CN113077415B - Tumor microvascular invasion detection device based on image analysis - Google Patents
Tumor microvascular invasion detection device based on image analysis Download PDFInfo
- Publication number
- CN113077415B CN113077415B CN202110238946.1A CN202110238946A CN113077415B CN 113077415 B CN113077415 B CN 113077415B CN 202110238946 A CN202110238946 A CN 202110238946A CN 113077415 B CN113077415 B CN 113077415B
- Authority
- CN
- China
- Prior art keywords
- image
- rectangular frame
- invasion
- tumor
- blood vessel
- 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.)
- Active
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
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- 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
- G06T7/66—Analysis of geometric attributes of image moments or centre of gravity
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20112—Image segmentation details
- G06T2207/20132—Image cropping
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30096—Tumor; Lesion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Quality & Reliability (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Geometry (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
- Investigating Or Analysing Biological Materials (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a tumor microvascular invasion detection device based on image analysis, which comprises: the input module is used for segmenting the pathological image into a plurality of regional images with preset sizes and inputting the regional images into the blood vessel invasion model; the marking module is used for carrying out image detection on the regional image by using the blood vessel invasion model to obtain a tumor micro-blood vessel invasion structure and marking each tumor micro-blood vessel invasion structure by adopting a rectangular frame; the duplication removing module is used for removing duplication of the two overlapped rectangular frames according to the coverage area and the confidence coefficient of the rectangular frames to obtain the duplicated rectangular frames; and the classification module is used for cutting each rectangular frame into a square image with the gravity center of each rectangular frame as the gravity center and the length of each rectangular frame as the side length, classifying the square image by adopting a classification model, and judging whether the square image is invaded by tumor capillaries. The embodiment of the invention can effectively improve the accuracy and reliability of the tumor microvascular invasion detection.
Description
Technical Field
The invention relates to the technical field of computer medical image information processing, in particular to a tumor microvascular invasion detection device based on image analysis.
Background
Microvascular invasion (MVI) is an important histopathological feature of patients with liver tumors and an independent risk factor of poor prognosis, and can increase postoperative recurrence risk and reduce long-term survival rate, and the recurrence risk of patients with liver tumors accompanied by microvascular invasion (MVI) can be greatly increased. The method for pre-detecting the microvascular invasion has important significance for selecting a clinical treatment scheme and improving the prognosis of a patient. The existing tumor microvascular invasion detection device usually uses a pathological image original sheet to train a deep learning model, and avoids memory overflow by reducing the resolution of an image, but the existing tumor microvascular invasion detection device has a fuzzy structure of microvascular invasion in the pathological image original sheet, so that the detection effect is inaccurate.
Disclosure of Invention
The invention provides a tumor microvascular invasion detection device based on image analysis, and aims to solve the technical problem that an existing tumor microvascular invasion detection device is inaccurate in detection effect due to fuzzy structure of microvascular invasion in a pathological image original sheet.
The embodiment of the invention provides a tumor microvascular invasion detection device based on image analysis, which comprises:
the input module is used for segmenting the pathological image into a plurality of regional images with preset sizes and inputting the regional images into the blood vessel invasion model;
the marking module is used for carrying out image detection on the region image by using the blood vessel invasion model to obtain a tumor micro-blood vessel invasion structure and marking each tumor micro-blood vessel invasion structure by using a rectangular frame, wherein the tumor micro-blood vessel invasion structure marked by each rectangular frame corresponds to a confidence coefficient;
the de-duplication module is used for performing target de-duplication processing on the two overlapped rectangular frames according to the coverage area and the confidence coefficient of the rectangular frames to obtain the de-duplicated rectangular frames;
and the classification module is used for cutting each rectangular frame into a square image with the gravity center of each rectangular frame as the gravity center and the length of each rectangular frame as the side length, classifying the square image by adopting a classification model, and judging whether the square image is invaded by tumor capillaries.
Further, the inputting a plurality of the region images into the blood vessel invasion model specifically includes:
and reducing the resolution of the area image, and inputting the area image with the reduced resolution into a blood vessel invasion model.
Further, the marking module is specifically configured to:
and carrying out image detection on the regional image by adopting a YOLO-v4 network structure through the blood vessel invasion model to obtain a tumor micro-blood vessel invasion structure, and marking each tumor micro-blood vessel invasion structure by adopting a rectangular frame.
Further, the image detection is performed on the region image through the blood vessel invasion model by using a YOLO-v4 network structure to obtain a tumor micro-blood vessel invasion structure, and each tumor micro-blood vessel invasion structure is marked by using a rectangular frame, which specifically comprises:
and carrying out target detection on the tumor microvascular invasion structure on the regional image by adopting a YOLO-v4 network structure under the magnification of 5 times, marking each tumor microvascular invasion structure by adopting a rectangular frame, and recording the coordinate point of the rectangular frame.
Further, the deduplication module is specifically configured to:
and detecting whether two rectangular frames of an overlapping area exist or not, if so, calculating the overlapping area of the two rectangular frames and the total coverage area of the two rectangular frames, and selecting the rectangular frame with the highest confidence level from the two rectangular frames as a final rectangular frame when the ratio of the overlapping area to the total coverage area exceeds a preset threshold.
Further, the cutting of each rectangular frame into a square image with the center of gravity of each rectangular frame as the center of gravity and the length of each rectangular frame as the side length specifically includes:
extracting the rectangular frame coordinate points, carrying out linear transformation on the rectangular frame coordinate points, mapping to two-dimensional plane coordinates under a high-power lens, determining the length of each rectangular frame according to the two-dimensional plane coordinates, and cutting each rectangular frame into a square image with the gravity center of each rectangular frame as the gravity center and the length of each rectangular frame as the side length.
According to the embodiment of the invention, the case image original sheet is divided into the area images with the preset size through the input module, so that the picture pixels in the input microvascular invasion detection model are reduced, and the problem that the accurate detection cannot be realized due to the overflow of the display memory of the display card when the input picture is too large is avoided; according to the embodiment of the invention, all tumor microvascular invasion structures in the pathological image are detected and obtained through the marking module, and the tumor microvascular invasion structures obtained through the preliminary detection are further screened and classified through the de-duplication module and the classification module, so that the accuracy of tumor microvascular invasion detection can be further improved.
Drawings
Fig. 1 is a schematic structural diagram of a tumor microvascular invasion detection apparatus based on image analysis according to an embodiment of the present invention;
fig. 2 is a schematic flow chart of a tumor microvascular invasion detection method based on image analysis according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In the description of the present application, it is to be understood that the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or as implying that the number of indicated technical features is indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present application, "a plurality" means two or more unless otherwise specified.
In the description of the present application, it should be noted that, unless otherwise explicitly stated or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be, for example, a fixed connection, a detachable connection, or an integral connection; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meaning of the above terms in the present application can be understood in a specific case by those of ordinary skill in the art.
Referring to fig. 1, an embodiment of the present invention provides a tumor microvascular invasion detection apparatus based on image analysis, including:
the input module 1 is used for segmenting the pathological image into a plurality of regional images with preset sizes and inputting the regional images into the blood vessel invasion model;
the marking module 2 is used for carrying out image detection on the region image by using the blood vessel invasion model to obtain a tumor micro-blood vessel invasion structure, and marking each tumor micro-blood vessel invasion structure by using a rectangular frame, wherein the tumor micro-blood vessel invasion structure marked by each rectangular frame corresponds to a confidence coefficient;
the duplication removing module 3 is used for performing target duplication removing processing on the two overlapped rectangular frames according to the coverage area and the confidence coefficient of the rectangular frames to obtain the duplicated rectangular frames;
and the classification module 4 is used for cutting each rectangular frame into a square image with the gravity center of each rectangular frame as the gravity center and the length of each rectangular frame as the side length, classifying the square image by adopting a classification model, and judging whether the square image is invaded by tumor capillaries.
According to the embodiment of the invention, the original case image is divided into the area images with the preset size through the input module 1, so that the input of the image pixels in the microvascular invasion detection model is reduced, and the problem that the accurate detection cannot be realized due to the overflow of the display memory of a display card when the input image is too large is avoided; according to the embodiment of the invention, all tumor microvascular invasion structures in the pathological image are detected and obtained through the marking module 2, and the tumor microvascular invasion structures obtained through the preliminary detection are further screened and classified through the de-duplication module 3 and the classification module 4, so that the accuracy of tumor microvascular invasion detection can be further improved.
As a specific implementation manner of the embodiment of the present invention, the method inputs a plurality of region images into a blood vessel invasion model, specifically:
and reducing the resolution of the area image, and inputting the area image with the reduced resolution into the blood vessel invasion model.
