CN114596256A - Image analysis method for extracting space curvature characteristics of fibrous structure based on space orientation change in neighborhood - Google Patents
Image analysis method for extracting space curvature characteristics of fibrous structure based on space orientation change in neighborhood Download PDFInfo
- Publication number
- CN114596256A CN114596256A CN202210065555.9A CN202210065555A CN114596256A CN 114596256 A CN114596256 A CN 114596256A CN 202210065555 A CN202210065555 A CN 202210065555A CN 114596256 A CN114596256 A CN 114596256A
- Authority
- CN
- China
- Prior art keywords
- image
- space
- dimensional
- curvature
- fibrous structure
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
Images
Classifications
-
- 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
- G06T7/60—Analysis of geometric attributes
- G06T7/64—Analysis of geometric attributes of convexity or concavity
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Geometry (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an image analysis method for extracting space curvature characteristics of a fibrous structure based on space orientation change in a neighborhood, which is used for quantitative analysis of an optical image of the fibrous structure of a biological tissue. The method can realize the full-automatic extraction of the space curvature characteristics of the fibrous structure with the pixel level resolution level, and also has excellent time efficiency while improving the precision and the accuracy; the invention also utilizes the pseudo-color coding technology to intuitively reflect the distribution and the change condition of the space curvature of the biological tissue, thereby enhancing the information readability; the invention realizes the quantitative analysis of the three-dimensional image, and greatly enhances the application range and potential of the space curvature parameter compared with the traditional two-dimensional analysis; by combining with other fibrous structure quantitative parameters, the invention can realize more comprehensive and complete understanding of biological tissues and has profound significance for researching and analyzing life activities and disease evolution processes.
Description
Technical Field
The invention belongs to the technical field of image processing and quantitative characterization of biological tissues, and particularly relates to an image analysis method for extracting space curvature characteristics of a fibrous structure based on space orientation change in a neighborhood.
Background
Fibrillar structures are ubiquitous in biological tissues and extend in size from molecular level to tissue level, including but not limited to endoplasmic reticulum, cellular microtubules, actin, neuronal axons, collagen fibrils, and blood vessels. These fibrous structures undergo dynamic remodeling in a series of life processes, which are expressed as subtle changes at the morphological and structural levels and are closely related to the development, damage, healing, and development and evolution of biological tissues. Research has shown that quantitative characterization of the morphological structure of fibrous tissues, including density, spatial orientation, spatial arrangement, etc., is of crucial importance for understanding the structure and function of living systems, cell-matrix interactions, and has been widely used in the study of cancer, diabetes, and cardiovascular and neurodegenerative diseases.
The spatial curvature is another important characteristic parameter describing the degree of bending of fibrous structures and is closely related to the mechanical properties of biological tissues. However, the method capable of quantitatively characterizing the spatial curvature is very lacking, and only a few methods based on semi-automatic fiber tracking technology are used to roughly estimate the spatial curvature of the fibrous structure by calculating the ratio of the longitudinal length and the transverse length of the fiber obtained by the tracking technology, however, such parameters hardly reflect the local curvature characteristics of the fibrous structure, and the accuracy thereof is highly dependent on the fiber tracking result, and is likely to be seriously influenced by the density, the diameter and other factors of the fibrous tissue. However, in processes such as disease progression, particularly early stages of the disease, the structural changes of the fibrous tissue tend to be very subtle and difficult to capture. In addition, many of the existing quantitative analysis tools for fibrous structures only stay at two-dimensional level, and with the development of imaging technology, more and more three-dimensional structures of biological tissues can be clearly revealed, so that the quantitative tools need to be expanded to three-dimensional. In view of the above, there is a need for a method for quantifying spatial curvature with higher precision, accuracy, automation, and pixel level resolution, which can be applied to three-dimensional fibrous structures.
Disclosure of Invention
Based on the defects of the prior art, the invention provides an image analysis method for extracting the space curvature characteristics of a fibrous structure based on the orientation change of the space in the neighborhood. The method can obtain the space curvature information of the fibrous structure with pixel level resolution, the value is 0-1, the higher the space curvature value is, the more curved fibrous structure is represented, and the space curvature information of the fibrous structure can be clearly displayed through different colors in a pseudo color coding pattern mode. The method can be used for processing and analyzing two-dimensional and three-dimensional fibrous structure images obtained by various optical microscopy technologies, and has the advantages of rapidness, simplicity, accuracy, strong applicability and the like.
