WO2020215593A1 - Artificial intelligence-based automatic evaluation method and system on quality check of gastrointestinal endoscopy - Google Patents

Artificial intelligence-based automatic evaluation method and system on quality check of gastrointestinal endoscopy Download PDF

Info

Publication number
WO2020215593A1
WO2020215593A1 PCT/CN2019/106029 CN2019106029W WO2020215593A1 WO 2020215593 A1 WO2020215593 A1 WO 2020215593A1 CN 2019106029 W CN2019106029 W CN 2019106029W WO 2020215593 A1 WO2020215593 A1 WO 2020215593A1
Authority
WO
WIPO (PCT)
Prior art keywords
colonoscopy
time
rate
quality control
artificial intelligence
Prior art date
Application number
PCT/CN2019/106029
Other languages
French (fr)
Chinese (zh)
Inventor
于天成
Original Assignee
武汉楚精灵医疗科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 武汉楚精灵医疗科技有限公司 filed Critical 武汉楚精灵医疗科技有限公司
Publication of WO2020215593A1 publication Critical patent/WO2020215593A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • 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
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06395—Quality analysis or management
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H30/00—ICT specially adapted for the handling or processing of medical images
    • G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS

Definitions

  • the invention belongs to the field of medical technology assistance, and specifically relates to an artificial intelligence-based method for automatically evaluating the quality of digestive endoscopy.
  • the "World Cancer Report” pointed out that cancer is one of the main causes of death in the world. In 2015, it caused 8.8 million deaths. The total annual economic cost caused by cancer is about 1.16 trillion U.S. dollars, which cannot be stopped by traditional surgery, radiotherapy and chemotherapy alone. The spread of the cancer crisis. Gastrointestinal tumors are one of the most common malignant tumors. In 2015, the number of patients with gastric cancer and colon cancer in my country was more than 1 million, and the death toll was nearly 700,000, accounting for 1/4 of the total cancer deaths. The fundamental reason that malignant tumors endanger human health is that it is difficult to detect early. If gastrointestinal tumors are diagnosed at an early stage, the 5-year survival rate of patients can be higher than 90%, and if it progresses to advanced stages, the 5-year survival rate of patients is only 5-25%. Therefore, early diagnosis is an important strategy to improve patient survival.
  • Endoscopy is the most commonly used powerful tool to find gastrointestinal cancer. Since the reform and opening up, my country's digestive endoscopy industry has developed rapidly. In 2015, more than 6,000 medical institutions in my country carried out digestive endoscopy, and nearly 30,000 digestive endoscopy doctors completed nearly 30 million digestive endoscopy examinations, ranking first in the world in the number of diagnosis and treatment cases. However, behind the rapid development of technology, some medical quality and safety risks also exist. The level of different endoscopists is uneven, and the quality of the examination cannot be fully guaranteed.
  • the technical problem to be solved by the present invention is to provide an intelligent and efficient digestive endoscopy evaluation tool and establish a scientific and rigorous artificial intelligence evaluation system.
  • an artificial intelligence-based digestive endoscopy quality assessment system including:
  • Medical imaging report system The doctor manually enters patient information, collects endoscopic images through gastrointestinal endoscopy equipment, and records the corresponding operation time;
  • Pre-training model based on big data Collect a large number of endoscopic images, professional endoscopists classify and annotate the images according to the anatomical parts and lesion features, construct pre-processed image sets, and use pre-trained images through transfer learning based on parameters/features. Process the image set to train the deep convolutional neural network model to obtain a pre-trained model that can automatically identify the digestive tract and lesions;
  • Quality control data statistics software Based on the calculation formula of the endoscopic operation quality control index, using the text data from the medical imaging reporting system and the prediction result of the picture from the pre-training model, automatically calculate multiple quality control indexes of the endoscopic operation;
  • the quality control indicators include, gastroscopy time, gastroscopy blind zone rate, colonoscopy cecal intubation rate, colonoscopy withdrawal time, colorectal polyp detection rate, colorectal adenoma detection rate, successful bowel preparation Rate, the detection rate of early gastrointestinal cancer;
  • gastroscopy time end time of gastroscope-start time of gastroscope-time for biopsy or treatment.
  • the specific definition of the rate of colonoscopy and cecal intubation is: the proportion of successful cases of colonoscopy and cecal intubation in the total number of colonoscopy during the same period.
  • colonoscopy withdrawal time time from the cecum to rectum-the time to find adenoma / polyp biopsy or treatment.
  • colorectal polyp detection rate colon Number of polyp cases detected by endoscopy/total number of all colonoscopy examinations in the same period ⁇ 100%.
  • colorectal adenoma detection Rate number of cases of adenoma detected by colonoscopy/total number of all colonoscopy during the same period ⁇ 100%.
  • the specific definition of the detection rate of early gastrointestinal cancer is: the number of patients with early esophageal cancer, gastric cancer or colorectal cancer detected by gastrointestinal endoscopy accounts for the proportion of the number of patients with esophageal cancer, gastric cancer or colorectal cancer in the same period.
  • the present invention also provides a method for automatically evaluating the quality of digestive endoscopy based on artificial intelligence, which includes the following steps:
  • S1 The doctor manually enters patient information, collects endoscopic images, and automatically records patient information, collected pictures, and corresponding operation time through the image reporting system;
  • S2 Obtain patient information, collected pictures, and corresponding operation time through a data exchange protocol, in which text data is input to quality control data statistics software, and picture data is input to a deep convolutional neural network model;
  • the deep convolutional neural network model receives the gastrointestinal endoscopy image, recognizes the location and lesion features of the image, and inputs the prediction result into the quality control data statistics software;
  • the quality control data statistics software uses the text data from the medical imaging report system and the prediction results of the picture from the pre-training model according to the calculation formula of the endoscopic operation quality control index to automatically calculate multiple quality control indexes of the endoscopic operation;
  • the quality control indicators include, gastroscopy time, gastroscopy blind zone rate, colonoscopy cecal intubation rate, colonoscopy withdrawal time, colorectal polyp detection rate, colorectal adenoma detection rate, successful bowel preparation Rate, the detection rate of early gastrointestinal cancer;
  • the quality control data statistical software outputs the statistical results through charts and other forms, and doctors and managers can view the statistical results by themselves.
  • the beneficial effects of the present invention are: through the present invention, the inspection quality of digestive endoscopy diagnosis and treatment institutions is evaluated and displayed. On the one hand, it can objectively and directly quantify the current operating level of physicians, encourage endoscopists to learn from each other, and continuously improve operations. Level. On the other hand, it is also possible for the superior medical management platform to obtain comprehensive and accurate quality reports of digestive endoscopy diagnosis and treatment institutions within its jurisdiction, and to perform quality control in a timely manner.
  • Figure 1 is a system structure diagram of the present invention.
  • an artificial intelligence-based digestive endoscopy quality assessment system developed by the present invention includes:
  • Medical imaging report system The doctor manually enters patient information and collects endoscopic images through gastrointestinal endoscopy equipment.
  • the hospital's medical imaging report system can automatically record patient information, collected pictures, and corresponding operating time.
  • Pre-training model based on big data A large number of endoscopic images are collected, and professional endoscopists classify and label the images according to the anatomical parts and the characteristics of the lesion, and build a pre-processed image set. Through transfer learning based on parameters/features, a pre-processed image set is used to train a model (usually a deep convolutional neural network model in deep learning) to obtain a pre-trained model that can automatically identify parts of the digestive tract and lesions.
  • a model usually a deep convolutional neural network model in deep learning
  • Quality control data statistical software Based on the calculation formula of the endoscopic operation quality control index, using the text data from the medical imaging report system and the prediction result of the picture from the pre-training model, multiple quality control indexes of the endoscopic operation are automatically calculated.
  • the present invention also provides a method for automatically evaluating the quality of digestive endoscopy based on artificial intelligence, which includes the following steps:
  • the doctor manually enters patient information and collects endoscopic images.
  • the hospital s imaging report system automatically records patient information, collected pictures and corresponding operating time;
  • the pre-training model receives the gastrointestinal endoscopy image, recognizes the location and lesion features of the image, and inputs the prediction result into the quality control data statistics software;
  • the quality control data statistics software uses the text data from the medical imaging report system and the prediction results of the picture from the pre-training model according to the calculation formula of the endoscopic operation quality control index to automatically calculate multiple quality control indexes for the endoscopic operation.
  • a number of statistical quality control indicators include but are not limited to: gastroscopy time, gastroscopy blind area rate, colonoscopy cecal intubation rate, colonoscopy withdrawal time, colorectal polyp detection rate, colorectal adenoma detection rate, Intestinal preparation success rate, early detection rate of digestive tract cancer;
  • the quality control data statistical software outputs statistical results in the form of charts and other forms, and doctors and managers can view the statistical results by themselves.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Theoretical Computer Science (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Physics & Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Development Economics (AREA)
  • General Physics & Mathematics (AREA)
  • Educational Administration (AREA)
  • Tourism & Hospitality (AREA)
  • Game Theory and Decision Science (AREA)
  • Data Mining & Analysis (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • General Business, Economics & Management (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Computational Linguistics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • Molecular Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Mathematical Physics (AREA)
  • Evolutionary Biology (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Epidemiology (AREA)
  • Medical Informatics (AREA)
  • Primary Health Care (AREA)