Optionally, the vessel invasion model is a hyposcopic tumor microvascular invasion target detection model (High sensitivity model). In a specific implementation manner, the pathological image is divided into a plurality of area images of 12800 × 12800 pixels, and the 12800 × 12800 pixels are converted into the area images of 1280 × 1280 pixels in a manner of reducing resolution, so that the phenomenon that a display card overflows due to too large pixels of the image, and therefore tumor vessel invasion detection cannot be accurately performed is avoided.
As a specific implementation manner of the embodiment of the present invention, the marking module 2 is specifically configured to:
and carrying out image detection on the regional image by adopting a YOLO-v4 network structure through a blood vessel invasion model to obtain a tumor micro-blood vessel invasion structure, and marking each tumor micro-blood vessel invasion structure by adopting a rectangular frame.
According to the embodiment of the invention, the regional image is subjected to image detection through the blood vessel invasion model to preliminarily extract the micro blood vessel invasion structure, so that a preliminary tumor blood vessel invasion structure is obtained. Furthermore, in the embodiment of the present invention, a rectangular frame marking manner is adopted, each tumor vessel invasion structure is respectively marked by different rectangular frames, and the YOLO-v4 network structure automatically calculates the confidence corresponding to each tumor vessel invasion structure during detection, where the confidence is a numerical value in an interval [0,1], and a value closer to 1 indicates that the detection reliability of the tumor vessel invasion structure is higher. According to the embodiment of the invention, all the microvascular invasion structures in the pathological image can be extracted through the blood vessel invasion model, so that comprehensive data support is provided for subsequent further detection, and therefore, the tumor microvascular invasion detection is more comprehensive and more accurate.
As a specific implementation manner of the embodiment of the present invention, a blood vessel invasion model is used to perform image detection on a region image by using a YOLO-v4 network structure to obtain a tumor microvascular invasion structure, and a rectangular frame is used to mark each tumor microvascular invasion structure, which specifically includes:
and carrying out target detection on the tumor microvascular invasion structure on the regional image by adopting a YOLO-v4 network structure under 5-time magnification, marking each tumor microvascular invasion structure by adopting a rectangular frame, and recording a rectangular frame coordinate point.
As a specific implementation manner of the embodiment of the present invention, the duplicate removal module 3 is specifically configured to:
and detecting whether two rectangular frames of an overlapping area exist or not, if so, calculating the overlapping area of the two rectangular frames and the total coverage area of the two rectangular frames, and selecting the rectangular frame with the highest confidence level in the two rectangular frames as a final rectangular frame when the ratio of the overlapping area to the total coverage area exceeds a preset threshold value.
In the embodiment of the present invention, when detecting that there are two rectangular frames in an overlapping area, the deduplication module 3 calculates a ratio of an overlapping area of the two rectangular frames to a total coverage area of the two rectangular frames, and compares the ratio with a preset threshold ratio, thereby determining whether the two rectangular frames are overlapping rectangular frames. In a specific implementation mode, the preset threshold ratio is 0.3, when the ratio of the overlapping area of the two rectangles to the total coverage area of the rectangular frames is equal to or greater than 0.3, the tumor microvascular invasion structures in the two rectangular frames are judged to be the same structure, and the rectangular frame with the highest confidence coefficient corresponding to the tumor microvascular invasion structures in the two rectangular frames is selected as the final rectangular frame, so that the influence of the overlapping area on the tumor microvascular invasion detection can be effectively reduced, the detection efficiency can be effectively improved, and the detection accuracy and reliability can be effectively improved.
As a specific implementation manner of the embodiment of the present invention, each rectangular frame is cut into a square image with the length of each rectangular frame as the side length, specifically:
extracting a rectangular frame coordinate point, carrying out linear transformation on the rectangular frame coordinate point and mapping the rectangular frame coordinate point to a two-dimensional plane coordinate under a high-power lens, determining the length of each rectangular frame according to the two-dimensional plane coordinate, and cutting each rectangular frame into a square image with the gravity center of each rectangular frame as the gravity center and the length of each rectangular frame as the side length.
According to the embodiment of the invention, each rectangular frame after the duplication elimination is cut into the regular square images, and the square images are input into the classification model for further tumor microvascular invasion detection, so that false positive microvascular invasion structures in the rectangular frames can be effectively screened, and the accuracy and reliability of microvascular invasion structure detection can be further improved. And each rectangular frame is cut to obtain a corresponding square image.
