WO2014207627A1 - Method and system for multi-modal tissue classification - Google Patents

Method and system for multi-modal tissue classification Download PDF

Info

Publication number
WO2014207627A1
WO2014207627A1 PCT/IB2014/062422 IB2014062422W WO2014207627A1 WO 2014207627 A1 WO2014207627 A1 WO 2014207627A1 IB 2014062422 W IB2014062422 W IB 2014062422W WO 2014207627 A1 WO2014207627 A1 WO 2014207627A1
Authority
WO
WIPO (PCT)
Prior art keywords
tissue
image
voxel
anatomical
feature
Prior art date
Application number
PCT/IB2014/062422
Other languages
French (fr)
Inventor
Amir Mohammad TAHMASEBI MARAGHOOSH
Shyam Bharat
Christopher Stephen Hall
Jochen Kruecker
Original Assignee
Koninklijke Philips N.V.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Koninklijke Philips N.V. filed Critical Koninklijke Philips N.V.
Priority to US14/891,768 priority Critical patent/US11660065B2/en
Priority to CN201480036465.0A priority patent/CN105338905B/en
Priority to JP2016522913A priority patent/JP6445548B2/en
Priority to EP14738901.9A priority patent/EP3013241B1/en
Publication of WO2014207627A1 publication Critical patent/WO2014207627A1/en

Links

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/08—Detecting organic movements or changes, e.g. tumours, cysts, swellings
    • A61B8/0833—Detecting organic movements or changes, e.g. tumours, cysts, swellings involving detecting or locating foreign bodies or organic structures
    • A61B8/085—Detecting organic movements or changes, e.g. tumours, cysts, swellings involving detecting or locating foreign bodies or organic structures for locating body or organic structures, e.g. tumours, calculi, blood vessels, nodules
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/05—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves 
    • A61B5/055—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves  involving electronic [EMR] or nuclear [NMR] magnetic resonance, e.g. magnetic resonance imaging
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/52—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/5215—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving processing of medical diagnostic data
    • A61B8/5223—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving processing of medical diagnostic data for extracting a diagnostic or physiological parameter from medical diagnostic data
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/52—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves
    • A61B8/5215—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving processing of medical diagnostic data
    • A61B8/5238—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving processing of medical diagnostic data for combining image data of patient, e.g. merging several images from different acquisition modes into one image
    • A61B8/5261—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves involving processing of medical diagnostic data for combining image data of patient, e.g. merging several images from different acquisition modes into one image combining images from different diagnostic modalities, e.g. ultrasound and X-ray
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T15/00—3D [Three Dimensional] image rendering
    • G06T15/08—Volume rendering
    • 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
    • G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
    • 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
    • G06T2207/10088—Magnetic resonance imaging [MRI]
    • 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
    • G06T2207/10132—Ultrasound 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
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30024—Cell structures in vitro; Tissue sections in vitro
    • 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
    • G06T2207/30081—Prostate

Definitions

  • Prostate cancer is one of the most common organ malignancies among
  • TRUS transrectal ultrasound
  • approximately 40% of the peripheral zone hypoechoic lesions have proven to be malignant.
  • T2W anatomical T2 -weighted
  • DCE dynamic contrast enhanced MR imaging
  • MRSI proton MR spectroscopic imaging
  • DWI diffusion-weighted MR imaging
  • the B-mode US image processing pipeline suppresses potential tissue signatures that may assist in differentiating cancerous tissue from normal tissue.
  • FIG. 4 illustrates a first exemplary implementation of the flowchart illustrated in FIG. 3 by the system illustrated in FIG. 2.
  • FIG. 5 illustrates a second exemplary implementation of the flowchart illustrated in FIG. 3 by the system illustrated in FIG. 2.
  • a MRI system 20 employs a scanner 21 and a workstation 22 to generate a MRI image of an anatomical tissue of a patient, such as, for example, a MRI image 23 of anatomical tissue of a prostate 11 of a patient 10 as shown.
  • the present invention utilizes one or more MRI systems 20 of various types to acquire MRI features of the anatomical tissue.
  • types of MRI systems and associated MRI features of an anatomical tissue include, but are not limited to, a T 2 W MRI system illustrating normalized intensity-value and/or texture based features of the anatomical tissue, a DWI-MRI system illustrating apparent diffusion coefficient ("ADC") of water in the anatomical tissue, a DCE-MRI system illustrating
  • each voxel of tissue classification volume 40a is classified as a being a healthy tissue voxel or an unhealthy tissue voxel (e.g., cancerous) based on the image feature vector of the voxel derived from the extracted image features of the spatially registered MRI images and ultrasound images.
  • the healthy tissue voxels are the white voxels and the unhealthy tissue voxels are the black voxels.
  • FIG. 1 Another example, not shown in FIG. 1, is the utilization of a grayscale of voxels ranging from white healthy tissue voxels through multiple gray voxels indicating various probabilities of unhealthiness of corresponding tissue to black unhealthy tissue voxels (e.g., cancerous).
  • exemplary embodiments of the present invention will now be described herein as directed to a generation of tissue classification volumes of anatomical tissue of a prostate.
  • image segmentation image (spatial) registration
  • image delineation image feature extraction
  • tissue classification tissue classification
  • image registrator 51 employs technique(s) for spatial registration(s) of the voxels of MRI image(s) 70 and ultrasound image(s) 80.
  • tissue classifier 53 employs technique(s) for classifying, unsupervised or supervised, a tissue type of each voxel of the tissue classification volume as indicated by the image feature vector of each voxel.
  • a user interface (not shown) of workstation 50 provides for the display of the tissue classification volume on a voxel-by-voxel basis including zoom and pan capabilities, such as, for example, the display of tissue classification volume 40a as shown in FIG. 2.
  • workstation 50 may provide tools for modules 51-53 including, but not limited, to an image segmentation tool to segment voxels of the anatomical tissue from MRI images 70 and US images 80 and an image delineation tool to delineate voxels of MRI images 70 and US images 80 suspicious of being unhealthy (e.g., cancerous).
  • workstation 50 may be a stand-alone workstation providing tissue classification volume(s) to an image diagnostic system 90 (e.g., a MRI system or an ultrasound system) for incorporation into a diagnostic procedure as needed and/or an interventional guidance system 91 (e.g., an electromagnetic tracking system, optical tracking system or image tracking system) for incorporation into an image diagnostic system 90 (e.g., a MRI system or an ultrasound system) for incorporation into a diagnostic procedure as needed and/or an interventional guidance system 91 (e.g., an electromagnetic tracking system, optical tracking system or image tracking system) for incorporation into an image diagnostic system 90 (e.g., a MRI system or an ultrasound system) for incorporation into a diagnostic procedure as needed and/or an interventional guidance system 91 (e.g., an electromagnetic tracking system, optical tracking system or image tracking system) for incorporation into an image diagnostic system 90 (e.g., a MRI system or an ultrasound system) for incorporation into a diagnostic procedure as needed and/or an interventional guidance system 91
  • workstation 50 may be incorporated within image diagnostic system 90 or incorporated within interventional guidance system 91.
  • FIGS. 4 and 5 will now be described herein in the context of MRI images 70 consisting of a T 2 W image 71, a DWI image 72, a DCE image 73 and a MRSI image 74, and US images 80 consisting of a B-mode image 81 and a RF echo image 82. From the description, those having ordinary skill in the art will appreciate alternative systems and devices for implementing multi-modal tissue classification methods of the present invention.
  • stage SI 02 of flowchart 100 encompasses feature extractor 52 extracting and concatenating image features from each voxel of the spatial registered MRI image(s) 70 and US image(s) 80 to generate a n-dimensional image feature vector for each voxel of the tissue classification volume.
  • a feature extractor 52a extracts normalized intensity-value and/or texture based features of the anatomical tissue from T 2 W image 71, an apparent diffusion coefficient ("ADC") of water of the anatomical tissue from DWI image 72, pharmacokinetic parameters of the anatomical tissue from DCE image 73, metabolic information of the anatomical tissue from MRS image 74, texture based image features of the anatomical tissue from B-mode image 81 and spectral image features of the anatomical tissue from RF image 82. From the spatially registered extractions, feature extractor 52a generates a vector 40a of six (6) dimensional image features for each voxel of the tissue classification volume.
  • ADC apparent diffusion coefficient
  • the tissue classification volume may be displayed by workstation 50, image diagnostic system 90 and/or interventional guidance system 91. Additionally, an overlay or fusion of the tissue classification volume may be displayed on/within an image of the anatomical tissue generated by image diagnostic system 90 and/or interventional guidance system 91.

Landscapes

  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Medical Informatics (AREA)
  • General Health & Medical Sciences (AREA)
  • Radiology & Medical Imaging (AREA)
  • Public Health (AREA)
  • Biomedical Technology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Pathology (AREA)
  • Molecular Biology (AREA)
  • Animal Behavior & Ethology (AREA)
  • Biophysics (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Veterinary Medicine (AREA)
  • Surgery (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Vascular Medicine (AREA)
  • High Energy & Nuclear Physics (AREA)
  • Quality & Reliability (AREA)
  • Physiology (AREA)
  • Computer Graphics (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Epidemiology (AREA)
  • Primary Health Care (AREA)
  • Databases & Information Systems (AREA)
  • Magnetic Resonance Imaging Apparatus (AREA)
  • Ultra Sonic Daignosis Equipment (AREA)

Abstract

A method for multi-modal tissue classification of an anatomical tissue involves a generation of a tissue classification volume (40) of the anatomical tissue derived from a spatial registration and an image extraction of one or more MRI features of the anatomical tissue and one or more ultrasound image features of the anatomical tissue. The method further involves a classification of each voxel of the tissue classification volume (40) as one of a plurality of tissue types including a healthy tissue voxel and an unhealthy tissue voxel.

Description

METHOD AND SYSTEM
FOR MULTI-MODAL TISSUE CLASSIFICATION
The present invention generally relates to an imaging of anatomical tissues. The present invention specifically relates to a utilization of multiple imaging modalities for screening, detecting and treating unhealthy (e.g., cancerous) anatomical tissues.
Prostate cancer is one of the most common organ malignancies among
American men. Historically, a "gold standard" for prostate cancer diagnosis is a histopathologic analysis of biopsy tissue samples whereby elevated prostate specific antigen ("PSA") levels and results of digital rectal exam ("DRE") tests are considered as the screening. However, a PSA test yields low specificity and sensitivity, and detection by palpation through DRE is limited to relatively large and superficial lesions.
In ultrasound ("US") imaging, tissue classification based on acoustic parameters
(e.g., attenuation and backscattered coefficients) extracted from radio frequency ("RF") echo signals has been studied since the early 1970s. More particularly, texture features extracted from B-mode US images and spectral features extracted from calibrated average spectrum of RF echo signals have been used along with numerous
classification approaches for tissue typing as healthy or cancerous. However, while US imaging provides high temporal resolution imaging of the anatomical tissue of interest, accuracy of ultrasound-based cancer detection techniques has proven to be limited due to poor signal-to-noise ratio and low spatial resolution.
More particularly, transrectal ultrasound ("TRUS") is sometimes used as an alternative step in the screening process for prostate cancer. However, approximately 40% of the peripheral zone hypoechoic lesions have proven to be malignant.
Therefore, a clinical value of TRUS is limited to a biopsy guidance tool.
Recently, there has been significant interest in using magnetic resonance imaging ("MRI") for diagnosis of prostate cancer due to its high anatomical resolution. However, a diagnostic value of anatomical T2 -weighted ("T2W") MRI in
distinguishing prostate cancer from benign prostate lesions is limited. For example, it has been shown that the accuracy, sensitivity and positive predictive values of prostate cancer detection using MRI for tumor foci greater than 1.0 cm in diameter are 79.8%, 85.3%), and 92.6%>, respectively. Moreover, the accuracy, sensitivity and positive predictive values of prostate cancer detection using MRI decreases to 24.2%, 26.2%, and 75.9%), respectively, for tumor foci smaller than 1.0 cm.
However, a combination of anatomic, biologic, metabolic and functional dynamic information offered by multi-parametric MRI has been shown to improve prostate cancer detection accuracy. A few of the common functional MR imaging techniques that have been utilized for prostate cancer detection are dynamic contrast enhanced MR imaging ("DCE"), proton MR spectroscopic imaging ("MRSI"), and diffusion-weighted MR imaging ("DWI"). More particularly, DCE visualizes tissue vascularity, MRSI provides metabolic information and DWI shows the Brownian motion of extracellular water molecules.
Importantly, a standard of care in prostate cancer therapy is moving from a 'whole gland' approach (e.g., radical prostatectomy, whole gland radiation and brachytherapy) to more focused and localized treatment paradigms designed to only treat specific cancerous regions within the prostate. A step in this direction is the ability to accurately identify these localized cancerous regions. However, in spite of the improved capabilities of multi-parametric MRI, it is still inherently difficult for observers to accurately identify cancerous regions in a consistent manner.
Furthermore, in ultrasound images, the B-mode US image processing pipeline suppresses potential tissue signatures that may assist in differentiating cancerous tissue from normal tissue.
In general, the present invention recognizes that an automated decision support system or computer-aided diagnosis that includes different algorithms to perform image registration, image segmentation, image feature extraction, image delineation and tissue classification on multi-modality images may provide a systematic and objective approach to fuse information from different types of images. Specifically, the present invention is premised on combining tissue information extracted from spatially registered MRI images (e.g., T1W, T2W, a DWI, a DCE and MRSI) and US images (e.g., a B-mode image and a RF echo image) in order to form a vector of unique image features for each voxel of the registered images. Such a multi-modal (US-RF-MRI) vector of features combines the advantages of the multi-modalities to achieve higher accuracy, sensitivity and specificity in detecting prostate cancer.
One form of the present invention is a system for multi-modal tissue classification of an anatomical tissue that employs one or more MRI systems, one or more ultrasound systems and a workstation. In operation, the MRI system(s) generate one or more MRI features of the anatomical tissue and the ultrasound system(s) generate one or more ultrasound image features of the anatomical tissue. The workstation generates a tissue classification volume of the anatomical tissue derived from a spatial registration and an image extraction of the MRI feature(s) and the ultrasound image feature(s) of the anatomical tissue. The workstation further classifies each voxel of the tissue classification volume as one of a plurality of tissue types including a healthy tissue voxel and an unhealthy tissue voxel.
A second form of the present invention is a workstation for multi-modal tissue classification of an anatomical tissue employing an image registrator, a feature extractor and a tissue classifier. In operation, the image registrator and the feature extract generate a tissue classification volume of the anatomical tissue derived from a spatial registration and an image extraction of one or more MRI features of the anatomical tissue and one or more ultrasound image features of the anatomical tissue. The tissue classifier classifies each voxel of the tissue classification volume as one of a plurality of tissue types including a healthy tissue voxel and an unhealthy tissue voxel.
A third form of the present invention is a method for multi-modal tissue classification of an anatomical tissue involving a generation of a tissue classification volume of the anatomical tissue derived from a spatial registration and an image extraction of one or more MRI features of the anatomical tissue and one or more ultrasound image features of the anatomical tissue. The method further involves a classification of each voxel of the tissue classification volume as one of a plurality of tissue types including a healthy tissue voxel and an unhealthy tissue voxel.
The foregoing forms and other forms of the present invention as well as various features and advantages of the present invention will become further apparent from the following detailed description of various embodiments of the present invention read in conjunction with the accompanying drawings. The detailed description and drawings are merely illustrative of the present invention rather than limiting, the scope of the present invention being defined by the appended claims and equivalents thereof.
FIG. 1 illustrates multi-modal tissue classification in accordance with the present invention.
FIG. 2 illustrates an exemplary embodiment of a multi-modal tissue
classification workstation in accordance with the present invention.
FIG. 3 illustrates a flowchart representative of an exemplary embodiment of a multi-modal tissue classification method in accordance with the present invention.
FIG. 4 illustrates a first exemplary implementation of the flowchart illustrated in FIG. 3 by the system illustrated in FIG. 2.
FIG. 5 illustrates a second exemplary implementation of the flowchart illustrated in FIG. 3 by the system illustrated in FIG. 2.
Referring to FIG. 1, a MRI system 20 employs a scanner 21 and a workstation 22 to generate a MRI image of an anatomical tissue of a patient, such as, for example, a MRI image 23 of anatomical tissue of a prostate 11 of a patient 10 as shown. In practice, the present invention utilizes one or more MRI systems 20 of various types to acquire MRI features of the anatomical tissue. Examples of types of MRI systems and associated MRI features of an anatomical tissue include, but are not limited to, a T2W MRI system illustrating normalized intensity-value and/or texture based features of the anatomical tissue, a DWI-MRI system illustrating apparent diffusion coefficient ("ADC") of water in the anatomical tissue, a DCE-MRI system illustrating
pharmacokinetic parameters of the anatomical tissue, and MRSI system illustrating metabolic information of the anatomical tissue.
An ultrasound system 30 employs a probe 31 and a workstation 32 to generate an ultrasound image of an anatomical tissue of a patient, such as, for example, a US image 33 of the anatomical tissue of prostate 11 of patient 10 as shown. In practice, the present invention utilizes one or more ultrasound systems 30 of various types to acquire US image features of the anatomical tissue. Examples of types of ultrasound imaging systems 30 and associated US image features of the anatomical tissue include, but are not limited to, a B-mode US imaging system illustrating texture based features of the anatomical tissue and US RF echo imaging system illustrating spectral features of the anatomical tissue. The present invention performs various known techniques including, but not limited to, image segmentation, image registration, image feature extraction, image delineation and tissue classification, on MRI images and ultrasound images of the anatomical tissue to provide a systematic and objective approach to fuse feature information from the MRI images and the ultrasound images. Specifically, the present invention is premised on combining image features extracted from spatially registered MRI images (e.g., T2W-MRI, DWI-MRI, DCE-MRI and MRSI) and ultrasound images (e.g., a B-mode image and a RF echo image) in order to form a vector of image features for each voxel of a tissue classification volume, such as, for example, tissue
classification volume 40a and tissue classification volume 40b as shown in FIG. 1.
In practice, each voxel of a tissue classification volume is classified between a healthy tissue voxel and an unhealthy tissue voxel (e.g., cancerous) based on the image feature vector of the voxel derived from the extracted image features of the spatially registered MRI images and ultrasound images. The quantity and linearity of classifications is not limited by the present invention.
For example, each voxel of tissue classification volume 40a is classified as a being a healthy tissue voxel or an unhealthy tissue voxel (e.g., cancerous) based on the image feature vector of the voxel derived from the extracted image features of the spatially registered MRI images and ultrasound images. As shown in FIG. 1, the healthy tissue voxels are the white voxels and the unhealthy tissue voxels are the black voxels.
By further example, each voxel of tissue classification volume 40b is classified as being either a healthy tissue voxel, an unhealthy tissue voxel (e.g., cancerous) or a borderline unhealthy tissue voxel based on the image feature vector of the voxel derived from the extracted image features of the spatially registered MRI images and ultrasound images. As shown in FIG. 1, the healthy tissue voxels are the white voxels, the unhealthy tissue voxels are the black voxels and the borderline unhealthy tissue voxels are the gray voxels.
Another example, not shown in FIG. 1, is the utilization of a grayscale of voxels ranging from white healthy tissue voxels through multiple gray voxels indicating various probabilities of unhealthiness of corresponding tissue to black unhealthy tissue voxels (e.g., cancerous). To facilitate an understanding of the present invention, exemplary embodiments of the present invention will now be described herein as directed to a generation of tissue classification volumes of anatomical tissue of a prostate. For purposes of the present invention, the terms "image segmentation", "image (spatial) registration", "image delineation", "image feature extraction", "tissue classification" and
"dimensionality reduction" as well as related terms are to be broadly interpreted as known in the art of the present invention. Also, in practice, the present invention applies to any anatomical region (e.g., head, thorax, pelvis, etc.) and to anatomical tissue of any anatomical structure (e.g., organs).
Referring to FIG. 2, an exemplary workstation 50 employs modules 51-53 for implementing a flowchart 100 (FIG. 3) representative of a multi-modal tissue classification method of the present invention.
In operation, workstation 50 provides a multi-modal data input channel (not shown) for receiving one or more MRI images 70 and one or more US images 80.
Second, image registrator 51 employs technique(s) for spatial registration(s) of the voxels of MRI image(s) 70 and ultrasound image(s) 80.
Third, feature extractor 52 employs technique(s) for extracting image features from the voxels of MRI image(s) 70 and ultrasound image(s) 80.
In practice, as will be further described herein, image registrator 51 and feature extractor 52 operate in tandem to extract image features from the spatially registered voxels of MRI image(s) 70 and ultrasound image(s) 80 for purposes of generating a tissue classification volume having a voxel correspondence with the spatially registered MRI image(s) 70 and ultrasound image(s) 80 and for generating a concatenated n- dimensional image feature vector for each voxel of the tissue classification volume with n equaling a total number of image features.
Fourth, tissue classifier 53 employs technique(s) for classifying, unsupervised or supervised, a tissue type of each voxel of the tissue classification volume as indicated by the image feature vector of each voxel.
Finally, a user interface (not shown) of workstation 50 provides for the display of the tissue classification volume on a voxel-by-voxel basis including zoom and pan capabilities, such as, for example, the display of tissue classification volume 40a as shown in FIG. 2. In practice, workstation 50 may provide tools for modules 51-53 including, but not limited, to an image segmentation tool to segment voxels of the anatomical tissue from MRI images 70 and US images 80 and an image delineation tool to delineate voxels of MRI images 70 and US images 80 suspicious of being unhealthy (e.g., cancerous).
Also in practice, workstation 50 may be structurally configured in any manner suitable for implementing a multi-modal tissue classification method of the present invention, particularly flowchart 100. In one embodiment, workstation 50 is structurally configured with hardware/circuitry (e.g., processor(s), memory, etc.) for executing modules 51-53 programmed and installed as software/firmware within workstation 50.
Additionally, in practice, workstation 50 may be a stand-alone workstation providing tissue classification volume(s) to an image diagnostic system 90 (e.g., a MRI system or an ultrasound system) for incorporation into a diagnostic procedure as needed and/or an interventional guidance system 91 (e.g., an electromagnetic tracking system, optical tracking system or image tracking system) for incorporation into an
interventional procedure as needed. Alternatively, workstation 50 may be incorporated within image diagnostic system 90 or incorporated within interventional guidance system 91.
Exemplary executions of flowchart 100 (FIG. 3) by workstation 50 as shown in
FIGS. 4 and 5 will now be described herein in the context of MRI images 70 consisting of a T2W image 71, a DWI image 72, a DCE image 73 and a MRSI image 74, and US images 80 consisting of a B-mode image 81 and a RF echo image 82. From the description, those having ordinary skill in the art will appreciate alternative systems and devices for implementing multi-modal tissue classification methods of the present invention.
Referring to FIG. 3, a stage S101 of flowchart 100 encompasses image registrator 51 performing an intra-modal spatial registration and/or a multi-modal spatial registration for purposes of generating a tissue classification volume having voxel correspondence with the spatially registered images. For example, as shown in FIGS. 4 and 5, image registrator 51 first performs an intra-modal spatial registration of MRI images 70 and then performs a multi-modal spatial registration of MRI images 70 and US images 80 for purposes of generating a tissue classification volume having voxel correspondence with the spatially registered MIR images 70 and US images 80. An image segmentation tool (not shown) may be utilized by image registrator 51 prior to or after the spatial registration(s) to segment voxels of the anatomical tissue (e.g., a prostrate) from MRI images 70 and US images 80.
The resulting spatial registration of MRI images 70 and US images 80 are communicated to feature extractor 52 whereby a stage SI 02 of flowchart 100 encompasses feature extractor 52 extracting and concatenating image features from each voxel of the spatial registered MRI image(s) 70 and US image(s) 80 to generate a n-dimensional image feature vector for each voxel of the tissue classification volume.
For example, as shown in FIG. 4, a feature extractor 52a extracts normalized intensity-value and/or texture based features of the anatomical tissue from T2W image 71, an apparent diffusion coefficient ("ADC") of water of the anatomical tissue from DWI image 72, pharmacokinetic parameters of the anatomical tissue from DCE image 73, metabolic information of the anatomical tissue from MRS image 74, texture based image features of the anatomical tissue from B-mode image 81 and spectral image features of the anatomical tissue from RF image 82. From the spatially registered extractions, feature extractor 52a generates a vector 40a of six (6) dimensional image features for each voxel of the tissue classification volume.
If not utilized by image registrator 51, the image segmentation tool (not shown) may be utilized by feature extractor 52a prior to or after the generation of an image feature vector 40a for each voxel.
By further example, as shown in FIG. 5, a feature extractor 52b utilizes a MRI image delineator 54 to delineate voxels of MRI images 70 representative of unhealthy (e.g., cancerous) portions of the anatomical tissue and generates a weight Wi of a distance from a boundary of the anatomical structure for each voxel from spatially registered MRI images 70. Feature extractor 52b then extracts texture based image features of the anatomical tissue from B-mode image 81 and spectral image features of the anatomical tissue from RF image 82. From the spatially registered extractions, feature extractor 52b generates a weighted vector 40b of two (2) dimensional image features for each voxel of the tissue classification volume. If not utilized by image registrator 51, the image segmentation tool (not shown) may be utilized by feature extractor 52b prior to or after the generation of an image feature vector 40b for each voxel.
Upon completion of stage SI 02, the resulting image feature vector for each voxel (e.g., image feature vectors 40a or image feature vectors 40b) of the tissue classification volume is communicated to a tissue classifier 53 whereby a stage SI 03 of flowchart 100 encompasses tissue classifier 53 classifying, unsupervised or supervised, a tissue type of each voxel for the tissue classification volume (e.g., tissue classification volume 40a and tissue classification volume 40b) as indicated by the image feature vector of each voxel.
For example, all of the image features of the image feature vector of a particular voxel may indicate the associated anatomical tissue is healthy (i.e., normal cells or void of any cellular irregularities). Conversely, one or more of the image features of the image feature vector of a particular voxel may indicate the associated anatomical tissue may be or is unhealthy to some degree (i.e., abnormal cells or inclusive of any cellular irregularity).
Referring to FIG. 2, upon completion of flowchart 100, the tissue classification volume may be displayed by workstation 50, image diagnostic system 90 and/or interventional guidance system 91. Additionally, an overlay or fusion of the tissue classification volume may be displayed on/within an image of the anatomical tissue generated by image diagnostic system 90 and/or interventional guidance system 91.
Referring to FIGS. 1-5, those having ordinary skill in the art will appreciate numerous benefits of the present invention including, but not limited to, an
improvement in a sensitivity and specificity of cancer detection for any anatomical tissue (e.g., a prostate). As such, the present invention is applicable in all diagnostic and therapeutic scenarios that would benefit from the ability to localize a location and size of tumors within a gland/section of tissue.
Those having ordinary skill in the art will further appreciate the implementation of the present invention with alternative imaging modalities (e.g., computed
tomography and SPECT, PET, etc.).
While various embodiments of the present invention have been illustrated and described, it will be understood by those skilled in the art that the embodiments of the present invention as described herein are illustrative, and various changes and modifications may be made and equivalents may be substituted for elements thereof without departing from the true scope of the present invention. In addition, many modifications may be made to adapt the teachings of the present invention without departing from its central scope. Therefore, it is intended that the present invention not be limited to the particular embodiments disclosed as the best mode contemplated for carrying out the present invention, but that the present invention includes all embodiments falling within the scope of the appended claims.

Claims

Claims
1. A system for multi-modal tissue classification of an anatomical tissue, the system comprising:
at least one MRI system (20) structurally configured to generate at least one MRI feature of the anatomical tissue;
at least one ultrasound system (30) structurally configured to generate at least one ultrasound image feature of the anatomical tissue; and
a workstation (60) structurally configured to generate a tissue classification volume (40) of the anatomical tissue derived from a spatial registration and an image extraction of the at least one MRI feature and the at least one ultrasound image feature of the anatomical tissue,
wherein the workstation (60) is further structurally operable to classify each voxel of the tissue classification volume as one of a plurality tissue types including at least a healthy tissue voxel and an unhealthy tissue voxel.
2. The system of claim 1, wherein a generation of the tissue classification volume (40) includes:
the workstation (60) spatially registering and image extracting the at least one MRI feature and the at least one ultrasound image feature of the anatomical tissue; and the workstation (60) generating an image feature vector (41) for each voxel, each image feature vector (41) including a concatenation of a spatially registered and image extracted at least one MRI feature and at least one ultrasound image feature of the anatomical tissue associated with a corresponding voxel.
3. The system of claim 2, wherein a classification of each voxel of the tissue classification volume (40) as one of the tissue types is derived from the concatenation of image features of the anatomical tissue included within a corresponding image feature vector (41).
4. The system of claim 1, wherein a generation of the tissue classification volume (40) includes: the workstation (60) image extracting at least two ultrasound image features of the anatomical tissue; and
the workstation (60) generating an image feature vector (41) for each voxel, each image feature vector (41) including a concatenation of the image extracted at least two ultrasound image features of the anatomical tissue associated with a corresponding voxel.
5. The system of claim 4, wherein the generation of the tissue classification volume (40) further includes:
the workstation spatially registering the at least one MRI feature and the at least two ultrasound image features;
the workstation (60) generating a weighted factor for each voxel, each weighted factor being derived from a delineation of each spatially registered at least one MRI feature of the anatomical tissue representative of an unhealthy portion of the anatomical tissue; and
the workstation (60) applying each weighted factor to a corresponding image feature vector (41).
6. The system of claim 5, wherein a classification of each voxel of the tissue classification volume (40) as one of the tissue types is derived from the concatenation of image features of the anatomical tissue within a corresponding image feature vector (41).
7. The system of claim 1, wherein the plurality of tissue types further includes a borderline unhealthy tissue voxel.
8. The system of claim 1, wherein the plurality of tissue types further includes a plurality of probable unhealthy tissue voxels ranging between the healthy tissue voxel and the unhealthy tissue voxel.
9. The system of claim 1, further comprising: an image diagnostic system (90) structurally configured for incorporating the tissue classification volume (40) into a diagnostic procedure.
10. The system of claim 1, further comprising:
an interventional guidance system (91) structurally configured for incorporating the tissue classification volume (40) into an interventional procedure.
11. A workstation (60) for multi-modal tissue classification of an anatomical tissue, the workstation (60) comprising:
an image registrator (51) and a feature extractor (52) structurally configured to generate a tissue classification volume (40) of the anatomical tissue derived from a spatial registration and an image extraction of at least one MRI feature of the anatomical tissue and at least one ultrasound image feature of the anatomical tissue; and
a tissue classifier (53) structurally configured to classify each voxel of the tissue classification volume (40) as one of a tissue type including at least a healthy tissue voxel and an unhealthy tissue voxel.
12. The workstation (60) of claim 11, a generation of the tissue classification volume (40) includes:
the image register (51) spatially registering the at least one MRI feature and the at least one ultrasound image feature of the anatomical tissue;
the feature extractor (52) extracting the spatially registered at least one MRI feature and at least one ultrasound image feature of the anatomical tissue; and
the feature generator (52) generating an image feature vector (41) for each voxel, each image feature vector (41) including a concatenation of the spatially registered and image extracted at least one MRI feature and at least one ultrasound image feature of the anatomical tissue associated with a corresponding voxel.
13. The system of claim 12, wherein a classification of each voxel of the tissue classification volume (40) as one of the tissue types is derived from the concatenation of image features of the anatomical tissue included within a corresponding image feature vector (41).
14. The workstation (60) of claim 11, wherein a generation of the tissue classification volume (40) includes:
the image register (51) spatially registering the at least one MRI feature and at least two ultrasound image features of the anatomical tissue;
the feature extractor (52) image extracting the at least two ultrasound image features of the anatomical tissue;
the feature extractor (52) generating an image feature vector (41) for each voxel, each image feature vector (41) including a concatenation of the image extracted at least two ultrasound image features of the anatomical tissue associated with a corresponding voxel;
the feature extractor (52) generating a weighted factor for each voxel, each weighted factor being derived from a delineation of each spatially registered at least one MRI feature of the anatomical tissue representative of a suspicious unhealthy portion of the anatomical tissue; and
the feature extractor (52) applying each weighted factor to each image feature vector (41) of the corresponding voxel.
15. The workstation (60) of claim 14, wherein a classification of each voxel of the tissue classification volume (40) as one of the tissue types is derived from the concatenation of image features of the anatomical tissue included within a
corresponding image feature vector (41).
16. A method for multi-modal tissue classification of an anatomical tissue, the method comprising:
generating a a tissue classification volume (40) of the anatomical tissue derived from a spatial registration and an image extraction of at least one MRI feature and at least one ultrasound image feature of the anatomical tissue; and classifying each voxel of the tissue classification volume (40) as one of a plurality of tissue types including at least a healthy tissue voxel and an unhealthy tissue voxel.
17. The method of claim 16, wherein the generation of the tissue classification volume (40):
spatially registering and image extracting the at least one MRI feature and the at least one ultrasound image feature of the anatomical tissue; and
generating an image feature vector (41) for each voxel, each image feature vector (41) including a concatenation of the spatially registered and image extracted at least one MRI feature and at least one ultrasound image feature of the anatomical tissue associated with a corresponding voxel.
18. The method of claim 17, wherein a classification of each voxel of the tissue classification volume (40) as one of the tissue types is derived from the concatenation of image features of the anatomical tissue included within a corresponding image feature vector (41).
19 The method of claim 16, wherein the generation of the tissue classification volume (40) includes:
spatially registering the at least one MRI feature and the at least one ultrasound image feature of the anatomical tissue;
image extracting the at least one ultrasound image feature of the anatomical tissue;
generating an image feature vector (41) for each voxel, each image feature vector (41) including a concatenation of the image extracted at least one ultrasound image feature of the anatomical tissue associated with a corresponding voxel;
generating a weighted factor for each voxel, each weighted factor being derived from a delineation of each spatially registered at least one MRI feature of the anatomical tissue representative of a suspicious unhealthy portion of the anatomical tissue; and applying each weighted factor to each image feature vector (41) of the corresponding voxel.
20. The method of claim 19, wherein a classification of each voxel of the tissue classification volume (40) as one of the tissue types is derived from the concatenation of image features of the anatomical tissue included within a corresponding image feature vector (41).
PCT/IB2014/062422 2013-06-26 2014-06-19 Method and system for multi-modal tissue classification WO2014207627A1 (en)

Priority Applications (4)

Application Number Priority Date Filing Date Title
US14/891,768 US11660065B2 (en) 2013-06-26 2014-06-19 Method and system for multi-modal tissue classification
CN201480036465.0A CN105338905B (en) 2013-06-26 2014-06-19 Method and system for multi-modal tissue typing
JP2016522913A JP6445548B2 (en) 2013-06-26 2014-06-19 System for multi-mode tissue classification
EP14738901.9A EP3013241B1 (en) 2013-06-26 2014-06-19 System for multi-modal tissue classification

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US201361839424P 2013-06-26 2013-06-26
US61/839,424 2013-06-26

Publications (1)

Publication Number Publication Date
WO2014207627A1 true WO2014207627A1 (en) 2014-12-31

Family

ID=51177105

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IB2014/062422 WO2014207627A1 (en) 2013-06-26 2014-06-19 Method and system for multi-modal tissue classification

Country Status (5)

Country Link
US (1) US11660065B2 (en)
EP (1) EP3013241B1 (en)
JP (1) JP6445548B2 (en)
CN (1) CN105338905B (en)
WO (1) WO2014207627A1 (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2017165801A1 (en) * 2016-03-24 2017-09-28 The Regents Of The University Of California Deep-learning-based cancer classification using a hierarchical classification framework
JP2018516703A (en) * 2015-06-12 2018-06-28 コーニンクレッカ フィリップス エヌ ヴェＫｏｎｉｎｋｌｉｊｋｅ Ｐｈｉｌｉｐｓ Ｎ．Ｖ． System and method for identifying cancer tissue
US20210327068A1 (en) * 2020-04-18 2021-10-21 HighRAD Ltd. Methods for Automated Lesion Analysis in Longitudinal Volumetric Medical Image Studies
CN117218419A (en) * 2023-09-12 2023-12-12 河北大学 Evaluation system and evaluation method for pancreatic and biliary tumor parting and grading stage

Families Citing this family (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10716536B2 (en) 2013-07-17 2020-07-21 Tissue Differentiation Intelligence, Llc Identifying anatomical structures
US10154826B2 (en) 2013-07-17 2018-12-18 Tissue Differentiation Intelligence, Llc Device and method for identifying anatomical structures
US11734592B2 (en) 2014-06-09 2023-08-22 Tecnotree Technologies, Inc. Development environment for cognitive information processing system
JP6612861B2 (en) * 2014-07-02 2019-11-27 コーニンクレッカ フィリップス エヌ ヴェ System and method for identifying an organization
US10909675B2 (en) * 2015-10-09 2021-02-02 Mayo Foundation For Medical Education And Research System and method for tissue characterization based on texture information using multi-parametric MRI
US11701086B1 (en) 2016-06-21 2023-07-18 Tissue Differentiation Intelligence, Llc Methods and systems for improved nerve detection
EP3923237A1 (en) * 2017-02-22 2021-12-15 The United States of America as represented by The Secretary Department of Health and Human Services Detection of prostate cancer in multi-parametric mri using random forest
EP3511866A1 (en) * 2018-01-16 2019-07-17 Koninklijke Philips N.V. Tissue classification using image intensities and anatomical positions
CN109316202B (en) * 2018-08-23 2021-07-02 苏州佳世达电通有限公司 Image correction method and detection device
US11645620B2 (en) 2019-03-15 2023-05-09 Tecnotree Technologies, Inc. Framework for explainability with recourse of black-box trained classifiers and assessment of fairness and robustness of black-box trained classifiers
WO2020223798A1 (en) * 2019-05-03 2020-11-12 Huron Technologies International Inc. Image diagnostic system, and methods of operating thereof
CN110674872B (en) * 2019-09-24 2022-03-01 广州大学 High-dimensional magnetic resonance image classification method and device
CN114650777A (en) 2019-11-08 2022-06-21 三星麦迪森株式会社 Medical image display apparatus and method of displaying medical image using the same
CN111080658A (en) * 2019-12-16 2020-04-28 中南民族大学 Cervical MRI image segmentation method based on deformable registration and DCNN

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130046168A1 (en) * 2011-08-17 2013-02-21 Lei Sui Method and system of characterization of carotid plaque

Family Cites Families (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3974946B2 (en) * 1994-04-08 2007-09-12 オリンパス株式会社 Image classification device
US6317617B1 (en) 1997-07-25 2001-11-13 Arch Development Corporation Method, computer program product, and system for the automated analysis of lesions in magnetic resonance, mammogram and ultrasound images
CA2529929A1 (en) * 2003-06-25 2005-01-06 Siemens Medical Solutions Usa, Inc. Systems and methods for automated diagnosis and decision support for breast imaging
US20050075566A1 (en) * 2003-09-19 2005-04-07 Fuji Photo Film Co., Ltd. Ultrasonice diagnosing apparatus
JP2005111258A (en) * 2003-09-19 2005-04-28 Fuji Photo Film Co Ltd Ultrasonic diagnostic apparatus
US20080132782A1 (en) 2004-02-02 2008-06-05 Rueckmann Bogdan Von Combined MR-ultrasound (US) coil for prostate-, cevix- and rectum cancer imaging diagnostics
WO2005111932A2 (en) 2004-05-14 2005-11-24 Philips Intellectual Property & Standards Gmbh Information enhanced image guided interventions
EP1901810B1 (en) * 2005-05-26 2010-10-27 Koninklijke Philips Electronics N.V. Radio-therapeutic treatment planning incorporating functional imaging information
WO2007019216A1 (en) 2005-08-04 2007-02-15 Teratech Corporation Integrated ultrasound and magnetic resonance imaging system
JP2008036284A (en) * 2006-08-09 2008-02-21 Toshiba Corp Medical image composition method and its apparatus
US20110178389A1 (en) 2008-05-02 2011-07-21 Eigen, Inc. Fused image moldalities guidance
JP5429517B2 (en) * 2008-09-03 2014-02-26 国立大学法人金沢大学 Diagnosis support system, method and computer program
US9521994B2 (en) 2009-05-11 2016-12-20 Siemens Healthcare Gmbh System and method for image guided prostate cancer needle biopsy
GB0913930D0 (en) * 2009-08-07 2009-09-16 Ucl Business Plc Apparatus and method for registering two medical images
US20110137148A1 (en) 2009-12-07 2011-06-09 Irvine Sensors Corporation Method and device comprising fused ultrasound and magnetic resonance imaging
JP2012075702A (en) * 2010-10-01 2012-04-19 Fujifilm Corp Apparatus, method, and program for reconstructing intra-tubular-structure image
JPWO2012063928A1 (en) * 2010-11-11 2014-05-12 オリンパスメディカルシステムズ株式会社 Ultrasonic observation apparatus, operation method of ultrasonic observation apparatus, and operation program of ultrasonic observation apparatus
US9208556B2 (en) 2010-11-26 2015-12-08 Quantitative Insights, Inc. Method, system, software and medium for advanced intelligent image analysis and display of medical images and information
DE102011085894B4 (en) * 2011-11-08 2014-03-27 Siemens Aktiengesellschaft Method and electronic computing device for imaging and imaging system
GB201121307D0 (en) * 2011-12-12 2012-01-25 Univ Stavanger Probability mapping for visualisation of biomedical images

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130046168A1 (en) * 2011-08-17 2013-02-21 Lei Sui Method and system of characterization of carotid plaque

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
BERNARD CHIU ET AL: "Characterization of carotid plaques on 3-dimensional ultrasound imaging by registration with multicontrast magnetic resonance imaging", JOURNAL OF ULTRASOUND IN MEDICINE : OFFICIAL JOURNAL OF THE AMERICAN INSTITUTE OF ULTRASOUND IN MEDICINE, 2 October 2012 (2012-10-02), United States, pages 1567 - 1580, XP055136953, Retrieved from the Internet <URL:http://www.ncbi.nlm.nih.gov/pubmed/23011620> [retrieved on 20140827] *
JONATHAN CHAPPELOW ET AL: "Improving supervised classification accuracy using non-rigid multimodal image registration: detecting prostate cancer", PROCEEDINGS OF SPIE, vol. 6915, 6 March 2008 (2008-03-06), pages 69150V, XP055136936, ISSN: 0277-786X, DOI: 10.1117/12.770703 *
KRUECKER ET AL: "Fusion ofreal-time trans-rectal ultrasound with pre-acquired MRI for multi-modality prostateimaging", PROCEEDINGS OF SPIE, S P I E - INTERNATIONAL SOCIETY FOR OPTICAL ENGINEERING, US, vol. 6509, 21 March 2007 (2007-03-21), pages 650912/1, XP009113917, ISSN: 0277-786X, DOI: 10.1117/12.710344 *
SHOGO NAKANO ET AL: "Impact of real-time virtual sonography, a coordinated sonography and MRI system that uses an image fusion technique, on the sonographic evaluation of MRI-detected lesions of the breast in second-look sonography", BREAST CANCER RESEARCH AND TREATMENT, KLUWER ACADEMIC PUBLISHERS, BO, vol. 134, no. 3, 24 July 2012 (2012-07-24), pages 1179 - 1188, XP035093085, ISSN: 1573-7217, DOI: 10.1007/S10549-012-2163-9 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2018516703A (en) * 2015-06-12 2018-06-28 コーニンクレッカ フィリップス エヌ ヴェＫｏｎｉｎｋｌｉｊｋｅ Ｐｈｉｌｉｐｓ Ｎ．Ｖ． System and method for identifying cancer tissue
WO2017165801A1 (en) * 2016-03-24 2017-09-28 The Regents Of The University Of California Deep-learning-based cancer classification using a hierarchical classification framework
US10939874B2 (en) 2016-03-24 2021-03-09 The Regents Of The University Of California Deep-learning-based cancer classification using a hierarchical classification framework
US20210327068A1 (en) * 2020-04-18 2021-10-21 HighRAD Ltd. Methods for Automated Lesion Analysis in Longitudinal Volumetric Medical Image Studies
CN117218419A (en) * 2023-09-12 2023-12-12 河北大学 Evaluation system and evaluation method for pancreatic and biliary tumor parting and grading stage
CN117218419B (en) * 2023-09-12 2024-04-12 河北大学 Evaluation system and evaluation method for pancreatic and biliary tumor parting and grading stage

Also Published As

Publication number Publication date
EP3013241A1 (en) 2016-05-04
US11660065B2 (en) 2023-05-30
US20160098621A1 (en) 2016-04-07
CN105338905A (en) 2016-02-17
JP6445548B2 (en) 2018-12-26
CN105338905B (en) 2019-10-18
JP2016523165A (en) 2016-08-08
EP3013241B1 (en) 2020-01-01

Similar Documents

Publication Publication Date Title
US11660065B2 (en) Method and system for multi-modal tissue classification
Sun et al. Multiparametric MRI and radiomics in prostate cancer: a review
Liu et al. Computer-aided detection of prostate cancer with MRI: technology and applications
Vilanova et al. Usefulness of prebiopsy multifunctional and morphologic MRI combined with free-to-total prostate-specific antigen ratio in the detection of prostate cancer
Orel et al. MR imaging of the breast for the detection, diagnosis, and staging of breast cancer
Jacobs et al. Benign and malignant breast lesions: diagnosis with multiparametric MR imaging
Jackson et al. Dynamic contrast-enhanced MRI for prostate cancer localization
EP3307173B1 (en) System for identifying cancerous tissue
Moradi et al. Computer-aided diagnosis of prostate cancer with emphasis on ultrasound-based approaches: A review
Chung et al. Prostate cancer detection via a quantitative radiomics-driven conditional random field framework
McClure et al. In-vitro and in-vivo diagnostic techniques for prostate cancer: A review
Chung et al. Discovery radiomics for multi-parametric MRI prostate cancer detection
US20090069665A1 (en) Automatic Lesion Correlation in Multiple MR Modalities
Firjani et al. A novel image-based approach for early detection of prostate cancer
Sammouda et al. Intelligent computer-aided prostate cancer diagnosis systems: state-of-the-art and future directions
Ampeliotis et al. A computer-aided system for the detection of prostate cancer based on magnetic resonance image analysis
Reda et al. A new fast framework for early detection of prostate cancer without prostate segmentation
Chen et al. A self-tuned graph-based framework for localization and grading prostate cancer lesions: An initial evaluation based on multiparametric magnetic resonance imaging
Freidlin et al. Application of an unsupervised multi-characteristic framework for intermediate-high risk prostate cancer localization using diffusion-weighted MRI
Shi et al. Histogram array and convolutional neural network of DWI for differentiating pancreatic ductal adenocarcinomas from solid pseudopapillary neoplasms and neuroendocrine neoplasms
Lenkinski The role of magnetic resonance imaging in prostate cancer imaging and staging
Gangeh et al. Quantitative ultrasound spectroscopy and a kernel-based metric in clinical cancer response monitoring
Fei 15 Precision Imaging of
Fei Precision Imaging of Prostate Cancer: Computer-Aided Detection and Their Clinical Applications
Chen et al. A graph-based multi-kernel feature weight learning framework for detection and grading of prostate lesions using multi-parametric MR images

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 201480036465.0

Country of ref document: CN

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 14738901

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 14891768

Country of ref document: US

WWE Wipo information: entry into national phase

Ref document number: 2014738901

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 2016522913

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE