CN113450310A - Analysis system and method for narrow-band light imaging cystoscopy image - Google Patents
Analysis system and method for narrow-band light imaging cystoscopy image Download PDFInfo
- Publication number
- CN113450310A CN113450310A CN202110602974.7A CN202110602974A CN113450310A CN 113450310 A CN113450310 A CN 113450310A CN 202110602974 A CN202110602974 A CN 202110602974A CN 113450310 A CN113450310 A CN 113450310A
- Authority
- CN
- China
- Prior art keywords
- image
- module
- nbi
- cystoscopy
- processing
- 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
- 238000002574 cystoscopy Methods 0.000 title claims abstract description 25
- 238000004458 analytical method Methods 0.000 title claims abstract description 21
- 238000003384 imaging method Methods 0.000 title claims abstract description 15
- 238000000034 method Methods 0.000 title claims abstract description 14
- 208000007097 Urinary Bladder Neoplasms Diseases 0.000 claims abstract description 37
- 206010005003 Bladder cancer Diseases 0.000 claims abstract description 35
- 201000005112 urinary bladder cancer Diseases 0.000 claims abstract description 35
- 238000012545 processing Methods 0.000 claims abstract description 28
- 238000013135 deep learning Methods 0.000 claims description 17
- 230000003902 lesion Effects 0.000 claims description 16
- 238000013527 convolutional neural network Methods 0.000 claims description 14
- 238000011176 pooling Methods 0.000 claims description 7
- 238000010606 normalization Methods 0.000 claims description 5
- 238000012360 testing method Methods 0.000 claims description 5
- 238000012549 training Methods 0.000 claims description 5
- 238000003672 processing method Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims description 2
- 238000005457 optimization Methods 0.000 claims description 2
- 238000003745 diagnosis Methods 0.000 abstract description 7
- 238000001000 micrograph Methods 0.000 abstract description 3
- 210000001519 tissue Anatomy 0.000 description 17
- 206010028980 Neoplasm Diseases 0.000 description 14
- 230000001575 pathological effect Effects 0.000 description 9
- 201000011510 cancer Diseases 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 6
- 238000013473 artificial intelligence Methods 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 4
- 230000001537 neural effect Effects 0.000 description 4
- 238000010276 construction Methods 0.000 description 3
- 238000010191 image analysis Methods 0.000 description 3
- 206010044412 transitional cell carcinoma Diseases 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 210000000981 epithelium Anatomy 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000000844 transformation Methods 0.000 description 2
- 208000023747 urothelial carcinoma Diseases 0.000 description 2
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 description 1
- 206010004446 Benign prostatic hyperplasia Diseases 0.000 description 1
- 206010006187 Breast cancer Diseases 0.000 description 1
- 208000026310 Breast neoplasm Diseases 0.000 description 1
- 208000009458 Carcinoma in Situ Diseases 0.000 description 1
- 206010063057 Cystitis noninfective Diseases 0.000 description 1
- 206010058467 Lung neoplasm malignant Diseases 0.000 description 1
- 241000227425 Pieris rapae crucivora Species 0.000 description 1
- 208000037062 Polyps Diseases 0.000 description 1
- 208000004403 Prostatic Hyperplasia Diseases 0.000 description 1
- 238000003556 assay Methods 0.000 description 1
- 238000005513 bias potential Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 206010005038 bladder diverticulum Diseases 0.000 description 1
- 210000004204 blood vessel Anatomy 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000003759 clinical diagnosis Methods 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 238000002059 diagnostic imaging Methods 0.000 description 1
- 230000002526 effect on cardiovascular system Effects 0.000 description 1
- 238000001839 endoscopy Methods 0.000 description 1
- 208000006750 hematuria Diseases 0.000 description 1
- 201000004933 in situ carcinoma Diseases 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 201000005202 lung cancer Diseases 0.000 description 1
- 208000020816 lung neoplasm Diseases 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000010827 pathological analysis Methods 0.000 description 1
- 208000000649 small cell carcinoma Diseases 0.000 description 1
- 206010041823 squamous cell carcinoma Diseases 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 238000001356 surgical procedure Methods 0.000 description 1
- 230000004083 survival effect Effects 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 210000000626 ureter Anatomy 0.000 description 1
- 210000003741 urothelium Anatomy 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30096—Tumor; Lesion
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention provides a system and a method for analyzing a narrow-band light imaging cystoscopy image. The system comprises the following modules: the device comprises a basic image module, an input module, an image processing module, an analysis learning module, an interpretation module and an output module. The system can effectively and accurately analyze the microscopy images of the NBI cystoscope, provides powerful auxiliary diagnosis information for bladder cancer, and has a clinical application prospect.