Optionally, before the square image is input into the classification model for classification detection, the resolution of the square image is increased, so that the definition of the input image can be further improved, the micro-vessel invasion structure of the square image in China is clear and visible, and the accuracy of tumor micro-vessel invasion detection can be effectively improved. Optionally, the classification model is the inclusion-v 3 model.
Referring to fig. 2, another embodiment of the present invention provides a tumor microvascular invasion detection method based on image analysis.
The embodiment of the invention has the following beneficial effects:
according to the embodiment of the invention, the case image original sheet is divided into the area images with the preset size through the input module 1, so that the picture pixels in the microvessel invasion detection model are reduced, and the problem that the accurate detection cannot be realized due to the overflow of the display memory of the display card when the input picture is too large is avoided; according to the embodiment of the invention, all tumor microvascular invasion structures in the pathological image are detected and obtained through the marking module 2, and the primarily detected tumor microvascular invasion structures are further screened and classified through the duplication removing module 3 and the classification module 4, so that the accuracy of tumor microvascular invasion detection can be further improved.
The foregoing is a preferred embodiment of the present invention, and it should be noted that it would be apparent to those skilled in the art that various modifications and enhancements can be made without departing from the principles of the invention, and such modifications and enhancements are also considered to be within the scope of the invention.
Claims (4)
1. A tumor microvascular invasion detection apparatus based on image analysis, comprising:
the input module is used for segmenting the pathological image into a plurality of regional images with preset sizes and inputting the regional images into the blood vessel invasion model;
the marking module is used for carrying out image detection on the region image by using the blood vessel invasion model to obtain a tumor micro-blood vessel invasion structure and marking each tumor micro-blood vessel invasion structure by using a rectangular frame, wherein the tumor micro-blood vessel invasion structure marked by each rectangular frame corresponds to a confidence coefficient;
the duplication removal module is used for carrying out target duplication removal processing on the two overlapped rectangular frames according to the coverage area and the confidence coefficient of the rectangular frames to obtain the duplicated rectangular frames;
the classification module is used for cutting each rectangular frame into a square image with the gravity center of each rectangular frame as the gravity center and the length of each rectangular frame as the side length, classifying the square image by adopting a classification model, and judging whether the square image is invaded by tumor microvasculature;
inputting a plurality of the region images into a blood vessel invasion model, specifically:
reducing the resolution of the area image, and inputting the area image with the reduced resolution into a blood vessel invasion model;
the marking module is specifically configured to:
and performing image detection on the region image by adopting a YOLO-v4 network structure through the blood vessel invasion model to obtain a tumor micro-blood vessel invasion structure, and marking each tumor micro-blood vessel invasion structure by adopting a rectangular frame.
2. The apparatus for detecting invasion of tumor microvasculature based on image analysis according to claim 1, wherein said image detection of said region image by using a YOLO-v4 network structure through said blood vessel invasion model obtains invasion of tumor microvasculature structures, and each of said invasion of tumor microvasculature structures is marked by using a rectangular frame, specifically:
and carrying out target detection on the tumor microvascular invasion structure on the regional image by adopting a YOLO-v4 network structure under the magnification of 5 times, marking each tumor microvascular invasion structure by adopting a rectangular frame, and recording the coordinate point of the rectangular frame.
3. The apparatus for detecting invasion of tumor microvasculature based on image analysis according to claim 1, wherein said deduplication module is specifically configured to:
and detecting whether two rectangular frames in an overlapping area exist or not, if so, calculating the overlapping area of the two rectangular frames and the total coverage area of the two rectangular frames, and selecting the rectangular frame with the highest confidence level in the two rectangular frames as a final rectangular frame when the ratio of the overlapping area to the total coverage area exceeds a preset threshold value.
4. The apparatus for detecting invasion of tumor microvasculature based on image analysis according to claim 2, wherein said each rectangular frame is cropped to a square image with the center of gravity of each rectangular frame as the center of gravity and the length of each rectangular frame as the side length, specifically:
and extracting the coordinate points of the rectangular frames, performing linear transformation on the coordinate points of the rectangular frames, mapping the coordinate points of the rectangular frames to two-dimensional plane coordinates under a high-power mirror, determining the length of the rectangular frames according to the two-dimensional plane coordinates, and cutting each rectangular frame into a square image with the gravity center of each rectangular frame as the gravity center and the length of each rectangular frame as the side length.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110238946.1A CN113077415B (en) | 2021-03-04 | 2021-03-04 | Tumor microvascular invasion detection device based on image analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110238946.1A CN113077415B (en) | 2021-03-04 | 2021-03-04 | Tumor microvascular invasion detection device based on image analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113077415A CN113077415A (en) | 2021-07-06 |
| CN113077415B true CN113077415B (en) | 2023-02-24 |
Family
ID=76609873
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110238946.1A Active CN113077415B (en) | 2021-03-04 | 2021-03-04 | Tumor microvascular invasion detection device based on image analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113077415B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110796656A (en) * | 2019-11-01 | 2020-02-14 | 上海联影智能医疗科技有限公司 | Image detection method, image detection device, computer equipment and storage medium |
| CN112215179A (en) * | 2020-10-19 | 2021-01-12 | 平安国际智慧城市科技股份有限公司 | In-vehicle face recognition method, device, apparatus and storage medium |
| CN112288672A (en) * | 2019-07-10 | 2021-01-29 | 富泰华工业（深圳）有限公司 | Tumor detection method and tumor detection device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6993175B2 (en) * | 2001-09-07 | 2006-01-31 | Quest Diagnostics Investments Incorporated | Methods for measuring microvascular density in tumors |
-
2021
- 2021-03-04 CN CN202110238946.1A patent/CN113077415B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112288672A (en) * | 2019-07-10 | 2021-01-29 | 富泰华工业（深圳）有限公司 | Tumor detection method and tumor detection device |
| CN110796656A (en) * | 2019-11-01 | 2020-02-14 | 上海联影智能医疗科技有限公司 | Image detection method, image detection device, computer equipment and storage medium |
| CN112215179A (en) * | 2020-10-19 | 2021-01-12 | 平安国际智慧城市科技股份有限公司 | In-vehicle face recognition method, device, apparatus and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 医学肿瘤影像分类算法研究及其在肝癌上的应用;管月;《中国优秀博硕士学位论文全文数据库(博士) 信息科技辑》;20190215(第2期);第I138-110页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113077415A (en) | 2021-07-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102294193B1 (en) | Apparatus and method for supporting computer aided diagonosis based on probe speed | |
| CN102138827B (en) | Image display device | |
| CN111340827A (en) | Lung CT image data processing and analyzing method and system | |
| CN110974306B (en) | System for discernment and location pancreas neuroendocrine tumour under ultrasonic endoscope | |
| CN110136153B (en) | Image processing method, device and storage medium | |
| JP7450848B2 (en) | Transparency detection method based on machine vision | |
| WO2019136922A1 (en) | Pulmonary nodule detection method, application server, and computer-readable storage medium | |
| KR20230157277A (en) | Apparatus, method and computer program for anayzing brain medical image using classification and segmentation | |
| JP2011165170A (en) | Object detection device and program | |
| CN110580697B (en) | Video image processing method and system for measuring thickness of fetal nape transparency from ultrasonic video image | |
| CN117333489B (en) | Film damage detection device and detection system | |
| CN113077415B (en) | Tumor microvascular invasion detection device based on image analysis | |
| CN110051384B (en) | Method and system for detecting position of transparency of neck and back of fetus by combining medical statistical information | |
| CN113469942B (en) | CT image lesion detection method | |
| CN113191351B (en) | Reading identification method and device of digital electric meter and model training method and device | |
| Kiyuna et al. | Automatic classification of hepatocellular carcinoma images based on nuclear and structural features | |
| CN108629786B (en) | Image edge detection method and device | |
| KR20090029430A (en) | Method and apparatus of digitizing electrocardiogram papers to interwork with electronic medical record system | |
| CN107403133A (en) | Determine equipment and determination method | |
| KR102460899B1 (en) | Method and System for People Count based on Deep Learning | |
| CN113469086A (en) | Method, device, equipment and medium for dividing areas in building plan | |
| CN108734746B (en) | Hand-foot-and-mouth disease detection device based on machine vision | |
| JP2010233737A (en) | Body type determination method, body type determination system, and monitoring system using the same | |
| CN111612755A (en) | Lung focus analysis method, device, electronic equipment and storage medium | |
| CN115148365B (en) | Methods and systems for predicting prognosis of CNS germ cell tumors |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |