WO2014082925A1 - A system and a method for processing post segmentation data of a medical image - Google Patents
A system and a method for processing post segmentation data of a medical image Download PDFInfo
- Publication number
- WO2014082925A1 WO2014082925A1 PCT/EP2013/074450 EP2013074450W WO2014082925A1 WO 2014082925 A1 WO2014082925 A1 WO 2014082925A1 EP 2013074450 W EP2013074450 W EP 2013074450W WO 2014082925 A1 WO2014082925 A1 WO 2014082925A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- vessels
- segmented
- subjects
- segmented subjects
- processor
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- 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
Definitions
- the invention relates to processing of post segmentation data of a medical image.
- Vessel segmentation in XRA images is a rudimentary step for high level computer vision tasks like diagnosis, computing risks of heart blocks, tracking pathologies, information fusion, surgery planning and intervention.
- the object of the invention is to reduce false positives pre- sent in the segmented images.
- the object of the invention is achieved by the system for processing post segmentation data comprising segmented subjects of a medical image of the claim 1 and the method of claim 9.
- the system includes a processor which receives the post segmentation data of the medical image and processes the post segmentation data on a basis of a physical feature of segmented subjects, and categorizes the segmented subjects into vessels and non-vessels.
- the physical features are region based features and the processor categorizes segmented subjects into vessels, if the segmented subjects are high likelihood regions.
- the segmented subjects are smallest units relevant for classifying the segmented subjects into vessels and non vessels.
- the physical features of the segmented subjects are defined by region based characteristics of the segmented subjects in the medical image which can create distinction between two segmented subjects. This provides for region based classification approach for classifying vessels and non-vessels.
- the physical features comprises geometry of the segmented subjects. Geometry based classification provides an easy way to identify vessels from non-vessels. According to yet another embodiment of the system, wherein the physical features comprises texture profiles of the segmented subjects in the medical image. Texture profile based classification provides for another way to classify vessels from non-vessels.
- the processor processes pixels of the segmented subjects on the basis of the physical features of the segmented subjects and determines a normalized central moment of the physical fea- tures, and categorizes the segmented subjects into vessels and non-vessels on a basis of the normalized central moment of the physical features. This maximize the inter class seperability, i.e. between vessels and non-vessels, and minimize the intra class seperability across the vasculature.
- the processor is captures a gradient orientation of the segmented subjects and categorizes the segmented subjects into vessels and non-vessels on a basis of the gradient orientation of the segmented subjects. This provides for invariance to geometric and photometric transformations, except for object orientation .
- the processor is processes the post segmentation data on a basis of symmetric shape of segmented subjects, and categorizes the segmented subjects into vessels and non-vessels. Symmetry of shape of segmented subjects provides for an easy identity of the segmented subjects with respect to each other and hence helps for easy classification.
- the sys- tern includes an image reconstructor and the processor transfers the segmented subjects categorized as vessels to an image reconstructor and the image reconstructor reconstructs a reconstructed medical image from the segmented subjects categorized as vessels.
- Such reconstruction of medical image pro- vides clarity for analysis of medical image.
- FIG 1 illustrates schematic diagram of a system for processing post segmentation data of a medical image.
- the embodiments herein and the various features and advantageous details thereof are explained more fully with reference to the non- limiting embodiments that are illustrated in the accompanying drawings and detailed in the following description. Descriptions of well-known components and processing techniques are omitted so as to not unnecessarily obscure the embodiments herein.
- the examples used herein are intended merely to facilitate an understanding of ways in which the embodiments herein may be practiced and to further enable those of skill in the art to practice the embodiments herein. Accordingly, the examples should not be construed as limiting the scope of the embodiments herein.
- processors are generally processors which are logic circuitry that responds to and processes the basic instructions for performing a function. They may be a central processing unit of a personal computer adapted to perform the function or microprocessors which are multipurpose, programmable devices that accepts digital data as in- put, processes it according to instructions stored in its memory, and provides results as output or any other computing device adapted to perform functions of the processor and/or image reconstructor according to current invention.
- processors and “Image reconstructor” are generally processors which are logic circuitry that responds to and processes the basic instructions for performing a function. They may be a central processing unit of a personal computer adapted to perform the function or microprocessors which are multipurpose, programmable devices that accepts digital data as in- put, processes it according to instructions stored in its memory, and provides results as output or any other computing device adapted to perform functions of the processor and/or image reconstructor according to current invention.
- Medical image is a visual representation of an anatomy of a human body which may include vessels and non-vessels.
- the vessels are the part of the circulatory system that transports blood throughout the body. Any other features of the human anatomy other than vessels are considered as non-vessels.
- Post segmentation data and "Segmented subjects”: For purposes of analyzing a medical image, the medical image is segmented into segmented subjects who are defined by various objects and boundaries (lines, curves, etc.) in images. More precisely, image segmentation is the process of assigning a label to every pixel in the medical image such that pixels with the same label share certain visual characteristics with respect to one of the segmented subjects. Any data obtained after segmentation of the medical image is considered as post segmentation data.
- FIG 1 shows a system 1 for processing post segmentation data 2 of a medical image.
- the post segmentation data 2 includes segmented subjects 3 which are smallest units relevant for classifying the segmented subjects 3 into vessels 5 and non vessels 6.
- the system 1 includes a processor 4 which receives the post segmentation data 2 of the medical image and processes the post segmentation data 2 on a basis of a physical feature of segmented subjects and categorizes the segmented subjects 3 into vessels 5 and non-vessels 6.
- the physical features of the segmented subjects 3 are defined by visual characteristics of the segmented subjects 3 in the medical image which can create distinction between segmented subjects 3.
- the physical features are region based and the processor 4 processes the segmented sub- jects 3 on the basis the physical features and identify the segmented subjects 3 which are high likelihood regions and categorizes the segmented subjects 3 into vessels, if the segmented subject 3 is high likelihood region.
- the physical features can be defined by characteristics of the segmented subjects 3 which can create distinction between segmented subjects 3 considering local pixel neighborhood like adaptive thresholding, Snakes, level set, watersheds, etc.
- the physical features may include geometry of the segmented subjects 3, or texture profiles like intensity pro- files profiles of the segmented subjects 3 in the medical image, or combination thereof. Use of such physical features in vessel extraction, helps to:
- the processor 4 processes pixels of the segmented subjects 3 on the basis of the physical features of the segmented subjects 3 and determines a normalized central moment 9 of the physical features, and uses the normalized central moment 9 of the physical features for categorizing the segmented subjects 3 into vessels 5 and non-vessels 6.
- the processor 4 need not generate the histogram from pixels based on the physical features of the segmented subjects; rather the processor 4 may directly use other re- gion or neighborhood based features like shape, texture profiles for categorizing vessels 5 and non-vessels 6.
- the processor 4 also captures a gradient orientation 10 of the segmented subjects 3 and uses the gradient orientation 10 of the segmented subjects 3 for categorizing the segmented subjects 3 into vessels 5 and non-vessels 6.
- the processor 4 need not capture the gradient orientation 10 of the segmented subjects 3; rather the processor 4 may directly use other region or neighborhood based features like shape, intensity profiles for categorizing vessels 5 and non-vessels 6.
- the processor 4 is also processes the post segmentation data 2 on a basis of symmetric shape of segmented subjects 3, and uses the symmetric shape for categorizing the segmented subjects 3 into vessels 5 and non-vessels 6.
- the processor 4 need not use the symmetric shape of the segmented subjects 3; rather the processor 4 may directly use other region or neighborhood based features like shape, intensity profiles for categorizing vessels 5 and non-vessels 6.
- the system 1 further includes an image reconstructor 7.
- the processor 4 further provides the segmented subjects 3 catego- rized as vessels 5 to the image reconstructor 7 and the image reconstructor 7 reconstructs a reconstructed medical image 8 of the segmented subjects 3 categorized as vessels 5.
- the system 1 need not include the image reconstructor 8, rather the processor 4 just transfers the segmented subjects 3 classified as vessels 5 to an external device which reconstructs the medical image to just include the segmented subjects 3 classified as vessels 5.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Image Analysis (AREA)
Abstract
A system (1) for processing post segmentation data (2) com- prising segmented subjects (3) of a medical image includes a processor (4) which receives the post segmentation data (2) of the medical image, processes the post segmentation data (2) on a basis of a physical feature of segmented subjects, and categorizes the segmented subjects (3) into vessels (5) and non-vessels (6). The segmented subjects (6) are smallest units relevant for classifying the segmented subjects (3) in- to vessels (5) and non vessels (6). The physical features of the segmented subjects (3) are defined by visual characteristics of the segmented subjects (3) in the medical image which can create distinction between segmented subjects (3).
Description
Description
A system and a method for processing post segmentation data of a medical image
The invention relates to processing of post segmentation data of a medical image.
Vessel segmentation in XRA images is a rudimentary step for high level computer vision tasks like diagnosis, computing risks of heart blocks, tracking pathologies, information fusion, surgery planning and intervention.
Many vessel segmentation methodologies have been proposed in the past. These methods can be broadly classified as: pattern recognition, model based, tracking and propagation, neural network, fuzzy, and artificial intelligence based methods. However vessel Segmentation in XRA images through any of the above- said methodologies is a challenging task. The challeng- es involve poor contrast, organ projections, noise or other degradations. This leads to many false positives present in the segmented images using the existing methods.
The object of the invention is to reduce false positives pre- sent in the segmented images.
The object of the invention is achieved by the system for processing post segmentation data comprising segmented subjects of a medical image of the claim 1 and the method of claim 9.
According to an embodiment of the system, the system includes a processor which receives the post segmentation data of the medical image and processes the post segmentation data on a basis of a physical feature of segmented subjects, and categorizes the segmented subjects into vessels and non-vessels. The physical features are region based features and the processor categorizes segmented subjects into vessels, if the
segmented subjects are high likelihood regions. The segmented subjects are smallest units relevant for classifying the segmented subjects into vessels and non vessels. The physical features of the segmented subjects are defined by region based characteristics of the segmented subjects in the medical image which can create distinction between two segmented subjects. This provides for region based classification approach for classifying vessels and non-vessels. According to another embodiment of the system, wherein the physical features comprises geometry of the segmented subjects. Geometry based classification provides an easy way to identify vessels from non-vessels. According to yet another embodiment of the system, wherein the physical features comprises texture profiles of the segmented subjects in the medical image. Texture profile based classification provides for another way to classify vessels from non-vessels.
According to one embodiment of the system, wherein the processor processes pixels of the segmented subjects on the basis of the physical features of the segmented subjects and determines a normalized central moment of the physical fea- tures, and categorizes the segmented subjects into vessels and non-vessels on a basis of the normalized central moment of the physical features. This maximize the inter class seperability, i.e. between vessels and non-vessels, and minimize the intra class seperability across the vasculature.
According to another embodiment of the system, wherein the processor is captures a gradient orientation of the segmented subjects and categorizes the segmented subjects into vessels and non-vessels on a basis of the gradient orientation of the segmented subjects. This provides for invariance to geometric and photometric transformations, except for object orientation .
According to yet another embodiment of the system, wherein the processor is processes the post segmentation data on a basis of symmetric shape of segmented subjects, and categorizes the segmented subjects into vessels and non-vessels. Symmetry of shape of segmented subjects provides for an easy identity of the segmented subjects with respect to each other and hence helps for easy classification.
According to an exemplary embodiment of the system, the sys- tern includes an image reconstructor and the processor transfers the segmented subjects categorized as vessels to an image reconstructor and the image reconstructor reconstructs a reconstructed medical image from the segmented subjects categorized as vessels. Such reconstruction of medical image pro- vides clarity for analysis of medical image.
FIG 1 illustrates schematic diagram of a system for processing post segmentation data of a medical image. The embodiments herein and the various features and advantageous details thereof are explained more fully with reference to the non- limiting embodiments that are illustrated in the accompanying drawings and detailed in the following description. Descriptions of well-known components and processing techniques are omitted so as to not unnecessarily obscure the embodiments herein. The examples used herein are intended merely to facilitate an understanding of ways in which the embodiments herein may be practiced and to further enable those of skill in the art to practice the embodiments herein. Accordingly, the examples should not be construed as limiting the scope of the embodiments herein.