The invention is realized by adopting the following technical scheme:
an image analysis method for extracting space curvature characteristics of a fibrous structure based on space orientation change in a neighborhood comprises the following steps:
1) the calculation of the pixel-level spatial orientation is performed for a two-dimensional/three-dimensional image of a fibrous structure, for the two-dimensional image, two-dimensional windows containing pixels in the neighborhood are generated respectively centering on all pixels of the image, and vectors along different directions are generated within the windows, then the vectors of different directions are weighted according to the intensity variation of the image signal within the window and the distance between the non-central pixel and the central pixel, finally all direction vectors are summed, and the spatial orientation value of the sum vector is given to the central pixel. For a three-dimensional image, a three-dimensional window containing pixels in the neighborhood is generated at all voxels of the image, three two-dimensional images are obtained by projection along the x, y and z directions, and then the spatial orientations of the three projection images are respectively calculated according to a two-dimensional calculation method, so that the three-dimensional orientation of the central pixel is obtained.
2) And performing thresholding processing on the two-dimensional/three-dimensional image of the fiber structure to obtain a binary matrix, wherein a 0 value represents a background pixel, and a 1 value represents a pixel containing fiber structure information and needing space curvature analysis.
3) Generating a two-dimensional/three-dimensional window at all effective pixels (1-value pixels), wherein the window contains the spatial orientation information obtained in the step 1). Then, the spatial curvature is calculated, and the calculation process is as follows: firstly, the spatial orientation values theta of all the non-central effective pixels in the window are respectively calculatednAnd central pixel spatial orientation thetacDifference value delta ofnThe change of the spatial orientation in the neighborhood of the window is quantified by calculating N times, wherein N is the number of non-central pixels in the window. For the fibrous structure data, when the difference is δnWhen the value is negative or more than 90 degrees, correction is needed when delta isnWhen the value is negative, the absolute value of delta is requirednWhen the angle is larger than 90 degrees, the supplementary angle is needed. Finally all corrected differences delta are addednThe sum is taken and the resulting value is divided by N and then by 90 to obtain the final normalized spatial curvature value.
4) And finally, carrying out pseudo-color coding on the image according to the calculated space curvature value so as to more clearly express the space curvature value, thereby revealing the bending degree of the fibrous structure in the image.
In the above technical solution, further, the generated vectors in different directions are weighted according to the intensity change of the image signal in the window and the distance between the pixel and the central pixel, and a specific calculation formula is as follows:
where W is the weighted vector modulus, L is the average distance of the non-central pixels relative to the central pixel, aiFor pixel signal intensity values in different vector directions,
is the average of pixel signals in all vector directions。
Compared with the background art, the invention has the beneficial effects that:
1. compared with a semi-automatic fiber tracking technology, the method can completely and automatically calculate the space curvature of the fibrous structure in the image without being influenced by the precision of fiber extraction, and improves the reliability of the space curvature parameter.
2. The method provided by the invention can process two-dimensional and three-dimensional images, and can be used for more completely analyzing the structural information of fibrous tissues provided by three-dimensional imaging compared with the method for calculating the two-dimensional images by using a fiber extraction technology in the background technology, so that the application scenes are very wide. The method can realize more comprehensive and complete understanding of biological tissues and has profound significance for researching and analyzing life activities and disease evolution processes.
3. Compared with the technology of quantifying the space curvature by calculating the ratio of the longitudinal length to the transverse length of the fiber obtained by tracking in the background art, the method can obtain the space curvature information of a pixel/voxel resolution level. For the former, a complete fibrous tissue can only obtain a single parameter to describe the integral bending degree of the fibrous structure, but the method provided by the invention can perform independent space curvature representation on each site of the fibrous tissue, so that the fibrous structure can be analyzed more accurately and precisely.
4. The fiber tracking technology has higher calculation cost, and the algorithm provided by the invention has higher time efficiency, and can complete the calculation of the space curvature of a three-dimensional image with the size of 512 multiplied by 100 within less than one minute, so that the algorithm has the potential of being a real-time clinical auxiliary diagnostic tool.
5. The pseudo-color coding technology used by the invention can more clearly display different space curvature information through different colors, so that the information about the distribution and the change of the space curvature is more easily obtained by a user, and the pseudo-color coding technology has stronger information readability compared with the background technology.
Drawings
Fig. 1 is a schematic diagram of a calculation process of spatial curvature: FIG. 1(a) is a binarized matrix of fibrous structures against the background, where white pixels (1 value) are cellulose pixels and black pixels (0 value) are background pixels; FIG. 1(b) is a pseudo-color-coded map of the spatial orientation of a fibrous structure; FIG. 1(c) is the binarization window generated in FIG. 1 (a); FIG. 1(d) is the window generated in FIG. 1(b), which contains spatial orientation information in a neighborhood, with arrows representing spatial orientation values; FIG. 1(e) is a flow chart of a process for calculating a spatial curvature; FIG. 1(f) is a two-dimensional curvature color-coded map; FIG. 1(g) is a three-dimensional curvature color-coded map.
FIG. 2 is a schematic representation of quantitative analysis of mouse cerebral vessels using spatial curvature: FIG. 2(a) is a demonstration of three-photon fluorescence imaging of craniotomy mice; FIG. 2(b) is a three-dimensional reconstructed image of mouse brain blood vessel three-photon fluorescence; FIG. 2(c) is a three-dimensional reconstructed image of a blood vessel space curvature pseudo-color coding map; FIG. 2(d) is a two-dimensional three-photon fluorescence image of a blood vessel, a spatial curvature pseudo-color encoding map; FIG. 2(e) is a line graph of spatial curvature of the cerebral vasculature as a function of depth; FIG. 2(f) is a graph showing the numerical relationship between vessel diameter and spatial curvature; FIG. 2(g) is a schematic diagram of a three-dimensional space curvature analysis of a blood vessel; FIG. 2(h) is a schematic view of a two-dimensional spatial curvature analysis of a blood vessel, wherein two-dimensional images are analyzed layer by layer according to depth; FIG. 2(i) (j) is a graph comparing the results of two-dimensional three-dimensional analysis of blood vessels.
FIG. 3 is a schematic representation of the use of spatial curvature to diagnose normal and cancerous pancreatic tissue: fig. 3(a) is a second harmonic (green)/two-photon fluorescence (purple) image of cancer pancreatic cancer tissue; FIG. 3(b) is a second harmonic (green)/two-photon fluorescence (purple) image of normal pancreatic cancer tissue; FIGS. 3(c) - (f) are second harmonic images, spatial orientation pseudo-colorcode, direction variance pseudo-colorcode and spatial curvature pseudo-colorcode of collagen fibers of cancer pancreatic tissue; FIGS. 3(g) - (j) are second harmonic images, spatial orientation pseudo-color patterns, direction variance pseudo-color patterns and spatial curvature pseudo-color patterns of normal pancreatic tissue collagen fibers; fig. 3(k) is a quantification of normal and cancer pancreatic tissue using image signal intensity, directional variance and spatial curvature.
Detailed Description
The following detailed description of the embodiments of the present invention is provided in conjunction with the accompanying drawings and examples, but the present invention is not limited thereto.
An image analysis method for extracting space curvature characteristics of a fibrous structure based on space orientation change in a neighborhood comprises the following steps:
1) the calculation of the pixel-level spatial orientation is performed for a two-dimensional/three-dimensional image of a fibrous structure, for the two-dimensional image, two-dimensional windows containing pixels in the neighborhood are generated respectively centering on all pixels of the image, and vectors along different directions are generated within the windows, then the vectors of different directions are weighted according to the intensity variation of the image signal within the window and the distance between the non-central pixel and the central pixel, finally all direction vectors are summed, and the spatial orientation value of the sum vector is given to the central pixel. For a three-dimensional image, a three-dimensional window containing pixels in the neighborhood is generated at all voxels of the image, three two-dimensional images are obtained by projection along the x, y and z directions, and then the spatial orientations of the three projection images are respectively calculated according to a two-dimensional calculation method, so that the three-dimensional orientation of the central pixel is obtained.