Description
Technical Field
The invention belongs to the technical field of medical image recognition, and particularly relates to a system and a method for analyzing a narrow-band light imaging cystoscopic image.
Background
The incidence and mortality of bladder cancer are high, according to 2018 data, the incidence of bladder cancer is the tenth of all tumors all the year round, and 82270 new cases of new cases in China are the first in the world. Bladder cancer is three to four times more common in men than in women, but because early symptoms (mainly hematuria) are not found in time, female patients are diagnosed in advanced stages and have higher mortality. Bladder cancer is a major hazard to the elderly population, with an average diagnostic age of 73 years and 90% of patients over 55 years. By the existing scientific technology, the diagnosis, treatment mode and five-year survival rate of bladder cancer have not changed obviously in the last thirty years, and diagnosis and follow-up relapse mostly depend on cystoscopy.
At present, Narrow Band Imaging (NBI) has been widely used in cystoscopy, which is to use a filter to filter white light into Narrow Band blue light, green light and red light to increase the contrast and definition of mucosal epithelium and submucosal blood vessels, and compared with common white light, the Narrow Band Imaging has more accurate diagnosis for early stage tumor and small tumor, and is considered as a powerful supplement to the traditional white light cystoscope.
However, at present, the microscopic examination result is only observed by naked eyes of a doctor, and the observation of the lesion position by human eyes has great inaccuracy. In recent years, artificial intelligence technology represented by deep learning is rapidly developed, and in the field of medical image diagnosis, the characteristics of data can be rapidly and accurately identified by utilizing a multilayer neural network, so that the workload of doctors can be reduced, and the inaccuracy of human eye observation can be greatly reduced. At present, deep learning and the like have been widely studied in the fields of medical imaging such as breast cancer pathological examination, lung cancer detection, cardiovascular imaging and the like.
At present, because the time of the NBI cystoscope technology for clinical use is still short, no prior art report combining the NBI cystoscope technology with artificial intelligence exists, a method capable of accurately analyzing images of NBI cystoscopy is developed, an artificial intelligence system for realizing the method is designed, and the NBI cystoscopy technology has very important significance for clinical diagnosis and treatment of bladder cancer.
Disclosure of Invention
The invention aims to provide a method and a system for analyzing a narrow-band light imaging cystoscopy image, which are used for analyzing an image acquired by NBI (negative bias potential indicator) by combining narrow-band light imaging with an artificial intelligence system so as to solve the problem of poor accuracy caused by manually identifying the image analysis at present.
The invention provides an analysis system for narrow-band imaging cystoscopy images, which comprises the following modules:
a basic image module: a basic NBI cystoscopy image library is formed by bladder cancer precancerous lesion images, bladder cancer tissue images and bladder normal tissue image samples which are acquired by an NBI cystoscope;
an input module: inputting the images acquired under the NBI cystoscope and the corresponding histopathological examination results into a system;
an image processing module: processing the image library of the basic image module to obtain a characteristic value set of a precancerous lesion image, a bladder cancer image and a normal tissue image; processing the image input by the image input module to obtain a test characteristic value;
an analysis learning module: constructing and training a deep learning network based on the feature value set; optimizing a deep learning network algorithm according to a result returned by the output module;
an interpretation module: inputting the test characteristic value into an analysis learning module for interpretation, and comparing the interpretation result with the histopathological examination result, wherein if the results are consistent, the interpretation result is accurate, and if the results are inconsistent, the interpretation result is inaccurate;
an output module: if the interpretation result is not accurate, returning the result to the learning module optimization algorithm; and if the interpretation result is accurate, outputting the interpretation result.
Further, the processing of the image processing module comprises the steps of carrying out normalization processing on the NBI cystoscopy image, delineating an interested region and detecting to obtain a characteristic value.
Further, the deep learning network is a convolutional neural network, and is composed of a convolutional layer, a pooling layer, and a full link layer.
The invention also provides a construction method of the analysis system, which comprises the following steps:
1) collecting bladder precancerous lesion images, bladder cancer tissue images and bladder normal tissue image samples collected by an NBI cystoscope to form a basic NBI cystoscope examination image library;
2) processing the image library of the basic image module to obtain a characteristic value set of a precancerous lesion image, a bladder cancer image and a normal tissue image;
3) constructing and training a deep learning network based on the feature value set;
4) the analysis system is optimized with inaccurate interpretation results.
The invention also provides an analysis method of the narrow-band imaging cystoscopy image, which uses the analysis system to analyze through the following steps:
(1) acquiring an NBI cystoscopy image;
(2) processing the NBI cystoscopy image to obtain a characteristic value;
(3) and inputting the characteristic value into the trained deep learning network for processing and distinguishing.
Further, the image processing method in the step (2) is as follows: and carrying out normalization processing, sketching the region of interest and detecting to obtain a characteristic value.
Further, the deep learning network in the step (3) is a convolutional neural network, and is composed of a convolutional layer, a pooling layer, and a full link layer.
Furthermore, the convolutional neural network is constructed on the basis of a characteristic value set obtained by image processing of a basic NBI cystoscopy image database;
the basic NBI cystoscopy image database consists of bladder precancerous lesion images, bladder cancer tissue images and bladder normal tissue image samples which are acquired by an NBI cystoscope.
The bladder cancer of the invention includes bladder cancer of a significant pathological type and bladder cancer of a non-significant pathological type.
Bladder cancer of the prominent pathological type: depending on the pathological origin, more than about 95% of bladder tumors originate from epithelial tissue, the vast majority of epithelial tumor constituents being the urothelium. Tumors derived therefrom are also known as urothelial cancer, referred to as significant bladder cancer lesions. Urothelial carcinoma of the bladder generally develops well at the side wall of the bladder and near the opening of the ureter in the trigone area of the bladder. Tumors can be single or multiple, and can be papillary, polyp-like or flat plaque-like in gross form.
Bladder cancer of non-prominent pathological type: in addition to urothelial carcinoma of the bladder, there are about 5% other pathological subtypes of bladder cancer of other origins or variations, which include the rare tissue subtypes related to nested variant cancer, lymphoepitheliomatous cancer, plasmacytoid cancer, giant cell variant cancer, plasmacytoid cancer, signet ring cell cancer, diffuse cancer, small cell cancer, sarcomatous cancer, etc., according to 2019 edition of Chinese guidance for urology. Bladder cancer of this tissue origin is rare but not negligible and its gross manifestations in the bladder are diverse, defined as non-significant bladder cancer lesions. In addition, in the staging of bladder tumors, Tis (carcinoma in situ), also known as "squamous carcinoma", whose gross appearance under cystoscopy is easily confused with bladder inflammation, should also belong to a category of bladder cancer that is not a significant pathological category.
Convolutional Neural Networks (CNN): is a kind of feed forward Neural Networks (fed forward Neural Networks) containing convolution calculation and having a deep structure, and is one of the representative algorithms of deep learning (deep learning).
The method can effectively and accurately analyze the microscopy image of the NBI cystoscope, solves the problem of poor accuracy caused by manually identifying the image analysis at present, and provides powerful auxiliary diagnosis information for bladder cancer.
Obviously, many modifications, substitutions, and variations are possible in light of the above teachings of the invention, without departing from the basic technical spirit of the invention, as defined by the following claims.
The present invention will be described in further detail with reference to the following examples. This should not be understood as limiting the scope of the above-described subject matter of the present invention to the following examples. All the technologies realized based on the above contents of the present invention belong to the scope of the present invention.
Drawings
FIG. 1 is a basic structure diagram of the NBI cystoscope image acquisition basic module of the application
FIG. 2 is a schematic diagram of the construction of a convolutional neural network for the pre-stage image processing of the NBI cystoscope image according to the present application
Detailed Description
Example 1 construction of an assay System according to the invention
1. Precancerous lesion images, bladder cancer tissue images and normal tissue images of bladder cancer under NBI (negative contrast imaging) bladder endoscopy in urology surgery of Wash's hospital are collected, and three different image representations are used for constructing a basic image module (figure 1).
2. The narrow optical band imaging machine is connected with a computer through an optical fiber to realize the transmission of images and input the images into the computer.
3. Processing the image library of the basic image module to obtain a characteristic value set of a precancerous lesion image, a bladder cancer image and a normal tissue image; processing the image input by the image input module to obtain a test characteristic value; the image processing method comprises the following steps: and carrying out normalization processing, sketching the region of interest and detecting to obtain a characteristic value.
Normalization treatment: the original image to be processed is converted into a corresponding unique standard form (the standard form has invariant characteristics to affine transformations such as translation, rotation and scaling) through a series of transformations.
Region of interest: in machine vision and image processing, a region to be processed is outlined in a manner of a square frame, a circle, an ellipse, an irregular polygon and the like from a processed image, and the region is called a region of interest (ROI).
4. And (3) constructing a convolutional neural network by depending on the result of the image processing in the step 3: building a convolutional layer to construct a neural architecture, periodically inserting a pooling layer and connecting all units by a full link layer; the method comprises the steps of guiding the cystoscopic performance and pathological diagnosis of a patient into an artificial intelligence system, further collecting and guiding a plurality of bladder precancerous lesion images, bladder cancer tissue images and normal tissue images (such as non-cancerous tissue images of bladder diverticulum changes caused by prostatic hyperplasia) under different NBI cystoscopes, and carrying out learning and training.
Convolutional neural networks are deep learning algorithms consisting of various layers arranged in order, each layer in the network using a differentiable function to pass data from one layer to the next. Convolutional neural networks are mainly composed of three types of layers: a convolutional layer, a pooling layer and a full-link layer. By adding these layers together, a complete convolutional neural network can be constructed.
Convolutional layers are one of the most important steps in convolutional neural architecture, and relate to the quality of feature expression, and also account for more than 95% of the computation of the whole network. Convolution is a linear, translation-invariant operation.
Pooling layers are periodically interposed between successive convolutional layers. The method has the function of gradually reducing the space size of the data volume, thereby reducing the number of parameters in the network, reducing the consumption of computing resources and effectively controlling overfitting.
The fully-linked layer is a traditional neural network, and each neural unit is densely connected with all the neural units of the previous layer. The fully-connected layer is used for final linear classification, which is equivalent to performing linear combination on the extracted high-layer feature vectors and outputting a final prediction result.
5. The new image input system is interpreted and the interpretation result is compared with the pathological tissue examination result of the patient.
6. If the interpretation result is accurate, outputting the result; and if the accuracy is not correct, correcting the convolutional neural network algorithm.
Through verification, the model constructed by the invention can accurately identify the NBI cystoscopy image.
The invention provides a system and a method for accurately analyzing a microscopy image of an NBI cystoscope, which can solve the problem of poor accuracy caused by manually identifying the image analysis at present, provide powerful auxiliary diagnosis information for bladder cancer and have excellent clinical application prospect.
Claims (8)
1. An analysis system for narrow band imaging cystoscopy images, comprising the following modules:
a basic image module: a basic NBI cystoscopy image library is formed by bladder cancer precancerous lesion images, bladder cancer tissue images and bladder normal tissue image samples which are acquired by an NBI cystoscope;
an input module: inputting the images acquired under the NBI cystoscope and the corresponding histopathological examination results into a system;
an image processing module: processing the image library of the basic image module to obtain a characteristic value set of a precancerous lesion image, a bladder cancer image and a normal tissue image; processing the image input by the image input module to obtain a test characteristic value;
an analysis learning module: constructing and training a deep learning network based on the feature value set; optimizing a deep learning network algorithm according to a result returned by the output module;
an interpretation module: inputting the test characteristic value into an analysis learning module for interpretation, and comparing the interpretation result with the histopathological examination result, wherein if the results are consistent, the interpretation result is accurate, and if the results are inconsistent, the interpretation result is inaccurate;
an output module: if the interpretation result is not accurate, returning the result to the learning module optimization algorithm; and if the interpretation result is accurate, outputting the interpretation result.
2. The analysis system of claim 1, wherein the processing by the image processing module comprises normalizing, region of interest delineating, and feature value detection of the NBI cystoscopy image.
3. The analysis system of claim 1, wherein the deep learning network is a convolutional neural network, consisting of convolutional layers, pooling layers, and full link layers.
4. A method of constructing an analysis system according to any one of claims 1 to 3, comprising the steps of:
1) collecting bladder precancerous lesion images, bladder cancer tissue images and bladder normal tissue image samples collected by an NBI cystoscope to form a basic NBI cystoscope examination image library;
2) processing the image library of the basic image module to obtain a characteristic value set of a precancerous lesion image, a bladder cancer image and a normal tissue image;
3) constructing and training a deep learning network based on the feature value set;
4) the analysis system is optimized with inaccurate interpretation results.
5. A method for analyzing narrow band imaging cystoscopic images, characterized in that the analysis is performed by using the analysis system of claims 1 to 3 by the following steps:
(1) acquiring an NBI cystoscopy image;
(2) processing the NBI cystoscopy image to obtain a characteristic value;
(3) and inputting the characteristic value into the trained deep learning network for processing and distinguishing.
6. The analysis method according to claim 5, wherein the image processing method of step (2) is: and carrying out normalization processing, sketching the region of interest and detecting to obtain a characteristic value.
7. The analysis method of claim 5, wherein the deep learning network of step (3) is a convolutional neural network, which is composed of convolutional layers, pooling layers, and full link layers.
8. The method of claim 7, wherein the convolutional neural network is constructed based on a set of feature values obtained by image processing of a base NBI cystoscopy image database;
the basic NBI cystoscopy image database consists of bladder precancerous lesion images, bladder cancer tissue images and bladder normal tissue image samples which are acquired by an NBI cystoscope.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110602974.7A CN113450310A (en) | 2021-05-31 | 2021-05-31 | Analysis system and method for narrow-band light imaging cystoscopy image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110602974.7A CN113450310A (en) | 2021-05-31 | 2021-05-31 | Analysis system and method for narrow-band light imaging cystoscopy image |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113450310A true CN113450310A (en) | 2021-09-28 |
Family
ID=77810517
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110602974.7A Pending CN113450310A (en) | 2021-05-31 | 2021-05-31 | Analysis system and method for narrow-band light imaging cystoscopy image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113450310A (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106991439A (en) * | 2017-03-28 | 2017-07-28 | 南京天数信息科技有限公司 | Image-recognizing method based on deep learning and transfer learning |
CN109523532A (en) * | 2018-11-13 | 2019-03-26 | 腾讯科技(深圳)有限公司 | Image processing method, device, computer-readable medium and electronic equipment |
WO2020206337A1 (en) * | 2019-04-03 | 2020-10-08 | The Board Of Trustees Of The Leland Stanford Junior University | Methods and systems for cystoscopic imaging incorporating machine learning |
CN111862075A (en) * | 2020-07-30 | 2020-10-30 | 西南医科大学 | Lung image analysis system and method based on deep learning |
WO2021016131A1 (en) * | 2019-07-19 | 2021-01-28 | The Jackson Laboratory | Convolutional neural networks for classification of cancer histological images |
KR102218255B1 (en) * | 2020-09-25 | 2021-02-19 | 정안수 | System and method for analyzing image based on artificial intelligence through learning of updated areas and computer program for the same |
CN112435743A (en) * | 2020-12-09 | 2021-03-02 | 上海市第一人民医院 | Bladder cancer pathological omics intelligent diagnosis method based on machine learning and prognosis model thereof |
-
2021
- 2021-05-31 CN CN202110602974.7A patent/CN113450310A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106991439A (en) * | 2017-03-28 | 2017-07-28 | 南京天数信息科技有限公司 | Image-recognizing method based on deep learning and transfer learning |
CN109523532A (en) * | 2018-11-13 | 2019-03-26 | 腾讯科技(深圳)有限公司 | Image processing method, device, computer-readable medium and electronic equipment |
WO2020206337A1 (en) * | 2019-04-03 | 2020-10-08 | The Board Of Trustees Of The Leland Stanford Junior University | Methods and systems for cystoscopic imaging incorporating machine learning |
WO2021016131A1 (en) * | 2019-07-19 | 2021-01-28 | The Jackson Laboratory | Convolutional neural networks for classification of cancer histological images |
CN111862075A (en) * | 2020-07-30 | 2020-10-30 | 西南医科大学 | Lung image analysis system and method based on deep learning |
KR102218255B1 (en) * | 2020-09-25 | 2021-02-19 | 정안수 | System and method for analyzing image based on artificial intelligence through learning of updated areas and computer program for the same |
CN112435743A (en) * | 2020-12-09 | 2021-03-02 | 上海市第一人民医院 | Bladder cancer pathological omics intelligent diagnosis method based on machine learning and prognosis model thereof |
Non-Patent Citations (4)
Title |
---|
ATSUSHI IKEDA 等: "Support System of Cystoscopic Diagnosis for Bladder Cancer Based on Artificial Intelligence" * |
HIROYA UEYAMA 等: "Application of artificial intelligence using a convolutional neural network for diagnosis of early gastric cancer based on magnifying endoscopy with narrow-band imaging" * |
LAN LI 等: "Convolutional neural network for the diagnosis of early gastric cancer based on magnifying narrow band imaging" * |
陈斌 等: "深度学习GoogleNet模型支持下的中分辨率遥感影像自动分类" * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Chen et al. | GasHis-Transformer: A multi-scale visual transformer approach for gastric histopathological image detection | |
CN111798425B (en) | Intelligent detection method for mitotic image in gastrointestinal stromal tumor based on deep learning | |
Xie et al. | Interpretable classification from skin cancer histology slides using deep learning: A retrospective multicenter study | |
Ghosh et al. | Deep transfer learning for automated intestinal bleeding detection in capsule endoscopy imaging | |
CN104424386A (en) | Multi-parameter magnetic resonance image based prostate cancer computer auxiliary identification system | |
US20210374953A1 (en) | Methods for automated detection of cervical pre-cancers with a low-cost, point-of-care, pocket colposcope | |
CN115019049A (en) | Bone imaging bone lesion segmentation method, system and equipment based on deep neural network | |
Du et al. | Improving the classification performance of esophageal disease on small dataset by semi-supervised efficient contrastive learning | |
Tan et al. | Lightweight pyramid network with spatial attention mechanism for accurate retinal vessel segmentation | |
Jing et al. | A comprehensive survey of intestine histopathological image analysis using machine vision approaches | |
Tenali et al. | Oral Cancer Detection using Deep Learning Techniques | |
CN117036288A (en) | Tumor subtype diagnosis method for full-slice pathological image | |
Singh et al. | Explainable information retrieval using deep learning for medical images | |
Garcia-Peraza-Herrera et al. | Interpretable fully convolutional classification of intrapapillary capillary loops for real-time detection of early squamous neoplasia | |
Sun et al. | Liver tumor segmentation and subsequent risk prediction based on Deeplabv3+ | |
Wang et al. | Controlling False-Positives in Automatic Lung Nodule Detection by Adding 3D Cuboid Attention to a Convolutional Neural Network | |
Li et al. | A dual attention-guided 3D convolution network for automatic segmentation of prostate and tumor | |
CN113450310A (en) | Analysis system and method for narrow-band light imaging cystoscopy image | |
CN115471512A (en) | Medical image segmentation method based on self-supervision contrast learning | |
CN111798426B (en) | Deep learning and detecting system for mitotic image in gastrointestinal stromal tumor of moving end | |
CN114271763A (en) | Mask RCNN-based gastric cancer early identification method, system and device | |
Mathina Kani et al. | Classification of skin lesion images using modified Inception V3 model with transfer learning and augmentation techniques | |
Ye et al. | Segmentation and feature extraction of endoscopic images for making diagnosis of acute appendicitis | |
Wang et al. | Three feature streams based on a convolutional neural network for early esophageal cancer identification | |
Liu et al. | Gastrointestinal stromal tumors diagnosis on multi-center endoscopic ultrasound images using multi-scale image normalization and transfer learning |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210928 |