Prior to explaining functioning of the system through various embodiments, some of the terminology used herein will be ex- plained.
"Processor" and "Image reconstructor" are generally processors which are logic circuitry that responds to and processes
the basic instructions for performing a function. They may be a central processing unit of a personal computer adapted to perform the function or microprocessors which are multipurpose, programmable devices that accepts digital data as in- put, processes it according to instructions stored in its memory, and provides results as output or any other computing device adapted to perform functions of the processor and/or image reconstructor according to current invention. However, technical difference between the processor and the image reconstructor are explained through there functionalities while explaining the figures.
"Medical image", "Vessels" and "Non-vessels": Medical image is a visual representation of an anatomy of a human body which may include vessels and non-vessels. The vessels are the part of the circulatory system that transports blood throughout the body. Any other features of the human anatomy other than vessels are considered as non-vessels. "Post segmentation data" and "Segmented subjects": For purposes of analyzing a medical image, the medical image is segmented into segmented subjects who are defined by various objects and boundaries (lines, curves, etc.) in images. More precisely, image segmentation is the process of assigning a label to every pixel in the medical image such that pixels with the same label share certain visual characteristics with respect to one of the segmented subjects. Any data obtained after segmentation of the medical image is considered as post segmentation data.
"Normalized central moment of physical features" is defined as function of eigen values of Hessian matrix of segmented subjects in the medical image. "Gradient orientation of segmented subjects" is defined by local appearance and shape of segmented subjects within the medical image and can be defined by the distribution of intensity gradients or edge directions.
FIG 1 shows a system 1 for processing post segmentation data 2 of a medical image. The post segmentation data 2 includes segmented subjects 3 which are smallest units relevant for classifying the segmented subjects 3 into vessels 5 and non vessels 6. The system 1 includes a processor 4 which receives the post segmentation data 2 of the medical image and processes the post segmentation data 2 on a basis of a physical feature of segmented subjects and categorizes the segmented subjects 3 into vessels 5 and non-vessels 6. The physical features of the segmented subjects 3 are defined by visual characteristics of the segmented subjects 3 in the medical image which can create distinction between segmented subjects 3. In an exemplary embodiment, the physical features are region based and the processor 4 processes the segmented sub- jects 3 on the basis the physical features and identify the segmented subjects 3 which are high likelihood regions and categorizes the segmented subjects 3 into vessels, if the segmented subject 3 is high likelihood region. Alternatively, the physical features can be defined by characteristics of the segmented subjects 3 which can create distinction between segmented subjects 3 considering local pixel neighborhood like adaptive thresholding, Snakes, level set, watersheds, etc. The physical features may include geometry of the segmented subjects 3, or texture profiles like intensity pro- files profiles of the segmented subjects 3 in the medical image, or combination thereof. Use of such physical features in vessel extraction, helps to:
• Maximize the inter class seperability, i.e between vessels 5 and non vessels 6.
· Minimize the intra class seperability, this is across the vasculature.
• Keep description robust to noise, illumination artifacts and other degradations.
• Invariant to small affine changes like scale, rotation and shifts.
• Provide the region with minimum number of elements.
The processor 4 processes pixels of the segmented subjects 3 on the basis of the physical features of the segmented subjects 3 and determines a normalized central moment 9 of the physical features, and uses the normalized central moment 9 of the physical features for categorizing the segmented subjects 3 into vessels 5 and non-vessels 6. In an alternate embodiment, the processor 4 need not generate the histogram from pixels based on the physical features of the segmented subjects; rather the processor 4 may directly use other re- gion or neighborhood based features like shape, texture profiles for categorizing vessels 5 and non-vessels 6.
The processor 4 also captures a gradient orientation 10 of the segmented subjects 3 and uses the gradient orientation 10 of the segmented subjects 3 for categorizing the segmented subjects 3 into vessels 5 and non-vessels 6. In an alternate embodiment, the processor 4 need not capture the gradient orientation 10 of the segmented subjects 3; rather the processor 4 may directly use other region or neighborhood based features like shape, intensity profiles for categorizing vessels 5 and non-vessels 6.
The processor 4 is also processes the post segmentation data 2 on a basis of symmetric shape of segmented subjects 3, and uses the symmetric shape for categorizing the segmented subjects 3 into vessels 5 and non-vessels 6. In an alternate embodiment, the processor 4 need not use the symmetric shape of the segmented subjects 3; rather the processor 4 may directly use other region or neighborhood based features like shape, intensity profiles for categorizing vessels 5 and non-vessels 6.
The system 1 further includes an image reconstructor 7. The processor 4 further provides the segmented subjects 3 catego- rized as vessels 5 to the image reconstructor 7 and the image reconstructor 7 reconstructs a reconstructed medical image 8 of the segmented subjects 3 categorized as vessels 5. In an alternate embodiment, the system 1 need not include the image
reconstructor 8, rather the processor 4 just transfers the segmented subjects 3 classified as vessels 5 to an external device which reconstructs the medical image to just include the segmented subjects 3 classified as vessels 5.
Claims
1. A system (1) for processing post segmentation data (2) comprising segmented subjects (3) of a medical image compris- ing :
- a processor (4) adapted to receive the post segmentation data (2) of the medical image, to process the post segmentation data (2) on a basis of a physical feature of segmented subjects (3) , and to categorize the segmented subjects (3) into vessels (5) and non-vessels (6) ,
Wherein the segmented subjects (6) are smallest units relevant for classifying the segmented subjects (3) into vessels (5) and non vessels (6) , and the physical features of the segmented subjects (3) are defined by visual characteristics of the segmented subjects (3) in the medical image which can create distinction between segmented subjects (3) .
2. The system (1) according to the claim 1, wherein the physical features are region based features and the processor is adapted to process the segmented subjects (3) based on the physical features to identify the segmented subjects (3) which are high likelihood regions and to categorize the segmented subjects (3) into vessels (5) , if the segmented subject (3) is identified to be high likelihood region.
3. The system (1) according to any of the claims 1 or 2 , wherein the physical features comprises geometry of the segmented subjects (3) .
4. The system (1) according to any of the claims 1 to 3 , wherein the physical features comprises texture profiles of the segmented subjects (3) in the medical image.
5. The system (1) according to any of the claims 1 to 4 , wherein the processor (4) adapted to determine a normalized central moment (9) of the physical features, and to categorize the segmented subjects (3) into vessels (5) and non-
vessels (6) on a basis of the normalized central moment (9) of the physical feature.
6. The system (1) according to any of the claims 1 to 5 , wherein the processor (4) is adapted to capture a gradient orientation (10) of the segmented subjects (3) and to categorize the segmented subjects (3) into vessels (5) and non- vessels (6) on a basis of the gradient orientation (10) of the segmented subjects (3) .
7. The system (1) according to any of the claims 1 to 6 , wherein the processor (4) is adapted to process the post segmentation data (2) on a basis of symmetric shape of segmented subjects (3) , and to categorize the segmented subjects (3) into vessels (5) and non-vessels (6) .
8. The system (1) according to any of the claims 1 to 6 , wherein the processor (4) is adapted to provide the segmented subjects (3) categorized as vessels (5) to an image
reconstructor (7), the system (1) comprising:
- the image reconstructor (7) adapted to receive the segmented subjects (3) categorized as vessels (5) and to reconstruct a reconstructed medical image (8) of the segmented subjects
(3) categorized as vessels (5) .
9. A method for processing post segmentation data (2) comprising segmented subjects (3) of a medical image comprising:
- receiving the post segmentation data (2) of the medical image by a processor (4) ,
- processing the post segmentation data (2) on a basis of a physical feature of segmented subjects (3) by the processor
(4) , and
- categorizing the segmented subjects (3) into vessels (5) and non-vessels (6) by the processor (4) ,
Wherein the segmented subjects (6) are smallest units relevant for classifying the segmented subjects (3) into vessels
(5) and non vessels (6) , and the physical features of the segmented subjects (3) are defined by visual characteristics
of the segmented subjects (3) in the medical image which can create distinction between segmented subjects (3) .
10. The method according to claim 9, wherein the physical features are region based features, the method comprising:
- processing the segmented subjects (3) based on the physical features and identifying the segmented subjects (3) which are high likelihood regions by the processor (4), and
- categorizing the segmented subjects (3) into vessels (5) by the processor (4) , if the segmented subject is identified to be high likelihood region.
11. The method according to any of the claims 9 or 10 comprising :
- determining a normalized central moment (9) of the physical features by the processor (4) , and
- categorizing the segmented subjects (3) into vessels (5) and non-vessels (6) by the processor (4) on a basis of the normalized central moment (9) of the physical feature.
12. The method according to any of the claims 9 to 11 comprising :
- capturing a gradient orientation (10) of the segmented subjects (3) by the processor (4) , and
- categorizing the segmented subjects (3) into vessels (5) and non-vessels (6) by the processor (4) on a basis of the gradient orientation (10) of the segmented subjects (3) .
13. The method according to any of the claims 9 to 12 com- prising:
- providing the segmented subjects (3) categorized as vessels (5) to an image reconstructor (7) , and
- reconstructing a reconstructed medical image (8) of the segmented subjects (3) categorized as vessels (5) .
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| IN1361KO2012 | 2012-11-27 | ||
| IN1361/KOL/2012 | 2012-11-27 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2014082925A1 true WO2014082925A1 (en) | 2014-06-05 |
Family
ID=49724544
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/EP2013/074450 WO2014082925A1 (en) | 2012-11-27 | 2013-11-22 | A system and a method for processing post segmentation data of a medical image |
Country Status (1)
| Country | Link |
|---|---|
| WO (1) | WO2014082925A1 (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106097293A (en) * | 2015-12-22 | 2016-11-09 | 中国计量学院 | A kind of image severability evaluation method for image capturing system optimization design |
| CN110537204A (en) * | 2017-06-28 | 2019-12-03 | 渊慧科技有限公司 | Using segmentation and Classification Neural can extensive medical image analysis |
| CN116863146A (en) * | 2023-06-09 | 2023-10-10 | 强联智创（北京）科技有限公司 | Method, apparatus and storage medium for extracting hemangio features |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090279758A1 (en) * | 2008-03-05 | 2009-11-12 | Dikici Engin | System and method for 3d vessel segmentation with minimal cuts |
-
2013
- 2013-11-22 WO PCT/EP2013/074450 patent/WO2014082925A1/en active Application Filing
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090279758A1 (en) * | 2008-03-05 | 2009-11-12 | Dikici Engin | System and method for 3d vessel segmentation with minimal cuts |
Non-Patent Citations (5)
| Title |
|---|
| ABRAMOFF M D ET AL: "Ridge-Based Vessel Segmentation in Color Images of the Retina", IEEE TRANSACTIONS ON MEDICAL IMAGING, IEEE SERVICE CENTER, PISCATAWAY, NJ, US, vol. 23, no. 4, 1 April 2004 (2004-04-01), pages 501 - 509, XP011110228, ISSN: 0278-0062, DOI: 10.1109/TMI.2004.825627 * |
| BUEHLER K ET AL: "Geometric Methods for Vessel Visualization and Quantification - A Survey", INTERNET CITATION, no. 35, pages 1 - 24, XP002482268, Retrieved from the Internet <URL:www.vrvis.at/TR/2002/TR_VRVis_2002_035_Full.pdf>> [retrieved on 20020101] * |
| KIRBAS C ET AL: "A REVIEW OF VESSEL EXTRACTION TECHNIQUES AND ALGORITHMS", INTERNET CITATION, 2003, XP007900870, Retrieved from the Internet <URL:ftp://cmp.felk.cvut.cz/pub/cmp/articles/matas/ps/Curvilinear/blood-vessel_review.pdf> [retrieved on 20060721] * |
| SYOJI KOBASHI ET AL: "<title>Automatic segmentation of blood vessels from MR angiography volume data by using fuzzy logic technique</title>", PROCEEDINGS OF SPIE, vol. 3661, 21 May 1999 (1999-05-21), pages 968 - 976, XP055103595, ISSN: 0277-786X, DOI: 10.1117/12.348489 * |
| WANG Y ET AL: "Vessel extraction in coronary X-ray Angiography", ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY, 2005. IEEE-EMBS 2005. 27T H ANNUAL INTERNATIONAL CONFERENCE OF THE SHANGHAI, CHINA 01-04 SEPT. 2005, PISCATAWAY, NJ, USA,IEEE, 1 September 2005 (2005-09-01), pages 1584 - 1587, XP010908086, ISBN: 978-0-7803-8741-6, DOI: 10.1109/IEMBS.2005.1616739 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106097293A (en) * | 2015-12-22 | 2016-11-09 | 中国计量学院 | A kind of image severability evaluation method for image capturing system optimization design |
| CN106097293B (en) * | 2015-12-22 | 2018-12-21 | 中国计量学院 | A kind of image severability evaluation method for image capturing system optimization design |
| CN110537204A (en) * | 2017-06-28 | 2019-12-03 | 渊慧科技有限公司 | Using segmentation and Classification Neural can extensive medical image analysis |
| US11954902B2 (en) | 2017-06-28 | 2024-04-09 | Google Llc | Generalizable medical image analysis using segmentation and classification neural networks |
| CN116863146A (en) * | 2023-06-09 | 2023-10-10 | 强联智创（北京）科技有限公司 | Method, apparatus and storage medium for extracting hemangio features |
| CN116863146B (en) * | 2023-06-09 | 2024-03-08 | 强联智创（北京）科技有限公司 | Method, apparatus and storage medium for extracting hemangio features |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shankar et al. | Hyperparameter tuning deep learning for diabetic retinopathy fundus image classification | |
| Shen et al. | Domain-invariant interpretable fundus image quality assessment | |
| CN107766786B (en) | Activity test method and activity test computing device | |
| Farnell et al. | Enhancement of blood vessels in digital fundus photographs via the application of multiscale line operators | |
| Zhao et al. | Retinal vessel segmentation: An efficient graph cut approach with retinex and local phase | |
| Al-Fahdawi et al. | A fully automatic nerve segmentation and morphometric parameter quantification system for early diagnosis of diabetic neuropathy in corneal images | |
| Jaiswal et al. | Deep learning-based smart IoT health system for blindness detection using retina images | |
| Candemir et al. | Multi-class regularization parameter learning for graph cut image segmentation | |
| Wan et al. | Retinal image enhancement using cycle-constraint adversarial network | |
| WO2019204700A1 (en) | Neonatal pain identification from neonatal facial expressions | |
| Hussain et al. | DilUnet: A U-net based architecture for blood vessels segmentation | |
| David et al. | Retinal Blood Vessels and Optic Disc Segmentation Using U‐Net | |
| Farooq et al. | Improved automatic localization of optic disc in Retinal Fundus using image enhancement techniques and SVM | |
| CN109544528A (en) | A kind of small pulmonary artery image-recognizing method and device | |
| Ahmedt-Aristizabal et al. | Vision-based mouth motion analysis in epilepsy: A 3d perspective | |
| WO2014082925A1 (en) | A system and a method for processing post segmentation data of a medical image | |
| Kumar et al. | Deep learning-assisted retinopathy of prematurity (ROP) screening | |
| Abdani et al. | Iris segmentation method of pterygium anterior segment photographed image | |
| Tan et al. | Classification of left and right eye retinal images | |
| Jana et al. | A semi-supervised approach for automatic detection and segmentation of optic disc from retinal fundus image | |
| Ali et al. | Segmenting retinal blood vessels with Gabor filter and automatic binarization | |
| Wu et al. | Biomedical video denoising using supervised manifold learning | |
| Honale et al. | A review of methods for blood vessel segmentation in retinal images | |
| Ardizzone et al. | Blood vessels and feature points detection on retinal images | |
| Fraz et al. | Computer vision algorithms applied to retinal vessel segmentation and quantification of vessel caliber |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 13801491 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 13801491 Country of ref document: EP Kind code of ref document: A1 |