2) Thresholding is performed on the two-dimensional/three-dimensional image of the fiber structure to obtain a binary matrix, where 0 represents a background pixel and 1 represents a pixel containing fiber structure information that needs to be subjected to space curvature analysis, as shown in fig. 1 (a).
3) A two-dimensional/three-dimensional window is generated at all the effective pixels (1-value pixels), and the window contains the spatial orientation information obtained in step 1), as shown in fig. 1(d), wherein the arrows represent the spatial orientation of the pixels. Next, the spatial curvature is calculated, and the calculation process is as shown in fig. 1 (e). Firstly, the spatial orientation values theta of all the non-central effective pixels in the window are respectively calculatednAnd central pixel spatial orientation thetacDifference value delta ofnThe change of the spatial orientation in the neighborhood of the window is quantified by calculating N times, wherein N is the number of non-central pixels in the window. For the fibrous structure data, when the difference is δnIs a negative value orWhen the angle is more than 90 degrees, correction is needed, and when delta is larger thannWhen the value is negative, the absolute value of delta is requirednWhen the angle is larger than 90 degrees, the supplementary angle is needed. Finally, all the corrected spatial orientation differences are summed, and the finally obtained value is divided by N and then divided by 90 to obtain the final normalized spatial curvature value.
4) Finally, the image is pseudo-color coded according to the calculated spatial curvature value to more clearly express the value of the spatial curvature, thereby revealing the degree of curvature of the fibrous structure in the image, as shown in fig. 1(f), (g).
The method can be used for quantitative analysis of optical images of fibrous structures of biological tissues. The blood vessel is one of the most typical fibrous structures in an animal body, the invention combines a three-dimensional image obtained by an in-vivo three-photon excitation fluorescence microscope to analyze the cerebral blood vessel of a mouse, the three-dimensional reconstruction model of the analyzed three-photon fluorescence image and a corresponding space curvature pseudo color code pattern is shown in figures 2(b) and (c), and the imaging depth is about 800 mu m. A representative two-dimensional blood vessel image at 490 μm depth and corresponding spatial curvature pseudo-color coded map is shown in fig. 2(d), in which a more curved blood vessel indicated by an arrow shows a higher spatial curvature value, and a more straight blood vessel indicated by an arrow shows a lower spatial curvature value, which shows the accuracy and precision of the spatial curvature parameter developed by the present invention for the analysis of the blood vessel structure. Fig. 2(e) shows the change in spatial curvature of mouse cerebral vessels with the depth of imaging. The spatial curvature generally tends to increase with increasing imaging depth, and the result is a sensitivity of the spatial curvature parameter to structural changes in the biological tissue. Fig. 2(f) shows the negative correlation between the spatial curvature value of the mouse cerebral blood vessel and the numerical value of the blood vessel diameter, thicker blood vessels have lower spatial curvature values, thinner blood vessels have higher spatial curvature values, and the synergistic analysis of the spatial curvature and the diameter of the blood vessels shows that the spatial curvature parameter provided by the invention can provide more comprehensive fibrous tissue structure information by combining with other quantitative parameters.
The image analysis method for extracting the space curvature characteristics of the fibrous structure based on the orientation change in the neighborhood space can be used for analyzing the three-dimensional biological structure, and has great advantages compared with two-dimensional analysis. FIG. 2(g) (h) shows the three-dimensional and two-dimensional spatial curvature analysis of mouse cerebral vessels, respectively. Fig. 2(i) (j) shows the results of three-dimensional and two-dimensional spatial curvature analysis of blood vessels parallel to the imaging plane and perpendicular to the imaging plane for two types of cerebral vessels, respectively. For vessels parallel to the imaging plane, the two-dimensional and three-dimensional spatial curvatures reveal similar values, while for vessels perpendicular to the imaging plane the two-dimensional spatial curvature results are much higher than the three-dimensional ones, which is the case because vessels perpendicular to the imaging plane exhibit a structure of dots in the 2D image instead of a fibrous structure, resulting in errors in the two-dimensional curvature analysis results. Such results show the importance of the three-dimensional image processing capabilities of the spatial curvature parameters developed by the present invention.
In addition to blood vessels, collagen fibers are another important fibrous structure in biological tissues that is ubiquitous in the organism. The present invention combines the analysis of spatial curvature with collagen fiber images of pancreatic tissue obtained by second harmonic microscopy and distinguishes normal from cancerous pancreatic tissue. Fig. 3(a) and (b) show second harmonic (green)/two-photon imaged (purple) images of cancer and normal pancreatic tissue, respectively. Fig. 3(c) and (f) are an enlarged second harmonic signal image and a corresponding spatial curvature image of the collagen fiber-rich region of the cancer pancreatic tissue in fig. 3 (a). FIG. 3(g) (j) is an enlarged second harmonic signal image and corresponding spatial curvature image of the collagen fiber-rich region of normal pancreatic tissue of FIG. 3 (b). The spatial curvature of the two has high discrimination, and compared with normal tissues, the spatial curvature of cancer tissues is lower, which reveals that the extracellular matrix abnormality influenced by the cancer progression causes the remodeling of collagen fibers, and shows the sensitivity of the spatial curvature parameter provided by the invention. Fig. 3(k) shows the spatial curvature together with the other two fibrous structure parameters: the image signal intensity and direction variance, and the result diagram of three parameter multidimensional characterization collagen fibers. The combination of the three parameters can correctly distinguish the cancer from the normal pancreatic tissue, so that the space curvature parameter developed by the invention has the potential of becoming a cancer auxiliary diagnosis tool.
Claims (4)
1. An image analysis method for extracting space curvature characteristics of a fibrous structure based on space orientation change in a neighborhood is characterized by comprising the following steps of:
(1) calculating the pixel-level spatial orientation of the fibrous structure image to obtain the spatial orientation of all pixels in the image;
(2) carrying out thresholding treatment on the fiber structure image to obtain a binarization matrix;
(3) generating a window at all effective pixels, and calculating the space curvature of all pixels in the image according to the space orientation information in the window to obtain the normalized space curvature value of the central pixel in the window;
(4) and performing pseudo color coding on the image according to the obtained normalized space curvature value to express the numerical value of the space curvature.
2. The image analysis method for extracting curvature features of a fibrous structure based on the change of the orientation of the space in the neighborhood according to claim 1, wherein in the step (1), the fibrous structure image is a two-dimensional/three-dimensional image;
for a two-dimensional image, generating a two-dimensional window containing pixels in the neighborhood by taking all pixels in the image as centers, generating weighting vectors along different directions in the window, finally summing the weighting vectors in all directions, and endowing the spatial orientation value of the obtained sum vector with a center pixel;
for a three-dimensional image, a three-dimensional window containing pixels in a neighborhood is generated by taking all voxels in the image as a center, three two-dimensional images are obtained by projection along the directions of x, y and z, and then the spatial orientation of the three projection images is respectively calculated according to the calculation method of the two-dimensional images, so that the three-dimensional orientation of the central voxel is obtained.
3. The image analysis method for extracting curvature features of a fiber-like structure space based on the orientation change of the space in the neighborhood according to claim 2, wherein the calculation method of the weighting vectors in different directions is as follows: the vectors in different directions are weighted according to the intensity variation of the image signal within the window and the distance between the non-central pixel and the central pixel.
4. The image analysis method for extracting curvature features of a fibrous structure based on spatial orientation variation in a neighborhood according to claim 1, wherein in the step (3), the specific method for calculating the spatial curvatures of all pixels in the image according to the spatial orientation information in the window is as follows: first, the spatial orientation values θ of all the non-central effective pixels in the window are calculated respectivelynAnd central pixel spatial orientation thetacDifference value delta ofnCalculating N times in total, wherein N is the number of non-central pixels in the window; when the difference value deltanWhen the value is negative or more than 90 degrees, correction is needed when delta isnWhen the value is negative, the absolute value of delta is requirednWhen the angle is larger than 90 degrees, the angle needs to be compensated; finally, all corrected differences δ are addednThe sum is divided by N and then divided by 90 to obtain the final normalized spatial curvature value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210065555.9A CN114596256A (en) | 2022-01-20 | 2022-01-20 | Image analysis method for extracting space curvature characteristics of fibrous structure based on space orientation change in neighborhood |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210065555.9A CN114596256A (en) | 2022-01-20 | 2022-01-20 | Image analysis method for extracting space curvature characteristics of fibrous structure based on space orientation change in neighborhood |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114596256A true CN114596256A (en) | 2022-06-07 |
Family
ID=81806969
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210065555.9A Pending CN114596256A (en) | 2022-01-20 | 2022-01-20 | Image analysis method for extracting space curvature characteristics of fibrous structure based on space orientation change in neighborhood |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114596256A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117530659A (en) * | 2023-11-06 | 2024-02-09 | 浙江大学 | Early cervical cancer transformation auxiliary diagnostic tool based on optical coherence tomography |
-
2022
- 2022-01-20 CN CN202210065555.9A patent/CN114596256A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117530659A (en) * | 2023-11-06 | 2024-02-09 | 浙江大学 | Early cervical cancer transformation auxiliary diagnostic tool based on optical coherence tomography |
| CN117530659B (en) * | 2023-11-06 | 2024-04-09 | 浙江大学 | Early cervical cancer transformation auxiliary diagnostic tool based on optical coherence tomography |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110934606B (en) | Cerebral apoplexy early-stage flat-scan CT image evaluation system and method and readable storage medium | |
| Li et al. | Biopsy-free in vivo virtual histology of skin using deep learning | |
| Tian et al. | Deep learning in biomedical optics | |
| Zaimi et al. | AxonSeg: open source software for axon and myelin segmentation and morphometric analysis | |
| Jaworek-Korjakowska | Computer‐aided diagnosis of micro‐malignant melanoma lesions applying support vector machines | |
| CA2752370C (en) | Segmentation of structures for state determination | |
| US8913061B2 (en) | Automatic tracing algorithm for quantitative analysis of continuous structures | |
| CN115359045A (en) | Image convolution neural network disease prediction system based on multi-mode magnetic resonance imaging | |
| Rebsamen et al. | Brain morphometry estimation: From hours to seconds using deep learning | |
| CN114596256A (en) | Image analysis method for extracting space curvature characteristics of fibrous structure based on space orientation change in neighborhood | |
| CN115222992A (en) | Preoperative intraoperative image-based breast cancer lymph node benign and malignant classification system and method | |
| Ong et al. | Detection of subtle white matter lesions in MRI through texture feature extraction and boundary delineation using an embedded clustering strategy | |
| CN115602311A (en) | Pancreatic cancer auxiliary inspection tool based on collagen fiber multivariate parameter analysis | |
| CN116228731A (en) | Multi-contrast learning coronary artery high-risk plaque detection method, system and terminal | |
| CN112184613A (en) | Magnetic resonance lung image recognition device and method based on artificial intelligence | |
| Pallawi et al. | Study of Alzheimer’s disease brain impairment and methods for its early diagnosis: a comprehensive survey | |
| Mihelic et al. | Segmentation-less, automated vascular vectorization robustly extracts neurovascular network statistics from in vivo two-photon images | |
| CN115063351A (en) | Deep learning-based fetal MRI brain tissue segmentation method and device | |
| Mosaliganti et al. | An imaging workflow for characterizing phenotypical change in large histological mouse model datasets | |
| Mir et al. | Smart Approaches for Diagnosis of Brain Disorders Using Artificial Intelligence | |
| CN117576045A (en) | Deep learning model construction and rectal cancer tumor heterogeneity characteristic system quantitative evaluation method based on rectal cavity contrast agent MRI image | |
| CN116523857B (en) | Hearing state prediction device and method based on diffusion tensor image | |
| CN117809839B (en) | Correlation analysis method for predicting hypertensive retinopathy and related factors | |
| Han | Rapid three-dimensional tracing of the mouse brain neurovasculature with local maximum intensity projection and moving windows | |
| Dey | Review on Novel Approach to observation of Brain Image Anomaly |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |