.
• adjustments may be based on contextual information, which may be used to update one or more probabilities impacting a determination or quantification of a biological property/analyte.
• contextual information for adjusting information related to a set of identified and quantified biological
• properties/ anal ytes in an a posteriori manner may include patient demographics, correlations between biological properties/analytes or correlations between identified/characterized pathologies and biological properties/analytes.
• the biological properties/ anal ytes may be related in the sense that the identification/quantification of a first biological property/analyte may impact a probability relating the identification/quantification of a second biological property/analyte.
• identification/characterization of a first pathology e.g., based on an initial set of identified/quantified biological
• properties/ anal ytes may impact a probability relating to the identification/quantification of a biological property/analyte in the initial set or even a biological property/analyte that wasn’t in the first set.
• pathologies may be related, e.g., wherein
• identification/characterization of a first pathology may impact a probability relating the identification/characterization of a first pathology.
• information related to identification and quantification of biological properties/analytes and/or information related to the identification and characterization of pathologies may be updated in an iterative manner, e.g., until data convergence or thresholds/benchmarks are achieved or for a selected number of cycles.
• a further advantage of the hierarchical analytics framework involves the ability to provide a user, e.g., a physician, with information relating both to a pathology as well as the underlying biology.
• This added context may facilitate clinical diagnosis / evaluation as well as assessing/determining next steps, e.g., therapeutic/treatment options or further diagnostics.
• the systems and methods may be configured to determine which biological parameters/analytes relevant to the identification/quantification of one or more pathologies are most indeterminate / have the highest degree of uncertainty (e.g., by reason of lack of data or conflicting data). In such instances, specific further diagnostics may be recommended.
• the added context of providing a user with information relating both to a pathology as well as the underlying biology may further help the user evaluate / error check various the clinical conclusions and recommendations reached by the analytics.
• a hierarchical analytics framework refers to an analytic framework wherein a one or more intermediary sets of data points are utilized as an intermediary processing layer or an intermediary transformation between initial set of data points and an end set of data points. This is similar to the concept of deep learning or hierarchical learning wherein algorithms are used to model higher level abstractions using multiple processing layers or otherwise utilizing multiple transformations such as multiple non-linear
• the hierarchical analytics framework of the systems and methods of the present disclosure includes data points relating to biological properties/analytes as an intermediary processing layer or intermediary transformation between imaging data points and pathology data points, in example, embodiments, multiple processing layers or multiple transformation (e.g., as embodied by multiple levels of data points) may be included for determining each of imaging information, underlying biological information and pathology information.
• multiple processing layers or multiple transformation e.g., as embodied by multiple levels of data points
• multiple processing layers or multiple transformation may be included for determining each of imaging information, underlying biological information and pathology information.
• example hierarchical analytic framework structures are introduced herein (e.g., with specific processing layers, transforms and datapoints)
• the systems and methods of the present disclosure are not limited to such implementations. Rather, any number of different types of analytic framework structures may be utilized without departing from the scope and spirit of the present disclosure.
• the hierarchical analytics frameworks of the subject application may be conceptualized as including a logical data layer as an intermediary between an empirical data layer (including imaging data) and a results layer (including pathology information).
• the empirical data layer represents directly sourced data
• the logical data layer advantageously adds a degree of logic and reasoning which distills this raw data into a set of useful analytes for the results layer in question.
• empirical information from diagnostics such as raw imaging information may be advantageously distilled down to a logical information relating to a particular set of biological features which is relevant for assessing a selected pathology or group of pathologies (for example, pathologies related to an imaged region of the patient’s body).
• the biological features/analytes of the subject application can also be thought of as pathology symptom s/indi cator s .
• biological features/analytes of the subject application may at times be referred to herein a biomarkers. While the term“biological” or prefix“bio” is used in characterizing biological features or biomarkers this in only intended to signify that the features or markers have a degree of relevance with respect to the patient’s body.
• biological features may be anatomical, morphological, compositional, functional, chemical, biochemical, physiological, histological, genetic or any number of other types of features related to the patient’s body.
• Example, biological features utilized by specific implementations of the systems and methods of the present disclosure are disclosed herein.
• example systems and methods of the present disclosure may be geared toward detecting, characterizing and treating pathologies/diseases
• the application of the systems and methods of the present disclosure are not limited to pathologies/diseases but rather may more generally applicable with respect to any clinically relevant medical conditions of a patient including, e.g., syndromes, disorders, traumas, allergic reactions, etc.
• the systems and methods of the present disclosure relate to Computer-Aided Phenotyping, e.g., by using knowledge about biology to analyze medical images to measure the differences between disease types that have been determined through research to indicate phenotypes which in turn predict outcomes.
• characterizing pathologies may include determining phenotypes for the pathologies which may in turn determine a predictive outcome.
• the analyzer module 120 advantageously implements a hierarchical analytics framework which first identifies and quantifies biological properties/analytes 130 utilizing a combination of (i) imaging features 122 from one or more acquired images 121 A of a patient 50 and (ii) non-imaging input data 121B for a patient 50 and then identifies and characterizes one or more pathologies (e.g., prognostic phenotypes) 124 based on the quantified biological properties/ anal ytes 123.
• the analyzer module 120 may operate independent of ground truth or validation references by implementing one or more pre-trained, e.g., machine learned algorithms for drawing its inferences.
• the analyzer may include algorithms for calculating imaging features 122 from the acquired images 121A of the patient 50.
• some of the image features 122 may be computed on a per-voxel basis while others may be computed on a region-of-interest basis.
• Example non-imaging inputs 121B which may be utilized along with acquired images 121 A may include data from laboratory systems, patient-reported symptoms, or patient history.
• the image features 122 and non -imaging inputs may be utilized by the analyzer module 120 to calculate the biological properties/analytes 123.
• the biological properties/analytes are typically quantitative, objective properties (e.g., objectively verifiable rather than being stated as impression or appearances) that may represent e.g., a presence and degree of a marker (such as a chemical substance) or other measurements such as structure, size, or anatomic characteristics of region of interest.
• the quantified biological properties/analytes 123 may be displayed or exported for direct consumption by the user, e.g., by a clinician, in addition to or independent of further processing by the analyzer module.
• properties/ anal ytes 123 may be used as inputs for determining phenotype. Phenotypes are typically defined in a disease-specific manner independent of imaging, often being drawn from ex vivo pathophysiological samples for which there is documented relationship to outcome expected. In example embodiments, the analyzer module 120 may also provide predicted outcomes 125 for determined phenotypes.
• analyzer module 120 is further described herein with respect to specific embodiments which follow the general description of the system 100.
• specific imaging features, biological properties/ anal ytes and pathologies/phenotypes are described with respect to specific medical applications such as with respect to the vascular system or with respect to the respiratory system.
• the cohort tool module 130 enables defining a cohort of patients for group analyses thereof, e.g., based on a selected set of criteria related to the cohort study in question.
• An example cohort analysis may be for a group of patients enrolled in a clinical trial, e.g., with the patient’s further being grouped based on one or more arms of the trial for example a treatment vs. control arm.
• Another type of cohort analysis may be for a set of subjects for which ground truth or references exist, and this type of cohort may be further decomposed into a training set or“development” set and a test or“holdout” set.
• Development sets may be supported so as to train 112 the algorithms and models within analyzer module 120, and holdout sets may be supported so as to evaluate/validate 113 the performance of the algorithms or models within analyzer module 120.
• the trainer module 110 may be utilized to train 112 the algorithms and models within analyzer module 120.
• the trainer module 110 may rely on ground truth 111 and/or reference annotations 114 so as to derive weights or models, e.g., according to established machine learning paradigms or by informing algorithm developers.
• classification and regression models are employed which may be highly adaptable, e.g., capable of uncovering complex relationships among the predictors and the response.
• their ability to adapt to the underlying structure within the existing data can enable the models to find patterns that are not reproducible for another sample of subjects.
• Adapting to irreproducible structures within the existing data is commonly known as model over-fitting.
• a systematic approach may be applied that prevents a model from finding spurious structure and enable the end-user to have confidence that the final model will predict new samples with a similar degree of accuracy on the set of data for which the model was evaluated.
• Successive training sets may be utilized to determine optimal tuning parameter(s), and a test set may be utilized to estimate an algorithm’s or model's predictive performance.
• Training sets may be used for training each of the classifiers via randomized cross-validation.
• Datasets may be repeatedly split into training and testing sets and may be used to determine classification performance and model parameters.
• the splitting of the datasets into training and test sets occurs using a stratified or maximum dissimilarity approaches.
• a re-sampling approach e.g. bootstrapping
• FIG. 2 outlines a re-sampling based model building approach 200 which may be utilized by the systems and methods of the present disclosure.
• a tuning parameter set may be defined.
• step 220 for each tuning parameter set data is resampled the model is fitted and hold-out samples are predicted.
• Resampling estimates are combined into a performance profile.
• step 240 final tuning parameters may be determined.
• step 250 the entire training set is re-fitted with the final tuning parameters.
• each model may be evaluated for predictive performance on the test set. Test set evaluation occur once for each model to ensure that the model building process does not over-fit the test set.
• the optimal tuning parameter estimates, the re-sampled training set performance, as well as the test set performance may be reported. The values of the model parameters over randomized splits are then be compared to evaluate model stability and robustness to training data.
• Model responses may include, for example, covariance-based techniques, non-covariance based techniques, and tree based models.
• endpoints may have continuous and categorical responses; some of the techniques in the above categories are used for both categorical and continuous responses, while others are specific to either categorical or continuous responses.
• Optimal tuning parameter estimates, the re-sampled training set performance, as well as the test set performance may be reported for each model.
• the analyzer module 120 may be configured to delineate fields, for example, to register multiple data streams across a field; to segment organs, vessels, lesions and other application-specific objects; and/or to reformat/reconfigure anatomy for specific analyses.
• the analyzer module 120 may further be configured for delineating a target, for example, a lesion, in a delineated field. Delineating a target may, for example, include registering multiple data streams at a locale; conducting fine-grained segmentation; measuring size and/or other characteristics of relevant anatomic structures; and/or extracting whole-target features (e.g., biological properties/analytes characteristic of the entire target region).
• one or more sub-target regions may also be delineated, for example, a target region may be split into sub-targets according to a particular application with sub-target specific calculations (e.g., biological properties/analytes characteristic of a sub-target region).
• the analyzer module 120 may also delineate components or relevant features (such as composition), for example, in a particular field, target or sub-target region. This may include segmenting or re-segmenting the
• the analyzer module 120 may be configured to compare data across multiple timepoints, e.g., one or more of the biological components/analytes may involve a time-based quantification.
• a wide scan field may be utilized to assess multi-focal pathologies, e.g., based on aggregate quantifications of biological properties/analytes across a plurality of targets in the delineated field.
• the analyzer module 120 may be configured to generate a patient report.
• a sample patient report 300 is depicted in Figure 3.
• the sample patient report 300 may include quantifications of biological parameters/analytes such as relating to structure 310 and composition 320 as well as data from non-imaging sources such as hemodynamics 330.
• the sample patient report may further include visualizations 340, e.g., 2D and/or 3D visualizations of imaging data as well as combined visualizations of non imaging data such as hemodynamic data overlaid onto imaging data.
• Various analytics 350 may be displayed for assessing the biological parameters/analytes including, e.g., a visualization of one or more model(s) (e.g., a decision tree model) for determining/ characterizing pathology.
• model(s) e.g., a decision tree model
• Patient background and identifying information may further be included.
• the analyzer module 120 of system 100 may advantageously provide a user, e.g., a clinician with comprehensive feedback for assessing the patient.
• Example vascular and lung applications are described in greater detail in the sections which follow (although it will be appreciated that the specific application described have general implications and interoperability with respect to numerous other applications).
• Table 2 provides an overview of vascular and lung related applications utilizing a hierarchical analytics framework as described herein.
• Anatomic Structure Vessel structural measurements, specifically those that lead to the determination of % stenosis, have long been and remain the single most used measurements in patient care. These were initially limited to inner lumen measurements, rather than wall measurements involving both the inner and outer surfaces of the vessel wall. However, all of the major non-invasive modalities, unlike X-ray angiography, can resolve the vessel wall and with this come expanded measurements that may be achieved. The category is broad and the measurements are of objects of varying sizes, so generalizations should be made with care. A primary consideration is the limit of spatial sampling or resolution. The minimally detectable changes in wall thickness may, however, be lower than the spatial sampling by taking advantage of subtle variations in intensity levels due to partial volume effect. Additionally, stated resolutions generally refer to grid size and field of view of post-acquisition
• in-plane vs. through- plane resolutions may or may not be the same and not only the size of a given feature but as well its proportions and shape will drive the measurement accuracy.
• categorical conclusions are drawn from applying thresholds to the measurements, which may then be interpreted according to signal detection theory with the ability to optimize the trade-off between sensitivity and specificity, terms that don’t otherwise refer to measurements in the normal sense.
• Tissue Characteristics The quantitative assessment of the individual constituent components of the atherosclerotic plaques, including for example lipid rich necrotic core (LRNC), fibrosis, intraplaque hemorrhage (IPH), permeability, and calcification, can provide crucial information concerning the relative structural integrity of the plaque that could aid the physician’s decisions on course of medical or surgical therapy. From the imaging technology point of view, the ability to do this lies less with spatial resolution as with contrast resolution and tissue discrimination made possible by differing tissues responding to incident energy differently so as to produce a differing receive signal.
• LRNC lipid rich necrotic core
• IPH intraplaque hemorrhage
• permeability permeability
• Each imaging modality does this to some extent; terms in ultrasound such as“echolucency”, the CT number in Hounsfield Units, and differentiated MR intensities as a function of various sequences such as (but not limited to) Tl, T2 and T2*.
• Dynamic tissue behavior (e.g., Permeability): In addition to morphological features of the vessel wall/plaque, there is increasing recognition that dynamic features are valuable quantitative indicators of vessel pathology. Dynamic sequences where the acquisition is taken at multiple closely-spaced times (known as phases) expand the repertoire beyond spatially- resolved values t include temporally-resolved values which may be used for compartment modeling or other techniques to determine the tissues’ dynamic response to stimulus (such as but not limited to wash-in and wash-out of contrast agent).
• Hemodynamics The basic hemodynamic parameters of the circulation have a direct effect on the vasculopathy. Blood pressures, blood flow velocity, and vessel wall shear stress may be measured by techniques ranging from very simple oscillometry to sophisticated imaging analysis. Using common principles of fluid dynamics, calculations of vessel wall shear stress can be ascertained for different regions of the wall. In similar fashion MRI, with or without the combination of US, has been used to calculate the wall shear stress (WSS) and correlate the results with structural changes in the vessel of interest. In addition, the effects of antihypertensive drugs on hemodynamics have been followed for short and long-term studies.
• plaque structure may advantageously include, e.g., lumen measurements (which improves stenosis measurement by providing area rather than only diameter measures) as well as wall measurements (e.g., wall thickness and vascular remodeling).
• Evaluating plaque composition may advantageously involve quantification of tissue characteristics (e.g., lipid core, fibrosis, calcification, permeability, etc.) rather than just “soft” or“hard” designations as typically found in the prior art.
• tissue characteristics e.g., lipid core, fibrosis, calcification, permeability, etc.
• Table 3 Structural calculations of vessel anatomy supported by vascular applications of the systems and methods disclosed herein.
• Example systems relating to evaluating the vascular system may advantageously include/employ algorithms for evaluating vascular structure.
• the systems may employ, e.g., a target/vessel segment/cross-section model for segmenting the underlying structure of an imaged vessel.
• a fast-marching competition filter may be applied to separate vessel segments.
• the systems may further be configured to handle vessel bifurcations.
• Image registrations may be applied utilizing Mattes mutual information (MR) or mean square error (CT) metric, rigid versor transform, LBFGSB optimizer, or the like.
• vessel segmentation may advantageously include lumen segmentation.
• An initial lumen segmentation may utilize a confidence connected filter (e.g., carotid, vertebral, femoral, etc.) to distinguish the lumen.
• Lumen segmentation may utilize MR imaging (such as a combination of normalized, e.g., inverted for dark contrast, images) or CT imaging (such as use of registered pre-contrast, post-contrast CT and 2D Gaussian distributions) to define a vessel-ness function.
• MR imaging such as a combination of normalized, e.g., inverted for dark contrast, images
• CT imaging such as use of registered pre-contrast, post-contrast CT and 2D Gaussian distributions
• Various connected components may be analyzed and thresholding may be applied.
• Vessel segmentation may further entail outer wall segmentation (e.g., utilizing a minimum curvature (k2) flow to account for lumen irregularities).
• an edge potential map is calculated as outward-downward gradients in both contrast and non contrast.
• outer wall segmentation may utilize cumulative distribution functions (incorporating prior distributions of wall thickness, e.g., from 1-2 adjoining levels) in a speed function to allow for median thickness in the absence of any other edge information.
• ferret diameters may be employed for vessel characterization.
• wall thickness may be calculated as the sum of the distance to lumen plus the distance to the outer wall.
• lumen and/or wall segmentations may be done using semantic segmentation using, for example, CNNs.
• Example systems relating to evaluating the vascular system may further
• composition may be determined based on image intensity and other image features.
• the lumen shape may be utilized, e.g., as relating to determining thrombosis.
• an analyte blob model may be employed for better analyzing composition of particular sub-regions of the vessel.
• an analyte blob may be a spatially contiguous region, in 2D, 3D, or 4D images, of one class of biological analyte.
• the blob model may utilize an anatomically aligned coordinate system using isocontours, e.g., in normalized radial distance from the lumen surface to the adventitial surface of the vessel wall.
• the model may advantageously identify one or more blobs and analyze each blobs location e.g., with respect to the overall vessel structure as well as relative to other blobs.
• a hybrid Bayesian/Markovian network may be utilized to model a relative location of a blob.
• the model may advantageously account for the observed image intensity at a pixel or voxel being influenced by a local neighborhood of hidden analyte category nodes thereby accounting for partial volume and scanner point spread function (PSF).
• PSF partial volume and scanner point spread function
• the model may further allow for dynamically delineating analyte blob boundaries from analyte probability maps during inference by the analyzer module.
• An iterative inference procedure may be applied that utilizes uses the current estimate of both analyte probability and blob boundaries.
• parametric modeling assumptions or kernel density estimation methods may be used to enable probability density estimates between the sparse data used to train the model.
• the multi-scale model computes the statistics of each contiguous region of a given analyte type, which may be referred to as a ‘blob’.
• blob analyte type
• the wall is defined by two boundaries, the inner boundary with the lumen and the outer boundary of the vessel wall, creating a donut shape in cross section.
• the donut shaped wall region there are a discrete number of blobs (different than the default background class of normal wall tissue which is not considered to be a blob). The number of blobs is modeled as a discrete random variable.
• each blob is assigned a label of analyte type and various shape descriptors are computed. Additionally, blobs are considered pairwise. Finally, within each blob, each pixel can produce a radiological imaging intensity value, which are modeled as independent and identically distributed (i.i.d.) samples that come from a continuously valued distribution specific to each analyte type. Note that in this last step, the parameters of the imaging intensity distributions are not part of the training process.
• This model accounts for the spatial relationship of analyte blobs within the vessel and also to each other, recognizing that point-wise image features (whether from histology and/or radiology) is not the only source of information for experts to determine plaque composition. While the model allows for the ability to train without explicit histology-to-radiology registration, it could also be applied in situations where that registration is known. It is believed that statistically modeling the spatial layout of atherosclerotic plaque components for classifying unseen plaques is a novel concept.
• Example techniques for estimating vessel wall composition from CT or MR images are further elaborated on in the following section.
• the methods may employ a multi-scale Bayesian analytic model.
• the basic Bayesian formulation is as follows:
• the hypothesis may be based on a multi-scale vessel wall analyte map, A, with observation combing from CT or MR image intensity information I.
• the multi-scale vessel wall analyte map may advantageously include wall-level segmentation 410 (e.g., a cross-sectional slice of the vessel), blob-level segmentation and pixel-level segmentation 430 (e.g., based on individual image pixels.
• wall-level segmentation 410 e.g., a cross-sectional slice of the vessel
• blob-level segmentation and pixel-level segmentation 430 e.g., based on individual image pixels.
• B b may be defined as a generic single blob where be[l..n B ] is an index of all blobs in A.
• B b a is a blob with label a.
• the individual blob descriptor operator D B ⁇ ⁇ is in an low- dimensional space.
• C c may be defined as a blob pair where ce[l..n B (n B -l)/2] is an index of all blob pairs in A.
• C c f,s is a blob pair with labels f and g.
• the blob pair descriptor operator Dc ⁇ ⁇ is in a low-dimensional space.
• I(x) is the continuous valued pixel intensity at pixel x.
• I(x) are modeled as independent. Note that because the model is used to classify wall composition in 3D radiological images, the word“pixel” is used to generically denote both 2D pixels and 3D voxels
• Characteristics of blob regions of like composition/structure may advantageously provide insights regarding the disease process.
• Each slice (e.g., cross-sectional slice) of a vessel may advantageously include a plurality of blobs. Relationships between blobs may be evaluated in a pairwise manner. The number of blobs within a cross-section is modeled as a discrete random variable and may also be of quantifiable significance.
• relevant characteristics e.g., biological properties/analytes
• blob level of segmentation characteristics of each blob, such as structural characteristics, e.g., size and shape, as well as compositional characteristics, etc., may be evaluated serving as a biological properties/analytes.
• pixel level analysis may be performed, e.g., based image intensity distribution.
• Probability mapping of characteristics may be applied with respect to the multi-scale vessel wall analyte map depicted in Figure 4.
• the probability mapping may advantageously establish a vector of probabilities for every pixel with components of the vector for the probability of each class of analyte and one component for the probability of background tissue.
• sets of probability vectors may represent mutually exclusive characteristics.
• each set of probability vectors representing mutually exclusive characteristics will sum to 1.
• the probability mapping does not assume independence of analyte class between pixels. This, is because neighboring pixels or pixels within a same blob may typically have same or similar characteristics.
• the probability mapping accounts, as described in greater detail herein, advantageously accounts for dependence between pixels.
• f(A) is the probability distribution function over all vessel walls.
• f(D B ⁇ B a ⁇ ) is the probability density function (pdf) of blob descriptors with label a.
• f(B) f(D B ⁇ B a ⁇ )
• f(D c ⁇ C f g ⁇ ) is the probability density function (pdf) of pairwise blob descriptors.
• PDF probability density function
• P(A(x)) is the probability mass function (pmf) of analytes (prevalence). It can be considered a vector of probabilities at a specific pixel x or as a probability map for a specific class label value.
• f(C c ) is the probability density function (pdf) of pairwise blob descriptors.
• f(B b ) is the probability density function (pdf) of blob descriptors.
• P(A(x)) is the probability mass function (pmf) of analytes (prevalence in a given map). It can be considered a vector of probabilities at a specific pixel x or as a spatial probability map for a specific analyte type.
• P(A(x) a
• P(I(x) i
• Figure 5 depicts an exemplary pixel-level probability mass function as a set of analyte probability vectors.
• the probability mass function Completeness: in example embodiments one may assume a sufficiently small pixel must fall into at least one of the analyte classes (including a catch-all‘background’ category) and thus the sum of probabilities sums to 1.
• Mutual exclusivity a sufficiently small pixel may be assumed to belong to only one class of analyte; if there are combinations (i.e., spiculated calcium on LRNC background), then a new combination class can be created in order to retain mutual exclusivity.
• An alternative view of the analyte map is as a spatial map of probability for a given analyte.
• analyte blobs can be defined using the full width half max rule. Using this rule, for each local maxima of probability for that analyte a region is grown outward to a lower threshold of half the local maxima value. Note that this 50% value is a tunable parameter. Spatial regularization of blobs can be done here by performing some curvature evolution on probability maps in order to keep boundaries more realistic (smooth with few topological holes). Note that different possible putative blobs of different analyte classes may in general have spatial overlap because until one collapses the
• the continuously valued probability maps can be presented directly to the user in one of several forms including but not limited to surface plots, iso-contour plots, or using image fusion similar to visualizing PET values as variation in hue and saturation on top of CT.
• a second alternative is to collapse the probability map at each pixel by choosing a single analyte label for each pixel. This can be done most straightforwardly by choosing the maximum a posteriori value at each pixel independently, thus creating a categorical map which could be visualized by assigning a distinct color to each analyte label and assigning either full or partial opacity on top of the radiological image.
• the label values might be assigned non-independently by resolving overlapping putative blobs based on a priority the probability of each blob. Hence, at a given pixel a lower priority analyte probability might be used for the label if it belongs to a higher probability blob.
• Figure 6 illustrates a technique for computing putative analyte blobs.
• putative blobs may have overlapping regions.
• analytical techniques to segmenting pixels by putative blobs. For a probability of a given analyte the local maxima in probability is determined.
• the full width half max rule may then be applied to determine discrete blobs.
• analyte blobs can be defined using the full width half max rule. Find local maxima, then region grow with a lower threshold of 0.5*max.
• spatial regularization of blobs may also be applied, e.g., by performing some curvature evolution on probability maps in order to keep boundaries smooth and avoid holes.
• an image-level analyte map be computed, e.g., based on a collapse of the probability map function.
• this collapse can be determined based on either the pixel -level analyte probability map, the putative blobs or a combination of both.
• the highest probability for each set of mutually exclusive probabilities vectors is locked in (e.g. with analyte priorities breaking possible ties). All other probabilities in the set may then be set to zero.
• probabilities for neighboring pixels/regions may be taken into account when collapsing data on a pixel level. With respect to putative blob level collapse, overlapping putative blobs may be resolved.
• an appropriate coordinate system can be chosen in order to provide rotational-, translational-, and scale-invariance between different images. These invariances are important to the model because they allow the ability to train on one type of vessel (e.g., carotids where
• the direction of the l-axis is determined by Vl(x).
• the radial distance may be defined based on the shortest distance to the inner luminal surface and the shortest distance to the outer adventitial surface.
• the expert-annotation of the histology images includes regions that define the lumen and the vessel (defined as the union of the lumen and vessel wall). A signed distance function can be created for each of these, L(x) and V(x), respectively. The convention is that the interior of these regions is negative so that in the wall L is positive and V is negative.
• the normalized tangential distance may be defined as lying along iso-contours of r (and of 1 if processing in 3D).
• the direction of the t- axis is determined by Vr c VI.
• the convention is that histology slices are assumed to be viewed looking from the proximal to the distal direction so that positive 1 points into the image. Note that unlike the others, t does not have a natural origin since it wraps onto itself around the vessel. Thus, one can define the origin of this coordinate differently for each blob relative to the centroid of the blob.
• Another wall coordinate that is used is normalized wall thickness. In some sense, this is a proxy for disease progression. Thicker wall is assumed to be due to more advanced disease. Assumption that statistical relationship of analytes changes with more advanced disease.
• the maximum diameter is the maximum Feret diameter of the vessel, D ma x.
• the degree to which the aforementioned coordinates may or may not be used in the model is in part dependent on the amount of training data available.
• the relative longitudinal distance may be ignored treating different sections through each plaque as though they come from the same statistical distribution. It has been observed that plaque composition changes along the longitudinal axis with more severe plaque appearance in the middle. However, instead of parameterizing the distributions by l(x), this dimension can be collapsed. Similarly, the relative wall thickness may also be collapsed. Observations have been made that certain analytes occur in “shoulder” regions of plaques where w(x) would have a middle value. However, this dimension can also be collapsed until enough training data is available.
• FIG. 7 depicts normalized vessel wall coordinates for an exemplary vessel wall composition model.
• 1 is relative longitudinal distance along vessel target from proximal to distal, which may be calculated, e.g., on a normalized the interval [0,1]
• l-axis direction is VI.
• r is normalized radial distance which may also be calculated on a normalized interval [0,1] from luminal to adventitial surface.
• r L/(L+(-V)) where L is lumen signed distance field (SDF) and V is vessel SDF.
• r-axis direction is Vr.
• t is normalized tangential distance which may be computed, e.g., on a normalized interval [-0.5, 0.5]
• the tangential distance is computed along iso-contour curves of 1 and of r. t-axis direction is Vr c VI.
• Figure 9 illustrates some complex vessel topologies which can be accounted for using the techniques described herein.
• different branches may be advantageously analyzed separately so that the relationship between analyte blobs in separate branches are properly ignored.
• a segmented view cross-sectional slice
• the proposed model may advantageously be independent of absolute vessel size, under the assumption that a disease process is similar and proportional for different caliber vessels.
• the model may be configured to characterize concentric vs. eccentric plaque.
• a normalized all thickness close to 1 may indicate highly eccentric place.
• inward vs. outward plaque characterization may be implemented.
• histological information on this characteristic is hindered by deformation.
• CT and training data may be utilized to establish an algorithm for determining inward vs. outward plaque characterization.
• non-imaging data such as histology data
• histology data may be utilized as a training set for establishing algorithms linking image features to biological properties/analytes.
• the data types that need to be addressed in ensuring a proper correlation.
• the following differences between histology and imaging may impact proper correlation: Carotid endarterectomy (CEA) leaves adventitia and some media behind in patient CT or MR image analysis presumed to find outer adventitial surface. (See e.g., Figure 8 depicting the margin between the plaque removed for the histology specimen relative to the outer vessel wall).
• CCA Carotid endarterectomy
• Figure 8 depicting the margin between the plaque removed for the histology specimen relative to the outer vessel wall.
• scientific literature shows uncertainty of whether calcification can occur in adventitia. The following techniques may be employed to account for this difference.
• Histology can be dilated outward, e.g., based on an assumption that little to no analyte in the wall is left behind.
• Image segmentation can be eroded inward, e.g., based on knowledge of typical or particular margins left.
• an average margin may be utilized.
• an average margin may be normalized a percentage of the overall diameter of the vessel.
• histology may be used to mask the imaging (e.g., overlay, based on alignment criteria). In such embodiments it may be necessary to apply one or more transformations to the histology data to match proper alignment.
• the difference may be ignored (which is equivalent to uniform scaling of removed plaque to entire wall). While this may induce some small error, presumably the wall left behind may be thin compared to plaque in CEA patients.
• histological data e.g., a training set
• longitudinal distance may be modeled/correlated explicitly.
• histology slice numbering A-G for example
• all histology slices may be treated as arising from the same distribution.
• some limited regularization may still be employed along the longitudinal direction.
• Normalized wall thickness in some sense is an imperfect proxy for disease progression.
• a thicker wall is assumed to be due to more advanced disease, e.g. based on an assumption that statistical relationship of analytes changes with more advanced disease.
• probabilities may be conditioned based on wall thickness, e.g., so that the distribution of analyte blobs would depend on wall thickness. This advantageously may model differences in analyte composition over the course of disease progression.
• Figure 10 depicts representing an exemplary analyte blob with a distribution of normalized vessel wall coordinates.
• the origin of t is placed at blob centroid.
• (r,t) coordinates are a random vector where the location/shape is fully represented by the joint distribution of points within. This can be simplified by considering the marginal distributions (since radial and tangential shape characteristics seem relatively independent). Marginal distributions may be calculated as projections along r and t (note that 1 and T coordinates can also be considered).
• the marginal distribution in the radial direction may advantageously represent/characterize the plaque growth in concentric layers (e.g., medial layer, adventitial layer and intima layer.)
• the marginal distribution in the tangential direction may advantageously represent a growth factor which may be indicative of the staging of the disease.
• analyte blob descriptors can be computed based on the marginal distributions. For example, on can take low order statistics on the marginal distributions (or use histograms or fit parametric probability distribution functions).
• analyte blob descriptors may be used, e.g., to capture location, shape or other structural characteristics of individual blobs:
• Thick (i.e., advanced) plaques assumed to have different statistics than thin plaques
• pair-wise blob descriptors may also be utilized. For example:
• the set of descriptors form a finite shape space that a blob lives in.
• Figure 11 depicts an exemplary distribution of blog descriptors.
• the distribution of blob descriptors may be computed over the whole training set.
• lower order statistics may be utilized on individual blob descriptors (assuming independence), e.g., Location: E[a r ], Var[a r ].
• a multi-dimensional Gaussian (mean vector + covariance matrix) analysis may be used to model the descriptors (e.g., wherein independence is not assumed).
• the distribution may be modeled with density estimation techniques.
• a number of blobs per cross section or the number of each class, e.g., h without regard to analyte class and rp counting number in each analyte class.
• Figure 14 depicts frequency distribution of the total number of blobs for each histology slide. A poison regression is applied as an overly. Note that the analytic chart of Figure 14 depicts the number of blobs per cross section N without regard to analyte class (number of blobs of each analyte type is represented by B).
• the overall vessel wall composition model may include the following:
• an imaging model may serve as the likelihood (e.g., P(I ⁇ A)) for the Bayesian analytic model.
• a maximum likelihood estimate may then be determined. In example embodiments, this may be done considering each pixel in isolation (e.g., without regard to the prior probability of the structure in the model).
• Estimated analyte maps are typically smooth only because images are smooth (which is why no prior smoothing is typically performed).
• Independent pixel-by-pixel analysis can be done, e.g., at least up to the point of accounting for scanner PSF.
• the imaging model is utilized to account for imperfect imaging data. For example, imaging small components of plaque adds independent noise on top of pixel values. Moreover, the partial volume effect and scanner PSF are well known as applying to small objects.
• simulating CT by Gaussian blurring with PSF is easy and fast.
• the imaging model described herein may also be applied to determine (or estimate) the distribution of true (not blurred) densities of different analytes. Notably this cannot come from typical imaging studies since these will have blurred image intensities. In some embodiments, wide variances could be used to represent the uncertainty.
• distribution parameters could be optimized from training set but the objective function would have to be based on downstream readings (of analyte areas), e.g., unless aligned histology data is available.
• Figure 12 depicts the exemplary model for imaging data (e.g., correlating between a hidden (categorical) state (A(x)) and an observed (continuous) state (I(x)) whereby random (e.g., analyte density distribution (H(A(x))) and deterministic (e.g., scanner blur *G(x)) noise factors are accounted for.
• random e.g., analyte density distribution (H(A(x))
• deterministic e.g., scanner blur *G(x) noise factors are accounted for.
• Q are the parameters of H (proportion & HU mean/variance of each analyte).
• q (ti,mi,si,... ,TN,PN,ON) for N different analyte classes assuming normal distributions.
• Q are patient specific and will be estimated in an expectation maximization (EM) fashion, e.g., wherein analyte labels are the latent variables and the image is observed data.
• EM expectation maximization
• Figure 13 depicts a diagram of an example Markov model/Viterbi algorithm for relating an observed state to a hidden state in an image model.
• the diagram depicts an observed state (gray) (observed image intensity, I(x)) and a hidden state (white) (pure analyte intensity, H(A(x))) which can be modeled either with empirical histogram or with Gaussian or boxcar probability distribution function.
• PSF of imaging system is modeled as Gaussian, G(x).
• EM expectation maximization
• data points are modeled as belonging to one of several classes, which is unknown.
• Each data point has a feature vector and for each class, this feature vector may be modeled with a parametric distribution such as a multidimensional Gaussian, represented by a mean vector and a covariance matrix.
• a straightforward EM implementation would work as follows:
• a modified Bayesian inference procedure may be applied with a much more sophisticated Bayesian prior.
• the variable tau represents the a priori relative proportion of each class but even this variable is unspecified and estimated during the inference procedure.
• the model prior is represented by the multi-scale analyte model. Tau becomes a function of position (and other variables), not just a global proportion.
• the membership probability function is defined as follows:
• the inference algorithm is as follows. At each step of iteration, the membership probability map is initialized to zero so that all classes have zero probability. Then for all possible model configurations, the membership probability map may be added to as follows:
• the probability vector may be normalized at each pixel in the membership probability map to restore the completeness assumption.
• N a relatively low value, for instance 9, at which point extremely few sections have ever been observed to have as many blobs.
• the putative blobs may be thresholded to a small number (N) based on their individual blob probabilities. Then, all of the permutations of N blobs are considered.
• N small number
• This procedure is obviously an approximate inference scheme since the full space of multi-scale model configurations may not be considered.
• This procedure also assumes that the weighted average of the most likely configurations provides a good estimate at each individual pixel.
• Another alternative is to perform a constrained search of model configurations and select the highest likelihood model as the MAP (maximum a posteriori) estimate.
• Posterior probability may be computed as:
• image information may influence analyte probability, Ai(x)
• I(x) is observed image intensity (possibly a vector)
• T(x) is observed relative wall thickness from image segmentation F(x) are CT image features
• S(x) are features of vessel wall shape (e.g., luminal bulge)
• a Metropolis-Hastings like approach may be utilized.
• a maximum a posteriori approach may be applied.
• the following are example algorithmic possibility for a statistical analysis model.
• the model may utilize Belief propagation (AKA max sum, max product, sum product messaging).
• AKA max sum, max product, sum product messaging Belief propagation
• a Viterbi (HMM) type approach may be utilized, e.g., wherein, hidden states are the analyte assignments, A, observed states are the image intensities, I.
• This approach may advantageously find a MAP estimate may be argmax P(A
• a soft output Viterbi algorithm SOVA
• reliability of each decision may be indicated by difference between chosen (survivor) path and discarded path. Thus, this could indicate reliability of each pixel analyte classification.
• a forward/backward Baum-Welch (HMM) approach may be utilized. For example, one can compute most likely state at any point in time but not the most likely sequence (see Viterbi).
• MCMC Metropolis-Hastings
• a simple MRF version for sampling may be utilized. Note that it may be particularly advantageous to sample the posterior directly. In example embodiments, one can build up per-pixel histograms of analyte class.
• an Expectation Maximization (EM) posterior approach Tinder this approach, observed data X is the imaging values, unknown parameters Q are due to the analyte map (but not including analyte probabilities) and latent variable Z is the analyte probability vector.
• EM Expectation Maximization
• a Markov model/chain may be applied.
• Figure 15 depicts exemplary implantation of a 1D Markov chain for Text/DNA.
• a higher order chain may be advantageous, however, due to conditional independence (Markov property).
• FIG. 16 depicts an example first order Markov chain for a text probability table.
• N number of letters
• histology samples have a very high resolution. However, since histology samples are not statistically independent, this may lead to overfitting as later described in greater detail. In general, the more conditional dependence that is modeled, the more predictive the model can be.
• a 2D Markov random field may be used for pixel values instead of a 1D sequence such as for letters.
• Figure 17 depicts conditional dependence of a first pixel (black) based on its neighboring pixels (gray).
• cliques may make use symmetry to reduce the number of dependencies in half.
• the values of pixels could be simple image intensities or could be probability values for classification problems. Problems exist with typical MRF use.
• Conventional MRF almost always is limited to the nearest neighbor pixels providing conditional dependence which greatly reduces the specificity of the probability space represented; usually just black/white blobs for very general
• an inference procedure e.g., a Bayes type rule of Posterior a
• the inference procedure implemented by the systems and methods of the subject application is a bit like trying to OCR a crossword puzzle from a noisy scan. Knowledge (even imperfect knowledge of several squares may help inform an unknown square in the crossword puzzle. Efficiently is improved even more by considering both vertical and horizontal direction simultaneously.
• the inference procedure may be heuristic. For example, one can initialize with uninformed prior, then, solve the easier ones first, which gives you clues about the harder ones which are solved later. Thus, relatively easy to detect biological properties such as dense calcium may inform the existence of other harder to detect analytes such as lipids. Each step of the inference procedure may narrow the probability distributions for unsolved pixels.
• an inference procedure may initialize with unconditional prior probabilities of analytes and then use a highest level of evidence to start narrowing down probabilities. For example, in some embodiments, an uncertain width may be associate with each analyte probability estimate. In other embodiments, closeness to l/N may represent such uncertainty.
• CT resolution is low compared to histology and plaque anatomy
• This may work well with the level set representation of probability maps since they naturally work well with sub-pixel interpolation.
• Discrete analyte states makes the model a discrete space model. However, if one represents continuous probabilities rather than analyte presence/absence, then it becomes a continuous space model.
• table 4 depicts exemplary biological properties/analytes which may utilized with respect to a hierarchical analytics framework for such applications.
• systems may be configured to detect lung lesions.
• an exemplary system may be configured for whole lung segmentation. In some embodiments, this may involve use of minimum curvature evolution to solve juxtapleural lesion problems.
• the system may implement lung component analysis (vessel, fissure, bronchi, lesion etc.).
• a Hessian filter may be utilized to facilitate lung component analysis.
• lung component analysis may further include pleural involvement, e.g., as a function of fissure geometry.
• attachment to anatomic structures may also be considered.
• separate analysis of ground glass vs. solid stated may also be applied. This may include determination of geometric features, such as volume, diameter, sphericity, image features, such as density and mass, and fractal analysis.
• Fractal analysis may be used to infer lepidic growth patterns.
• our method adaptively modifies the support for convolution kernels to limit them to the region of interest (i.e., lung nodule).
• Intersecting vessels/bronchi as well as non-lesion feature may be masked out for the purposes of fractal analysis. This is done by applying HR Gaussian filters over masked local neighborhoods and normalizing with IIR blurred binary masking.
• fractal analysis may further include determining lacunarity (based on variance of the local mean). This may be applied with respect to lung lesions, subparts of lesions.
• HR Gaussian filters or circular neighborhoods may be applied.
• IIR may be utilized to compute variance.
• Average of local variance (AVL) may also be computed, e.g., as applied to lung lesions. Likewise, a variance of local variance may be calculated.
• both lesion structure and composition may be calculated.
• Advantageously calculating lesion structure may utilize full volumetry of thin sections thereby improving on calculating size measurement change.
• Measurements such as sub-solid and ground glass opacity (GGO) volume may also be determined as part of assessing lesion structure.
• tissue characteristics such as consolidation, invasion, proximity and perfusion may be calculated e.g., thereby reducing false positive rate relative to conventional analytics.
• FIG. 18 a further exemplary hierarchical analytics framework 1800 for the systems of the present disclosure is depicted.
• Figure 18 may be understood as an elaboration of Figure 1 elucidating greater detail with respect to exemplary intermediate processing layers of the hierarchical inference system.
• the hierarchical inferences still flow from imaging data 1810 to underlying biological information 1820 to clinical disease 1800.
• the framework 1800 includes multiple levels of data points for processing imaging data in order to determine biological
• a pre-processing level 1812 physical parameters, registrations transformations and region segmentations may be determined. This preprocessed imaging information may then be utilized to extract imaging features at the next level of data points 1814 such as intensity features, shape, texture, temporal characteristics, and the like.
• Extracted image features may next utilized at level 1816 to fit one or more biological models to the imaged anatomy.
• Example models may include a Bayes/Markov net lesion
• the biological model may advantageously act as a bridge for correlating imaging features to underlying biological properties/analytes at level 1822.
• Example biological properties/analytes include anatomic structure, tissue composition, biological function, gene expression correlates, and the like.
• the biological properties/analytes may be utilized to determine clinical findings related to the pathology including, e.g., related to disease subtype, prognosis, decision support and the like.
• Figure 19 is an example application of phenotyping purpose in directing vascular therapy, using the Stary plaque typing system adopted by the AHA as an underlay with in vivo determined types shown in color overlays.
• the left panel shows an example that labels according to the likely dynamic behavior of the plaque lesion based on its physical characteristics
• the right panel illustrates an example which uses the classification result for directing patient treatment .
• the systems and methods of the present disclosure may merge disparate typing systems, the class map may be changed, or other variations.
• FFR phenotypes values such as normal or abnormal may be used, and/or numbers may be used, to facilitate comparison with physical FFR for example.
• Figure 20 is example for a different disease, e.g., lung cancer.
• the subtypes of masses are determined so as to direct the most likely beneficial treatment for the patient based on the manifest phenotype.
• CNNs are expected to perform better than readings-vector classification because CNNs contain filters which extract spatial context which isn't included in (only) analyte area measurements. It may be practical to use a CNN despite the reduced training set because
• example systems and methods described herein may apply spatial unwrapping (for example, training and testing CNNs with (unwrapped dataset) and without (donut dataset) spatial unwrapping). Unwrapping was observed to improve the validation accuracy
• the image volume is preprocessed. This may include target initialization, normalization, and other pre-processing such as deblurring or restoring to form a region of interest containing a physiological target that is to be phenotyped.
• Said region is a volume composed of cross sections through that volume.
• Body site is either automatically determined or is provided explicitly by user.
• Targets for body sites that are tubular in nature are accompanied with a centerline. Centerlines, when present, can branch. Branches can be labelled either automatically or by user. Generalizations on centerline concept may be represented for anatomy that is not tubular but which benefit by some structural directionality, e.g., regions of a tumor. In any case, a centroid is determined for each cross section in the volume.
• Figure 21 is representative of an exemplary image pre-processing step, in this case deblurring or restoring using a patient-specific point spread determination algorithm to mitigate artifacts or image limitations that result from the image formation process that may decrease the ability to determine characteristics predictive of the phenotype.
• the figure demonstrates a portion of the radiology analysis application analysis of a plaque from CT. Shown here are the deblurred or restored images that are a result of iteratively fitting a physical model of the scanner point spread function with regularizing assumptions about the true latent density of different regions of the image.
• the (optionally deblurred or restored) image is represented in a Cartesian data set where x is used to represent how far from centroid, y represents a rotational theta, and z represents the cross section.
• Cartesian data set where x is used to represent how far from centroid, y represents a rotational theta, and z represents the cross section.
• x is used to represent how far from centroid
• y represents a rotational theta
• z represents the cross section.
• One such Cartesian set will be formed per branch or region.
• a“null” value will be used for overlapping regions, that is, each physical voxel will be represented only once across the sets, in such a way as to geometrically fit together.
• Each data set will be paired with an additional data set with sub-regions labelled by objectively verifiable tissue composition (see, e.g., Figure 36).
• Example labels for vascular tissue can be lumen, calcification, LRNC, etc.
• Example labels for lesions could be necrotic, neovascularized, etc. These labels can be validated objectively, e.g. by histology (see, e.g., Figure 37). Paired data sets will used as input to a training step to build a convolutional neural network.
• Two levels of analysis are supported, one at an individual cross-section level, optionally where the output varies continuously across adjacent cross-sections, and a second at the volume level (where individual cross-sections may be thought of as still frames, and the vessel tree traversal could be considered as analogous to movies).
• AlexNet is a CNN, which competed in the ImageNet Large Scale Visual Recognition Challenge in 2012. The network achieved a top-5 error of 15.3%. AlexNet was designed by the Supervision group, consisting of Alex Krizhevsky, Geoffrey Hinton, and Ilya Sutskever at U Toronto at the time. AlexNet was trained from scratch to classify an independent set of images (not used in training and validation steps during the network training). For the unwrapped data an AlexNet style network with 400x200 pixel input was used, and the donut network is AlexNet style with 280x280 pixel input (roughly the same resolution but different aspect ratio). All of the convolutional filter values were initialized with weights taken from AlexNet trained on the ImageNet dataset.
• the ImageNet dataset is a natural image dataset, this simply serves as an effective method of weight initialization. Once training begins, all weights are adjusted to better fit the new task. Most of the training schedule was taken directly from the open source AlexNet implementation, but some adjustment was needed. Specifically, the base learning rate was reduced to 0.001 (solver. prototxt) and the batch size was reduced to 32 (train val. prototxt) for both the AlexNet-donut and AlexNet- unwrapped networks. All models were trained to 10,000 iterations and were compared to snapshots when trained till just 2,000 iterations. While a more in depth study on overfitting could be done, it was generally found that both training and validation error decreased between 2k and lOk iterations.
• Run-time optimizations such as frame-to-frame redundancy between cross-sections (sometimes referred to as "temporal" redundancy, but in our case, being a form of inter-cross- section redundancy) could be leveraged to save on computation (e.g.,
• AlexNet was trained to classify an independent set of images between two categories of clinical significance, e.g.,‘unstable’ plaques and‘stable’ plaques, based on histology ground truth plaque types of V and VI, while the latter includes plaque types VII and VIII following the industry de-facto standard plaque classification nomenclature accepted by the American Heart Association (AHA), and on a related but distinct typing system by Virmani
• Physician annotated data is expensive, so it is desirable to artificially increase medical datasets (e.g., for use in training and/or validation).
• Two different augmentation techniques were used in example embodiments described herein. Donuts were horizontally flipped randomly, as well as rotated to a random angle from 0 to 360. The resulting rotated donut was then cropped to the range in which the donut was present, and then padded with black pixels to fill the image to have a square aspect ratio. The result was then scaled to the 280x280 size and saved to a PNG.
• the unwrapped dataset was augmented by randomly horizontally flipping, and then “scrolled” by a random number of pixels in the range from 0 to the width of the image. The result was then scaled to the 400x200 size and saved to a PNG.
• Figure 22 illustrates an exemplary application that used to demonstrate aspects of the present invention, in this case being for classification of atherosclerotic plaque phenotype classification (using specific subject data by way of example).
• Different colors represent different tissue analyte types with dark gray showing the otherwise normal wall.
• the figure illustrates the result of ground truth annotation of tissue characteristics that are indicative of plaque phenotype as well as the spatial context of how they present in a cross section taken orthogonal to the axis of the vessel. It also illustrates a coordinate system that has been developed in order to provide a common basis for analysis of a large number of histological cross sections. Grid lines added to demonstrate coordinate system (tangential vs. radial distance) and overlaid on top of color-coded pathologist annotations.
• data sets of this kind may be used efficiently in deep learning approaches because they simplify the information using a relatively simpler false color image in place of a higher- resolution full image but without losing spatial context, e.g., to have a formal representation for such presentations as“napkin ring sign”, juxtaluminal calcium, thin (or thick) caps (spacing between LRNC and lumen), etc.
• Figure 23 illustrates tangential and radial direction variable internally represented using unit phasors and here, phasor angle shown coded in gray scale, which exemplifies the use of normalized axes for tubular structures relevant to the vascular and other
• the vertical bar from black to white is a purely arbitrary boundary due to the gray scale encoding, and the normalized radial distance has a value of 0 at the luminal boundary and value of 1 at the outer boundary.
• Figure 24 illustrates an exemplary overlay of radiology analysis application generated annotations from CTA (unfilled color contours) on top of pathologist generated annotations from histology (solid color regions).
• CTA unfilled color contours
• pathologist generated annotations from histology solid color regions.
• classification scheme so as to determine phenotype non-invasively, where the classifier is trained on known ground truth. Specifically, filling in the contours which are shown unfilled in this figure (so as to not obscure the relationship with the ex vivo annotation for this specific section which is provided to show the correspondence) creates input data for the classifier.
• Figure 25 demonstrates a further step of data enrichment, specifically, utilizing the normalized coordinate system to avoid non-relevant variation associated with the wall thickness and radial presentation.
• the“donut” is“unwrapped” while retaining the pathologist annotations.
• the left panel illustrates pathologist region annotations of a histological slice of a plaque after morphing to convert the cut open“C” shape back into the in vivo“O” shape of the intact vessel wall.
• Horizontal axis is the tangential direction around the wall.
• Vertical axis is the normalized radial direction (bottom is luminal surface, top is outer surface).
• the finer granularity of the pathologist annotations has been collapsed to match the granularity intended for extraction by the in vivo radiology analysis application (e.g., LRNC, CALC, IPH).
• the right panel illustrates the comparable unwrapped radiology analysis application annotations. Axes and colors are the same as the pathologist annotations.
• Figure 26 represents the next refinement relevant to the plaque phenotyping example.
• luminal irregularity as results, for example from ulceration or thrombus
• local varying wall thickening are represented.
• the light grey at bottom represents the lumen (added in this step so as to represent that luminal surface irregularity) and the black used in the prior step is now replaced with dark grey, to represent the varying wall thickening. Black now represents area outside of the wall entirely.
• Figure 27 represents an additional example, so as to include, for example, intra-plaque hemorrhage and/or other morphology aspects as needed (using specific subject data by way of example).
• the left panel shows the donut representation and the right panel the unwrapped with the luminal surface and localized wall thickening represented.
• the convolutional filter values were initialized with weights taken from AlexNet trained on the ImageNet (cite here) dataset. While the ImageNet dataset is a natural image dataset, this simply serves as an effective method of weight initialization. Once training begins, all weights are adjusted to better fit the new task.
• the base learning rate was reduced to 0.001 (solver. prototxt) and the batch size was reduced to 32 (train val.prototxt) for both the alexnet-donut and alexnet-unwrapped networks.
• AlexNet network model was trained from scratch for 4 (four) different combinations of ground-truth results of two leading pathologists, two different ways of processing images (see above), as well as unwrapped images and donut images.
• the results are listed in Figure 28.
• Each dataset variation had its training data augmented by 15c with class normalization enabled.
• a network was trained on this augmented data and then was tested on the corresponding un-augmented validation data corresponding to that variation.
• For the unwrapped data an AlexNet style network with 400x200 pixel input was used, and the donut network is AlexNet style with 280x280 pixel input (roughly the same resolution but different aspect ratio). Note that in test embodiments the dimensions of the conventional layers as well as the fully connected layers were changed.
• the network in the AlexNet test embodiments can be described as a five convolutional layer, three fully connected layer network. Without loss of generality, here are some high-level conclusions that are illustrated from these results:
• unwrapped data is easier for the data to analyze as it receives higher validation accuracy across the board
• Transfer-learning re-training of an Inception v3 CNN was started with the August 08 2016 version of the network uploaded on the TensorFlow site for public use. The network was trained for 10,000 steps. Training and Validation sets were normalized in number of images via image augmentation so both sub-sets amounted to the same number of annotated images. All other network parameters were taken to be at their default values.
• Pre-trained CNNs can be used to classify imaging features using the output from the last convolution layer, which is a numeric tensor with dimensionality of 2048 x 2 in the case of the Google Inception v3 CNN. We then train an SVM classifier to recognize the object.
• This process is normally performed on the Inception model after a transfer-learning and Fine- Tuning steps in which the model initially trained on the ImageNet 2014 dataset has its last, softmax layer removed and re-trained to recognize the new categories of images.
• Figure 29 provides an alternative example, for phenotyping potentially cancerous lung lesions.
• the left-most panel indicates the outlines of a segmented lesion, with pre-processing to separate out into solid vs. semi-solid (“ground glass) sub regions.
• the middle panels indicate its location in the lung, and the right-most panel shows it with false color overlay.
• 2-D cross-sections separately, techniques such as video interpretation from computer vision may be applied for the classifier input data set.
• processing multiple cross-sections sequentially, as if in a“movie” sequence along a centerline can generalize these methods for tubular structures.
• Figure 30 shows a set of features, sometimes described as so-called“radiomics” features that can be calculated for each voxel. Such a set of values may exist in arbitrary number of pre-processed overlays and be fed into the phenotype classifier.
• change data for example as collected from multiple timepoints, rather than (only) data from a single timepoint.
• change data for example as collected from multiple timepoints, rather than (only) data from a single timepoint.
• change data for example, if the amount or nature of a negative cell type increased, it may be said to be a“progressor” phenotype, vs. a“regressor” phenotype for decreases.
• the regressor might be, for example, due to response to a drug.
• the rate of change for, say, LRNC is rapid, this may imply a different phenotype.
• the extension of the example to use delta values or rates of change is obvious to one skilled in the art.
• non-spatial information such as which are derived from other assays (e.g., lab results), or demographics/risk factors, or other
• measurements taken from the radiological image may be fed into the final layers of the CNN to combine the spatial information with non-spatial information.
• localized information such as use of a pressure wire with readings at one or more certain locations along a vessel from a reference such as a bifurcation or ostia, by inference of full 3D coordinates at imaging may be determined.
• FIG. 34 provides a further example implementation of a hierarchical analytics framework.
• Biological properties are identified and quantified by semantic segmentation to identify biological properties singly or in combination, in the example application lipid-rich necrotic core, cap thickness, stenosis, dilation, remodeling ratio, tortuosity (e.g., entrance and exit angles), calcification, IPH, and/or ulceration, which is represented numerically as well as in enriched data sets with spatial unwrapping to convert the vein/artery cross-section into a rectangle, and then medical conditions (example.
• Ground truth data for HRP may exist as expert pathologist determined plaque types at given cross-sections, having been determined ex vivo.
• Ground truth data for FFR may be from physical pressure wire, with one or more measured values, and network training for locations along the centerline proximal of a given measurement being constrained to be greater than or equal to the measured value, locations distal being constrained to be less than or equal, and when two measurements on the same centerline are known, that the values between the two measured values be constrained within the interval.
• biological properties can include one or more of the following:
• Pressure e.g., of blood in channel or one tissue pressing against another
• Cell types e.g., macrophages
• Shear stress e.g., of blood in channel
• Analysis can include determining one or more of quantity, degree and/or character for each of the aforementioned biological properties.
• Conditions that can be determined based on the biological properties may include one or more of:
• Perfusion/ischemia (as limited) (e.g., of brain or heart tissue)
• High-risk plaque e.g., HRP(+) vs. (-) and/or labelled phenotype
• Risk stratification (whether as probability of event, or time to event) (e.g., MACCE, mentioned explicitly)
• Validation in the form of truth bases can include the following:
• Physiological monitoring e.g., ECG, Sa02, etc.
• Figure 31 illustrates a high-level view of the users and other systems that interact with an analytics platform, as per the systems and methods of the present disclosure. Stakeholders of this view include System Administrators, Support Technicians, which have Interoperability, Security, Failover & Disaster Recovery, Regulatory concerns.
• the platform can be deployed in two main configurations; on-premises, or remote server ( Figure 32).
• the platform deployment may be a stand-alone configuration (Left Upper), on-premises server configuration (Left Lower), or remote server configuration (Right).
• the on-premises deployment configuration can have two sub-configurations; desktop only or rackmount.
• the remote configuration the platform may be deployed on a FUPAA compliant data center. Clients access that API server over a secure HTTP connection. Clients can be desktop or tablet browsers. No hardware except for the computers running the web browsers is deployed on the customer site.
• the deployed server may be on a public cloud or an extension of the customer’s private network using a VPN.
• FIG. 33 An exemplary embodiment is comprised of a client and a server.
• Figure 33 illustrates a client as a C++ application and the server as a Python application.
• These components interact using HTML 5.0, CSS 5.0 and JavaScript.
• HTML 5.0 CSS 5.0
• JavaScript Wherever possible open standards are used for interfaces including but not limited to; HTTP(S), REST, DICOM, SPARQL, and JSON.
• Third party libraries are also used as shown in this view which shows the primary pieces of the technology stack. Many variations and different approaches may be understood by people skilled in the art.
• the implementation can be as a computer program product (i.e., a computer program tangibly embodied in an information carrier).
• the implementation can, for example, be in a machine- readable storage device and/or in a propagated signal, for execution by, or to control the operation of, data processing apparatus.
• the implementation can, for example, be a programmable processor, a computer, and/or multiple computers.
• a computer program can be written in any form of programming language, including compiled and/or interpreted languages, and the computer program can be deployed in any form, including as a stand-alone program or as a subroutine, element, and/or other unit suitable for use in a computing environment.
• a computer program can be deployed to be executed on one computer or on multiple computers at one site.
• Method steps can be performed by one or more programmable processors executing a computer program to perform functions of the invention by operating on input data and generating output. Method steps can also be performed by and an apparatus can be implemented as special purpose logic circuitry.
• the circuitry can, for example, be a FPGA (field programmable gate array) and/or an ASIC (application specific integrated circuit).
• Modules, subroutines, and software agents can refer to portions of the computer program, the processor, the special circuitry, software, and/or hardware that implements that functionality.
• processors suitable for the execution of a computer program include, by way of example, both general and special purpose microprocessors, and any one or more processors of any kind of digital computer.
• a processor receives instructions and data from a read-only memory or a random access memory or both.
• the essential elements of a computer are a processor for executing instructions and one or more memory devices for storing instructions and data.
• a computer can include, can be operatively coupled to receive data from and/or transfer data to one or more mass storage devices for storing data (e.g., magnetic, magneto-optical disks, or optical disks).
• Data transmission and instructions can also occur over a communications network.
• Information carriers suitable for embodying computer program instructions and data include all forms of non-volatile memory, including by way of example semiconductor memory devices.
• the information carriers can, for example, be EPROM, EEPROM, flash memory devices, magnetic disks, internal hard disks, removable disks, magneto-optical disks, CD- ROM, and/or DVD-ROM disks.
• the processor and the memory can be supplemented by, and/or incorporated in special purpose logic circuitry.
• the above described techniques can be implemented on a computer having a display device.
• the display device can, for example, be a cathode ray tube (CRT) and/or a liquid crystal display (LCD) monitor.
• CTR cathode ray tube
• LCD liquid crystal display
• the interaction with a user can, for example, be a display of information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer (e.g., interact with a user interface element).
• Other kinds of devices can be used to provide for interaction with a user.
• Other devices can, for example, be feedback provided to the user in any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback).
• Input from the user can, for example, be received in any form, including acoustic, speech, and/or tactile input.
• the above described techniques can be implemented in a distributed computing system that includes a back-end component.
• the back-end component can, for example, be a data server, a middleware component, and/or an application server.
• the above described techniques can be implemented in a distributing computing system that includes a front-end component.
• the front-end component can, for example, be a client computer having a graphical user interface, a Web browser through which a user can interact with an example implementation, and/or other graphical user interfaces for a transmitting device.
• the components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), the Internet, wired networks, and/or wireless networks.
• LAN local area network
• WAN wide area network
• the Internet wired networks, and/or wireless networks.
• the system can include clients and servers.
• a client and a server are generally remote from each other and typically interact through a communication network.
• the relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.
• Packet-based networks can include, for example, the Internet, a carrier internet protocol (IP) network (e.g., local area network (LAN), wide area network (WAN), campus area network (CAN), metropolitan area network (MAN), home area network (HAN)), a private IP network, an IP private branch exchange (IPBX), a wireless network (e.g., radio access network (RAN), 802.11 network, 802.16 network, general packet radio service (GPRS) network, HiperLAN), and/or other packet-based networks.
• IP carrier internet protocol
• LAN local area network
• WAN wide area network
• CAN campus area network
• MAN metropolitan area network
• HAN home area network
• IP network IP private branch exchange
• wireless network e.g., radio access network (RAN), 802.11 network, 802.16 network, general packet radio service (GPRS) network, HiperLAN
• GPRS general packet radio service
• HiperLAN HiperLAN
• Circuit-based networks can include, for example, the public switched telephone network (PSTN), a private branch exchange (PBX), a wireless network (e.g., RAN, Bluetooth, code-division multiple access (CDMA) network, time division multiple access (TDMA) network, global system for mobile communications (GSM) network), and/or other circuit-based networks.
• PSTN public switched telephone network
• PBX private branch exchange
• CDMA code-division multiple access
• TDMA time division multiple access
• GSM global system for mobile communications
• the computing device can include, for example, a computer, a computer with a browser device, a telephone, an IP phone, a mobile device (e.g., cellular phone, personal digital assistant (PDA) device, laptop computer, electronic mail device), and/or other communication devices.
• the browser device includes, for example, a computer (e.g., desktop computer, laptop computer) with a World Wide Web browser (e.g., Microsoft® Internet Explorer® available from Microsoft Corporation, Mozilla® Firefox available from Mozilla Corporation).
• the mobile computing device includes, for example, a Blackberry®, iPAD®, iPhone® or other smartphone device.

Landscapes

• Engineering & Computer Science (AREA)
• Theoretical Computer Science (AREA)
• Physics & Mathematics (AREA)
• General Physics & Mathematics (AREA)
• Computer Vision & Pattern Recognition (AREA)
• Evolutionary Computation (AREA)
• Artificial Intelligence (AREA)
• Data Mining & Analysis (AREA)
• General Health & Medical Sciences (AREA)
• Health & Medical Sciences (AREA)
• Life Sciences & Earth Sciences (AREA)
• Multimedia (AREA)
• General Engineering & Computer Science (AREA)
• Medical Informatics (AREA)
• Molecular Biology (AREA)
• Biomedical Technology (AREA)
• Bioinformatics & Cheminformatics (AREA)
• Software Systems (AREA)
• Bioinformatics & Computational Biology (AREA)
• Computing Systems (AREA)
• Evolutionary Biology (AREA)
• Biodiversity & Conservation Biology (AREA)
• Databases & Information Systems (AREA)
• Probability & Statistics with Applications (AREA)
• Radiology & Medical Imaging (AREA)
• Geometry (AREA)
• Quality & Reliability (AREA)
• Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
• Biophysics (AREA)
• Computational Linguistics (AREA)
• Mathematical Physics (AREA)
• Image Analysis (AREA)
• Measuring And Recording Apparatus For Diagnosis (AREA)
• Apparatus For Radiation Diagnosis (AREA)
• Investigating Or Analysing Biological Materials (AREA)
• Image Processing (AREA)

Applications Claiming Priority (3)

Publications (2)

Family

ID=68697312

Family Applications (1)

Country Status (5)

Families Citing this family (19)

Family Cites Families (16)

• 2019
  • 2019-05-24 EP EP19811880.4A patent/EP3803687A4/de active Pending
  • 2019-05-24 KR KR1020207037314A patent/KR102491988B1/ko active IP Right Grant
  • 2019-05-24 WO PCT/US2019/033930 patent/WO2019231844A1/en unknown
  • 2019-05-24 CN CN201980049912.9A patent/CN112567378A/zh active Pending
  • 2019-05-24 JP JP2020567160A patent/JP7113916B2/ja active Active
• 2022
  • 2022-06-22 JP JP2022100036A patent/JP2022123103A/ja active Pending

Also Published As

Similar Documents

Legal Events