Abstract

Disclosed are an artificial intelligence-based automatic evaluation method and system on the quality check of a gastrointestinal endoscopy, aimed at accurately, comprehensively and rapidly evaluating the quality check of the gastrointestinal endoscopy in medical institutions, and providing a feasible supervision basis for improving the quality of the gastrointestinal endoscopy. The quality check of gastrointestinal endoscopy diagnosis and treatment institutions is evaluated and displayed. On one hand, the current operant level of a physician is quantitatively expressed objectively and directly, thereby encouraging endoscopists to learn from each other to continously improve the operant level; on the other hand, a superior medical management platform is allowed to comprehensively and accurately obtain the quality reports of gastrointestinal endoscopy diagnosis and treatment institutions within its jurisdiction and to perform quality control timely.

Description

一种基于人工智能的消化内镜检查质量自动评估方法和系统A method and system for automatically evaluating the quality of digestive endoscopy based on artificial intelligence 技术领域Technical field
本发明属于医疗技术辅助领域，具体涉及一种基于人工智能的消化内镜检查质量自动评估方法。The invention belongs to the field of medical technology assistance, and specifically relates to an artificial intelligence-based method for automatically evaluating the quality of digestive endoscopy.
背景技术Background technique
《世界癌症报告》指出，癌症是全球主要死因之一，2015年导致880万人死亡，由癌症导致的年度经济总费用约为1.16万亿美元，仅靠传统的外科手术、放疗和化疗无法遏制癌症危机的蔓延。消化道肿瘤是最常见的恶性肿瘤之一。2015年我国胃癌、结肠癌患病人数大于100万，死亡人数近70万，占癌症总死亡人数的1/4。恶性肿瘤危害人类健康的根本原因是难以早期发现。消化道肿瘤若在早期阶段得到诊断，患者5年生存率可高于90％，若进展至中晚期，患者5年生存率仅为5-25％。因此，早期诊断是提高患者生存率的重要策略。The "World Cancer Report" pointed out that cancer is one of the main causes of death in the world. In 2015, it caused 8.8 million deaths. The total annual economic cost caused by cancer is about 1.16 trillion U.S. dollars, which cannot be stopped by traditional surgery, radiotherapy and chemotherapy alone. The spread of the cancer crisis. Gastrointestinal tumors are one of the most common malignant tumors. In 2015, the number of patients with gastric cancer and colon cancer in my country was more than 1 million, and the death toll was nearly 700,000, accounting for 1/4 of the total cancer deaths. The fundamental reason that malignant tumors endanger human health is that it is difficult to detect early. If gastrointestinal tumors are diagnosed at an early stage, the 5-year survival rate of patients can be higher than 90%, and if it progresses to advanced stages, the 5-year survival rate of patients is only 5-25%. Therefore, early diagnosis is an important strategy to improve patient survival.
内镜检查是发现胃肠癌最常用的有力工具。改革开放以来，我国消化内镜行业迅速发展。2015年，我国共有6000多家医疗机构开展消化内镜，共有近3万名消化内镜医生，完成近3000万例次的消化内镜检查，诊疗例数位居世界第一。然而，在技术蓬勃发展的背后，一些医疗质量和安全隐患也现实存在。不同内镜医师的水平参差不齐，检查质量难以完全保证。据国家卫健委内镜诊疗技术专家组报道，2017年我国早期胃癌发现率仅为13％，远远落后于疾病谱相仿的日本(70％)和韩国(50％)等邻国；而湖北省的早期胃癌发现率约为10％，低于全国平均值，在部分基层医院，甚至一年也发现不了一例早癌。这严重威胁着我省乃至全国广大患者的生命健康。Endoscopy is the most commonly used powerful tool to find gastrointestinal cancer. Since the reform and opening up, my country's digestive endoscopy industry has developed rapidly. In 2015, more than 6,000 medical institutions in my country carried out digestive endoscopy, and nearly 30,000 digestive endoscopy doctors completed nearly 30 million digestive endoscopy examinations, ranking first in the world in the number of diagnosis and treatment cases. However, behind the rapid development of technology, some medical quality and safety risks also exist. The level of different endoscopists is uneven, and the quality of the examination cannot be fully guaranteed. According to the National Health Commission’s Endoscopic Diagnosis and Treatment Technical Expert Group, the detection rate of early gastric cancer in my country in 2017 was only 13%, which is far behind neighboring countries such as Japan (70%) and South Korea (50%) with similar disease spectrums; while Hubei The detection rate of early gastric cancer in the province is about 10%, which is lower than the national average. In some primary hospitals, even one case of early cancer is not found in one year. This seriously threatens the lives and health of patients in our province and even the whole country.
为了提高我国消化内镜检查质量，国家卫健委和消化内镜各领域专家们付出了大量的努力。2009年，在卫生部的指导下，各省市相继成立消化内镜质控中心。2015年，国家消化内镜质控专家组制订《消化内镜医疗质量控制指标(草案)》，明确了多层次的消化内镜质量控制指标。2016年开始，国家消化质控中心每年组织全国消化内镜质量控制的普查，并进行实时的抽样调研，监督我国医疗机构消化内镜质量。然而，目前使用的质控措施包括专家组抽查、人工调研和医院自行上报等存在一定的弊端。2017年，全国有大于7000家医疗机构开展消化内镜，其中有4654家医院自行填报消化内镜技术质量数据，经二次抽样后仅有1294家医 院进入调查，无法实现消化内镜质量的全面评估，数据准确性难以完全保证。此外，依靠质量控制中心成员和消化专家走访各级医疗机构做人工调查，耗时又耗力，且突击检查无法客观反映医疗机构日常的诊疗水平。In order to improve the quality of digestive endoscopy in my country, the National Health Commission and experts in various fields of digestive endoscopy have made great efforts. In 2009, under the guidance of the Ministry of Health, various provinces and cities successively established digestive endoscopy quality control centers. In 2015, the National Digestive Endoscopy Quality Control Expert Group formulated the "Digestive Endoscopy Medical Quality Control Indicators (Draft)", which clarified multi-level digestive endoscopy quality control indicators. Since 2016, the National Digestive Quality Control Center has organized a national general survey of digestive endoscopy quality control every year, and conducted real-time sampling surveys to supervise the quality of digestive endoscopy in medical institutions in my country. However, the current quality control measures, including random inspections by expert groups, manual investigations, and self-reporting by hospitals, have certain drawbacks. In 2017, more than 7000 medical institutions across the country carried out digestive endoscopy, of which 4654 hospitals self-reported the technical quality data of digestive endoscopy. After sub-sampling, only 1294 hospitals entered the survey, failing to achieve comprehensive digestive endoscopy quality Evaluation, data accuracy is difficult to guarantee completely. In addition, relying on members of the Quality Control Center and digestive experts to visit medical institutions at all levels for manual investigations is time-consuming and labor-intensive, and surprise inspections cannot objectively reflect the daily diagnosis and treatment level of medical institutions.
近年来科技迅猛发展，人工智能掀起了新一波的技术浪潮。随着自动驾驶汽车的测试成功、AlphaGo击败围棋世界冠军,短短几年时间内，人工智能逐步进入公众视野。在汽车领域，计算机视觉已成功用于监控驾驶员行为，分析驾驶员的策略和状态，并在发生风险时实时反馈和发出警报。在食品行业，人工智能已成功应用于食品质量监测预警系统。然而，尚无利用人工智能监控消化内镜检查质量的应用研究。基于此，我们拟发明一种基于人工智能的消化内镜检查质量自动评估方法，准确、全面、快速的评估医疗机构的胃肠镜检查质量，为提高消化内镜质量提供切实可行的督导依据。In recent years, science and technology have developed rapidly, and artificial intelligence has set off a new wave of technology. With the successful testing of self-driving cars and AlphaGo defeating the world champion of Go, in just a few years, artificial intelligence has gradually entered the public eye. In the automotive field, computer vision has been successfully used to monitor driver behavior, analyze the driver's strategy and status, and provide real-time feedback and alarms when risks occur. In the food industry, artificial intelligence has been successfully applied to food quality monitoring and early warning systems. However, there is no applied research on using artificial intelligence to monitor the quality of digestive endoscopy. Based on this, we intend to invent an artificial intelligence-based automatic evaluation method for the quality of digestive endoscopy, which accurately, comprehensively and quickly evaluates the quality of gastrointestinal endoscopy in medical institutions, and provides a practical and feasible supervision basis for improving the quality of digestive endoscopy.
发明内容Summary of the invention
本发明要解决的技术问题是：提供一种智能高效的消化内镜评估工具，建立科学严谨的人工智能评估体系。The technical problem to be solved by the present invention is to provide an intelligent and efficient digestive endoscopy evaluation tool and establish a scientific and rigorous artificial intelligence evaluation system.
本发明为解决上述技术问题所采取的技术方案为：一种基于人工智能的消化内镜检查质量评估系统，包括：The technical solution adopted by the present invention to solve the above technical problems is: an artificial intelligence-based digestive endoscopy quality assessment system, including:
医学影像报告系统：医生手动录入病人信息，并通过胃肠镜设备采集内镜图像，记录相应的操作时间；Medical imaging report system: The doctor manually enters patient information, collects endoscopic images through gastrointestinal endoscopy equipment, and records the corresponding operation time;
基于大数据的预训练模型：收集大量的内镜图像，由专业内镜医师根据解剖学部位和病灶特征对图像进行分类标注，构建预处理图像集，通过基于参数/特征的迁移学习，使用预处理图像集训练深度卷积神经网络模型，得到可自动辨认消化道部位和病灶的预训练模型；Pre-training model based on big data: Collect a large number of endoscopic images, professional endoscopists classify and annotate the images according to the anatomical parts and lesion features, construct pre-processed image sets, and use pre-trained images through transfer learning based on parameters/features. Process the image set to train the deep convolutional neural network model to obtain a pre-trained model that can automatically identify the digestive tract and lesions;
质控数据统计软件：基于内镜操作质量控制指标的计算公式，使用来自医学影像报告系统的文字资料和来自预训练模型对于图片的预测结果，自动计算内镜操作的多项质量控制指标；Quality control data statistics software: Based on the calculation formula of the endoscopic operation quality control index, using the text data from the medical imaging reporting system and the prediction result of the picture from the pre-training model, automatically calculate multiple quality control indexes of the endoscopic operation;
其中，所述质量控制指标包括，胃镜检查时间，胃镜检查盲区率，肠镜盲肠插管率，肠镜退镜时间，结直肠息肉检出率，结直肠腺瘤检出率，肠道准备成功率，消化道早癌发现率；Among them, the quality control indicators include, gastroscopy time, gastroscopy blind zone rate, colonoscopy cecal intubation rate, colonoscopy withdrawal time, colorectal polyp detection rate, colorectal adenoma detection rate, successful bowel preparation Rate, the detection rate of early gastrointestinal cancer;
用户：包括医生和管理者，用来查看质控数据统计软件输出的统计结果。User: including doctors and managers, used to view the statistical results output by the quality control data statistical software.
进一步的，胃镜检查时间的具体定义为：从内窥镜插管到拔管的时间，计算公式为：胃镜检查时间＝胃镜结束时间-胃镜开始时间-进行活检或治疗的时间。Further, the specific definition of gastroscopy time is: the time from endoscope intubation to extubation, and the calculation formula is: gastroscopy time = end time of gastroscope-start time of gastroscope-time for biopsy or treatment.
进一步的，胃镜检查盲区率的具体定义为：胃镜未检查到的部位占胃镜26部位的比率，计算公式为：胃镜检查盲区率＝1-被观察到的胃镜下部位数/胃镜下部位总数×100％。Further, the specific definition of the rate of the blind area of gastroscopy is: the ratio of the parts that are not detected by the gastroscope in the 26 parts of the gastroscope, and the calculation formula is: the rate of the blind area of the gastroscopy = 1-the number of lower parts of the gastroscope observed / the total number of parts under the gastroscope × 100%.
进一步的，肠镜盲肠插管率的具体定义为：肠镜检查盲肠插管成功例数占同期肠镜检查总数的比例，计算公式为：肠镜盲肠插管率＝肠镜检查盲肠插管成功例数/同期肠镜检查总数×100％。Further, the specific definition of the rate of colonoscopy and cecal intubation is: the proportion of successful cases of colonoscopy and cecal intubation in the total number of colonoscopy during the same period. The calculation formula is: colonoscopy and cecum intubation rate = successful colonoscopy and cecum intubation Number of cases/total number of colonoscopy during the same period × 100%.
进一步的，肠镜退镜时间的具体定义为：结肠镜检查过程中，从进镜到达盲肠开始到退镜至直肠之间的实际时间，不包括对息肉进行活检等额外操作的时间，即阴性结肠镜退镜时间，计算公式为：肠镜退镜时间＝从盲肠退镜至直肠的时间-发现腺瘤/对息肉进行活检或治疗的时间。Further, the specific definition of colonoscopy withdrawal time is: the actual time from the time the colonoscopy reaches the cecum to the withdrawal to the rectum during colonoscopy, excluding the time for additional operations such as polyp biopsy, ie negative Colonoscopy withdrawal time, the calculation formula is: colonoscopy withdrawal time = time from the cecum to rectum-the time to find adenoma / polyp biopsy or treatment.
进一步的，结直肠息肉检出率的具体定义为：肠镜检查中至少检出一枚结直肠息肉的患者数占同期结肠镜检查总数的比例，计算公式为：结直肠息肉检出率＝结肠镜检查出息肉病例数/同期所有结肠镜检查总数×100％。Further, the specific definition of the detection rate of colorectal polyps is: the proportion of the number of patients with at least one colorectal polyp detected in colonoscopy to the total number of colonoscopy in the same period, the calculation formula is: colorectal polyp detection rate = colon Number of polyp cases detected by endoscopy/total number of all colonoscopy examinations in the same period × 100%.
进一步的，结直肠腺瘤检出率的具体定义为：肠镜检查中至少检出一枚结直肠腺瘤的患者数占同期结肠镜检查总数的比例，计算公式为：结直肠腺瘤检出率＝结肠镜检查出腺瘤病例数/同期所有结肠镜检查总数×100％。Furthermore, the specific definition of the detection rate of colorectal adenoma is: the proportion of the number of patients with at least one colorectal adenoma detected in colonoscopy to the total number of colonoscopy during the same period, the calculation formula is: colorectal adenoma detection Rate = number of cases of adenoma detected by colonoscopy/total number of all colonoscopy during the same period×100%.
进一步的，肠道准备成功率的具体定义为：肠腔内有少量或者无粪渣且不影响肠镜观察的患者数占同期结肠镜检查总数的比例，计算公式为：肠道准备成功率＝1-肠腔内包含不合格图像的案例数量/同期结肠镜检查总数×100％。Further, the specific definition of the success rate of bowel preparation is: the proportion of the number of patients with a small amount or no fecal residue in the intestinal cavity that does not affect colonoscopy observations to the total number of colonoscopy examinations in the same period, the calculation formula is: bowel preparation success rate = 1- The number of cases containing unqualified images in the intestinal cavity/total number of colonoscopy in the same period × 100%.
进一步的，消化道早癌发现率的具体定义为：胃肠镜检查发现早期食管癌、胃癌或结直肠癌的患者数分别占同期食管癌、胃癌或结直肠癌患者数的比例，计算公式为：消化道早癌发现率＝胃肠镜检查发现早癌的患者数/同期早癌和中晚期癌患者总数×100％。Further, the specific definition of the detection rate of early gastrointestinal cancer is: the number of patients with early esophageal cancer, gastric cancer or colorectal cancer detected by gastrointestinal endoscopy accounts for the proportion of the number of patients with esophageal cancer, gastric cancer or colorectal cancer in the same period. The calculation formula is : Detection rate of early gastrointestinal cancer = number of patients with early cancer found by gastrointestinal endoscopy/total number of patients with early cancer and advanced cancer in the same period × 100%.
本发明还提供一种基于人工智能的消化内镜检查质量自动评估的方法，包括如下步骤：The present invention also provides a method for automatically evaluating the quality of digestive endoscopy based on artificial intelligence, which includes the following steps:
S1，医生手动录入病人信息、采集内镜图像，通过影像报告系统自动记录病人信息、被采集的图片和相应的操作时间；S1: The doctor manually enters patient information, collects endoscopic images, and automatically records patient information, collected pictures, and corresponding operation time through the image reporting system;
S2，通过数据交换协议获取病人信息、被采集的图片和相应的操作时间，其中文字资料输入质控数据统计软件，图片资料输入深度卷积神经网络模型；S2: Obtain patient information, collected pictures, and corresponding operation time through a data exchange protocol, in which text data is input to quality control data statistics software, and picture data is input to a deep convolutional neural network model;
S3，深度卷积神经网络模型接收胃肠镜图像，对图像进行部位特征和病灶特征识别，将预测结果输入质控数据统计软件；S3, the deep convolutional neural network model receives the gastrointestinal endoscopy image, recognizes the location and lesion features of the image, and inputs the prediction result into the quality control data statistics software;
S4，质控数据统计软件根据内镜操作质量控制指标的计算公式，使用来自医学影像报告系统的文字资料和来自预训练模型对于图片的预测结果，自动计算内镜操作的多项质量控制指标；S4, the quality control data statistics software uses the text data from the medical imaging report system and the prediction results of the picture from the pre-training model according to the calculation formula of the endoscopic operation quality control index to automatically calculate multiple quality control indexes of the endoscopic operation;
其中，所述质量控制指标包括，胃镜检查时间，胃镜检查盲区率，肠镜盲肠插管率，肠镜退镜时间，结直肠息肉检出率，结直肠腺瘤检出率，肠道准备成功率，消化道早癌发现率；Among them, the quality control indicators include, gastroscopy time, gastroscopy blind zone rate, colonoscopy cecal intubation rate, colonoscopy withdrawal time, colorectal polyp detection rate, colorectal adenoma detection rate, successful bowel preparation Rate, the detection rate of early gastrointestinal cancer;
S5，质控数据统计软件通过图表等形式将统计结果输出，医生和管理者可自行查看统计结果。S5, the quality control data statistical software outputs the statistical results through charts and other forms, and doctors and managers can view the statistical results by themselves.
本发明的有益效果为：通过本发明对消化内镜诊疗机构检查质量进行评估并在进行显示，一方面客观直接的将医师当前的操作水平定量化表达，激励内镜医师互相学习，不断提高操作水平。另一方面也可以让上级医疗管理平台全面准确地获取管辖内的消化内镜诊疗机构的质量报表，及时做好质量控制。The beneficial effects of the present invention are: through the present invention, the inspection quality of digestive endoscopy diagnosis and treatment institutions is evaluated and displayed. On the one hand, it can objectively and directly quantify the current operating level of physicians, encourage endoscopists to learn from each other, and continuously improve operations. Level. On the other hand, it is also possible for the superior medical management platform to obtain comprehensive and accurate quality reports of digestive endoscopy diagnosis and treatment institutions within its jurisdiction, and to perform quality control in a timely manner.
附图说明Description of the drawings
图1为本发明的系统结构图。Figure 1 is a system structure diagram of the present invention.
具体实施方式Detailed ways
下面结合附图和实施例对本发明的技术方案作进一步说明。The technical scheme of the present invention will be further described below in conjunction with the drawings and embodiments.
如图1所示，本发明开发的一种基于人工智能的消化内镜检查质量评估系统，它包括：As shown in Figure 1, an artificial intelligence-based digestive endoscopy quality assessment system developed by the present invention includes:
医学影像报告系统：医生手动录入病人信息，并通过胃肠镜设备采集内镜图像。医院的医学影像报告系统可以自动记录病人信息、被采集的图片和相应的操作时间。Medical imaging report system: The doctor manually enters patient information and collects endoscopic images through gastrointestinal endoscopy equipment. The hospital's medical imaging report system can automatically record patient information, collected pictures, and corresponding operating time.
基于大数据的预训练模型：收集大量的内镜图像，由专业内镜医师根据解剖学部位和病灶特征对图像进行分类标注，构建预处理图像集。通过基于参数/特征的迁移学习，使用预处理图像集训练模型(一般采用深度学习中的深度卷积神经网络模型)，得到可自动辨认消化道部位和病灶的预训练模型。Pre-training model based on big data: A large number of endoscopic images are collected, and professional endoscopists classify and label the images according to the anatomical parts and the characteristics of the lesion, and build a pre-processed image set. Through transfer learning based on parameters/features, a pre-processed image set is used to train a model (usually a deep convolutional neural network model in deep learning) to obtain a pre-trained model that can automatically identify parts of the digestive tract and lesions.
质控数据统计软件：基于内镜操作质量控制指标的计算公式，使用来自医学影像报告系统的文字资料和来自预训练模型对于图片的预测结果，自动计算内镜操作的多项质量控制指标。Quality control data statistical software: Based on the calculation formula of the endoscopic operation quality control index, using the text data from the medical imaging report system and the prediction result of the picture from the pre-training model, multiple quality control indexes of the endoscopic operation are automatically calculated.
Figure PCTCN2019106029-appb-000001
Figure PCTCN2019106029-appb-000001
Figure PCTCN2019106029-appb-000002
Figure PCTCN2019106029-appb-000002
Figure PCTCN2019106029-appb-000003
Figure PCTCN2019106029-appb-000003
用户：包括医生和管理者，用户可以查看质控数据统计软件输出的统计结果。User: Including doctors and managers, users can view the statistical results output by the quality control data statistical software.
本发明还提供一种基于人工智能的消化内镜检查质量自动评估的方法，包括以下步骤：The present invention also provides a method for automatically evaluating the quality of digestive endoscopy based on artificial intelligence, which includes the following steps:
S1.医生手动录入病人信息、采集内镜图像。医院的影像报告系统自动记录病人信息、被采集的图片和相应的操作时间；S1. The doctor manually enters patient information and collects endoscopic images. The hospital’s imaging report system automatically records patient information, collected pictures and corresponding operating time;
S2.通过数据交换协议获取病人信息、被采集的图片和相应的操作时间，其中文字资料输入质控数据统计软件，图片资料输入基于大数据的预训练模型；S2. Obtain patient information, collected pictures, and corresponding operation time through the data exchange protocol. The text data is input into the quality control data statistics software, and the picture data is input into the pre-training model based on big data;
S3.预训练模型接收胃肠镜图像，对图像进行部位特征和病灶特征识别，将预测结果输入质控数据统计软件；S3. The pre-training model receives the gastrointestinal endoscopy image, recognizes the location and lesion features of the image, and inputs the prediction result into the quality control data statistics software;
S4.质控数据统计软件根据内镜操作质量控制指标的计算公式，使用来自医学影像报告系统的文字资料和来自预训练模型对于图片的预测结果，自动计算内镜操作的多项质量控制指标。S4. The quality control data statistics software uses the text data from the medical imaging report system and the prediction results of the picture from the pre-training model according to the calculation formula of the endoscopic operation quality control index to automatically calculate multiple quality control indexes for the endoscopic operation.
所统计的多项质量控制指标包括但不限于：胃镜检查时间，胃镜检查盲区率，肠镜盲肠插管率，肠镜退镜时间，结直肠息肉检出率，结直肠腺瘤检出率，肠道准备成功率，消化道 早癌发现率；A number of statistical quality control indicators include but are not limited to: gastroscopy time, gastroscopy blind area rate, colonoscopy cecal intubation rate, colonoscopy withdrawal time, colorectal polyp detection rate, colorectal adenoma detection rate, Intestinal preparation success rate, early detection rate of digestive tract cancer;
S5.质控数据统计软件通过图表等形式将统计结果输出，医生和管理者可自行查看统计结果。S5. The quality control data statistical software outputs statistical results in the form of charts and other forms, and doctors and managers can view the statistical results by themselves.
其中胃镜检查时间，胃镜检查盲区率，肠镜盲肠插管率，肠镜退镜时间，结直肠息肉检出率，结直肠腺瘤检出率，肠道准备成功率，消化道早癌发现率的具体定义、计算公式以及实现步骤和质控数据统计软件中的具体实现方式相同。Including gastroscopy time, gastroscopy blind area rate, colonoscopy cecal intubation rate, colonoscopy withdrawal time, colorectal polyp detection rate, colorectal adenoma detection rate, bowel preparation success rate, early digestive tract cancer detection rate The specific definition, calculation formula, and implementation steps of is the same as the specific implementation in the quality control data statistics software.
本文中所描述的具体实施例仅仅是对本发明精神作举例说明。本发明所属技术领域的技术人员可以对所描述的具体实施例做各种各样的修改或补充或采用类似的方式替代，但并不会偏离本发明的精神或者超越所附权利要求书所定义的范围。The specific embodiments described herein are merely examples to illustrate the spirit of the present invention. Those skilled in the art to which the present invention pertains can make various modifications or additions to the specific embodiments described or use similar alternatives, but they will not deviate from the spirit of the present invention or exceed the definition of the appended claims. Range.

Claims (10)

  1. 一种基于人工智能的消化内镜检查质量评估系统，其特征在于，包括：An artificial intelligence-based digestive endoscopy quality assessment system, which is characterized in that it includes:
    医学影像报告系统：医生手动录入病人信息，并通过胃肠镜设备采集内镜图像，记录相应的操作时间；Medical imaging report system: The doctor manually enters patient information, collects endoscopic images through gastrointestinal endoscopy equipment, and records the corresponding operation time;
    基于大数据的预训练模型：收集大量的内镜图像，由专业内镜医师根据解剖学部位和病灶特征对图像进行分类标注，构建预处理图像集，通过基于参数/特征的迁移学习，使用预处理图像集训练深度卷积神经网络模型，得到可自动辨认消化道部位和病灶的预训练模型；Pre-training model based on big data: Collect a large number of endoscopic images, professional endoscopists classify and annotate the images according to the anatomical parts and lesion features, construct pre-processed image sets, and use pre-trained images through transfer learning based on parameters/features. Process the image set to train the deep convolutional neural network model to obtain a pre-trained model that can automatically identify the digestive tract and lesions;
    质控数据统计软件：基于内镜操作质量控制指标的计算公式，使用来自医学影像报告系统的文字资料和来自预训练模型对于图片的预测结果，自动计算内镜操作的多项质量控制指标；Quality control data statistics software: Based on the calculation formula of the endoscopic operation quality control index, using the text data from the medical imaging reporting system and the prediction result of the picture from the pre-training model, automatically calculate multiple quality control indexes of the endoscopic operation;
    其中，所述质量控制指标包括，胃镜检查时间，胃镜检查盲区率，肠镜盲肠插管率，肠镜退镜时间，结直肠息肉检出率，结直肠腺瘤检出率，肠道准备成功率，消化道早癌发现率；Among them, the quality control indicators include, gastroscopy time, gastroscopy blind zone rate, colonoscopy cecal intubation rate, colonoscopy withdrawal time, colorectal polyp detection rate, colorectal adenoma detection rate, successful bowel preparation Rate, the detection rate of early gastrointestinal cancer;
    用户：包括医生和管理者，用来查看质控数据统计软件输出的统计结果。User: including doctors and managers, used to view the statistical results output by the quality control data statistical software.
  2. 如权利要求1所述的一种基于人工智能的消化内镜检查质量评估系统，其特征在于：胃镜检查时间的具体定义为：从内窥镜插管到拔管的时间，计算公式为：胃镜检查时间＝胃镜结束时间-胃镜开始时间-进行活检或治疗的时间。The artificial intelligence-based digestive endoscopy quality assessment system according to claim 1, wherein the specific definition of gastroscopy time is: the time from endoscopic intubation to extubation, and the calculation formula is: gastroscopy Examination time = end time of gastroscope-start time of gastroscope-time for biopsy or treatment.
  3. 如权利要求1所述的一种基于人工智能的消化内镜检查质量评估系统，其特征在于：胃镜检查盲区率的具体定义为：胃镜未检查到的部位占胃镜26部位的比率，计算公式为：胃镜检查盲区率＝1-被观察到的胃镜下部位数/胃镜下部位总数×100％。The artificial intelligence-based digestive endoscopy quality assessment system according to claim 1, wherein the specific definition of the rate of the blind area of gastroscopy is: the ratio of the undetected parts of the gastroscope to the 26 parts of the gastroscope, and the calculation formula is ：Gastroscopy blind area rate = 1-the number of lower parts of the gastroscope observed/total number of parts under the gastroscope × 100%.
  4. 如权利要求1所述的一种基于人工智能的消化内镜检查质量评估系统，其特征在于：肠镜盲肠插管率的具体定义为：肠镜检查盲肠插管成功例数占同期肠镜检查总数的比例，计算公式为：肠镜盲肠插管率＝肠镜检查盲肠插管成功例数/同期肠镜检查总数×100％。The artificial intelligence-based digestive endoscopy quality assessment system according to claim 1, wherein the specific definition of the rate of colonoscopy and cecal intubation is: the number of successful colonoscopy and cecal intubation accounts for the number of successful cases of colonoscopy in the same period. The ratio of the total number, the calculation formula is: enteroscopy and cecal intubation rate = successful cases of enteroscopy and cecum intubation / total number of colonoscopy in the same period × 100%.
  5. 如权利要求1所述的一种基于人工智能的消化内镜检查质量评估系统，其特征在于：肠镜退镜时间的具体定义为：结肠镜检查过程中，从进镜到达盲肠开始到退镜至直肠之间的实际时间，不包括对息肉进行活检等额外操作的时间，即阴性结肠镜退镜时间，计算公式为：肠镜退镜时间＝从盲肠退镜至直肠的时间-发现腺瘤/对息肉进行活检或治疗的时间。The artificial intelligence-based digestive endoscopy quality assessment system according to claim 1, wherein the specific definition of colonoscopy withdrawal time is: during colonoscopy, from entering the cecum to withdrawing the mirror The actual time between the rectum and the rectum does not include the time for additional operations such as polyp biopsy, that is, the negative colonoscopy withdrawal time. The calculation formula is: colonoscopy withdrawal time = time from cecal withdrawal to rectum-adenoma found /Time for biopsy or treatment of polyps.
  6. 如权利要求1所述的一种基于人工智能的消化内镜检查质量评估系统，其特征在于：结直肠息肉检出率的具体定义为：肠镜检查中至少检出一枚结直肠息肉的患者数占同期结肠镜检查总数的比例，计算公式为：结直肠息肉检出率＝结肠镜检查出息肉病例数/同期所有结 肠镜检查总数×100％。The artificial intelligence-based digestive endoscopy quality assessment system according to claim 1, wherein the detection rate of colorectal polyps is specifically defined as: patients with at least one colorectal polyp detected in colonoscopy The ratio of the number to the total number of colonoscopy in the same period, the calculation formula is: colorectal polyp detection rate = number of polyp cases detected by colonoscopy/total number of all colonoscopy in the same period × 100%.
  7. 如权利要求1所述的一种基于人工智能的消化内镜检查质量评估系统，其特征在于：结直肠腺瘤检出率的具体定义为：肠镜检查中至少检出一枚结直肠腺瘤的患者数占同期结肠镜检查总数的比例，计算公式为：结直肠腺瘤检出率＝结肠镜检查出腺瘤病例数/同期所有结肠镜检查总数×100％。The artificial intelligence-based digestive endoscopy quality assessment system according to claim 1, wherein the detection rate of colorectal adenoma is specifically defined as: at least one colorectal adenoma is detected in colonoscopy The ratio of the number of patients to the total number of colonoscopy during the same period, the calculation formula is: colorectal adenoma detection rate = number of colonoscopy adenoma cases / total number of colonoscopy during the same period × 100%.
  8. 如权利要求1所述的一种基于人工智能的消化内镜检查质量评估系统，其特征在于：肠道准备成功率的具体定义为：肠腔内有少量或者无粪渣且不影响肠镜观察的患者数占同期结肠镜检查总数的比例，计算公式为：肠道准备成功率＝1-肠腔内包含不合格图像的案例数量/同期结肠镜检查总数×100％。The artificial intelligence-based digestive endoscopy quality assessment system according to claim 1, wherein the specific definition of the success rate of intestinal preparation is: there is little or no fecal residue in the intestinal cavity and does not affect the colonoscopy observation The ratio of the number of patients to the total number of colonoscopy in the same period, the calculation formula is: bowel preparation success rate = 1-the number of cases with unqualified images in the intestinal cavity / total number of colonoscopy in the same period × 100%.
  9. 如权利要求1所述的一种基于人工智能的消化内镜检查质量评估系统，其特征在于：消化道早癌发现率的具体定义为：胃肠镜检查发现早期食管癌、胃癌或结直肠癌的患者数分别占同期食管癌、胃癌或结直肠癌患者数的比例，计算公式为：消化道早癌发现率＝胃肠镜检查发现早癌的患者数/同期早癌和中晚期癌患者总数×100％。The artificial intelligence-based digestive endoscopy quality assessment system according to claim 1, wherein the specific definition of the detection rate of early gastrointestinal cancer is: gastrointestinal endoscopy detects early esophageal cancer, gastric cancer or colorectal cancer The number of patients accounted for the proportion of patients with esophageal cancer, gastric cancer or colorectal cancer in the same period. The calculation formula is: early gastrointestinal cancer detection rate = number of patients with early cancer detected by gastrointestinal endoscopy/total number of patients with early cancer and advanced cancer in the same period ×100%.
  10. 一种基于人工智能的消化内镜检查质量自动评估的方法，其特征在于，包括如下步骤：An artificial intelligence-based method for automatically evaluating the quality of digestive endoscopy, which is characterized in that it includes the following steps:
    S1，医生手动录入病人信息、采集内镜图像，通过影像报告系统自动记录病人信息、被采集的图片和相应的操作时间；S1: The doctor manually enters patient information, collects endoscopic images, and automatically records patient information, collected pictures, and corresponding operation time through the image reporting system;
    S2，通过数据交换协议获取病人信息、被采集的图片和相应的操作时间，其中文字资料输入质控数据统计软件，图片资料输入深度卷积神经网络模型；S2: Obtain patient information, collected pictures, and corresponding operation time through a data exchange protocol, in which text data is input to quality control data statistics software, and picture data is input to a deep convolutional neural network model;
    S3，深度卷积神经网络模型接收胃肠镜图像，对图像进行部位特征和病灶特征识别，将预测结果输入质控数据统计软件；S3, the deep convolutional neural network model receives the gastrointestinal endoscopy image, recognizes the location and lesion features of the image, and inputs the prediction result into the quality control data statistics software;
    S4，质控数据统计软件根据内镜操作质量控制指标的计算公式，使用来自医学影像报告系统的文字资料和来自预训练模型对于图片的预测结果，自动计算内镜操作的多项质量控制指标；S4, the quality control data statistics software uses the text data from the medical imaging report system and the prediction results of the picture from the pre-training model according to the calculation formula of the endoscopic operation quality control index to automatically calculate multiple quality control indexes of the endoscopic operation;
    其中，所述质量控制指标包括，胃镜检查时间，胃镜检查盲区率，肠镜盲肠插管率，肠镜退镜时间，结直肠息肉检出率，结直肠腺瘤检出率，肠道准备成功率，消化道早癌发现率；Among them, the quality control indicators include, gastroscopy time, gastroscopy blind zone rate, colonoscopy cecal intubation rate, colonoscopy withdrawal time, colorectal polyp detection rate, colorectal adenoma detection rate, successful bowel preparation Rate, the detection rate of early gastrointestinal cancer;
    S5，质控数据统计软件通过图表等形式将统计结果输出，医生和管理者自行查看统计结果。S5, the quality control data statistical software outputs the statistical results in the form of charts and other forms, and the doctors and managers can view the statistical results by themselves.
PCT/CN2019/106029 2019-04-22 2019-09-16 Artificial intelligence-based automatic evaluation method and system on quality check of gastrointestinal endoscopy WO2020215593A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910325418.2A CN110097105A (en) 2019-04-22 2019-04-22 A kind of digestive endoscopy based on artificial intelligence is checked on the quality automatic evaluation method and system
CN201910325418.2 2019-04-22

Publications (1)

Publication Number Publication Date
WO2020215593A1 true WO2020215593A1 (en) 2020-10-29

Family

ID=67445576

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/106029 WO2020215593A1 (en) 2019-04-22 2019-09-16 Artificial intelligence-based automatic evaluation method and system on quality check of gastrointestinal endoscopy

Country Status (2)

Country Link
CN (1) CN110097105A (en)
WO (1) WO2020215593A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113327238A (en) * 2021-06-10 2021-08-31 紫东信息科技(苏州)有限公司 Gastroscope image classification model construction method and gastroscope image classification method
CN117198473A (en) * 2023-11-03 2023-12-08 首都医科大学宣武医院 Intestinal tract preparation quality judging system

Families Citing this family (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110097105A (en) * 2019-04-22 2019-08-06 上海珍灵医疗科技有限公司 A kind of digestive endoscopy based on artificial intelligence is checked on the quality automatic evaluation method and system
CN110473619B (en) * 2019-08-16 2022-05-27 电子科技大学 Bronchofiberscope intubation assistant decision-making system based on deep learning
CN110916606A (en) * 2019-11-15 2020-03-27 武汉楚精灵医疗科技有限公司 Real-time intestinal cleanliness scoring system and method based on artificial intelligence
CN113288007B (en) * 2019-12-06 2022-08-09 腾讯科技（深圳）有限公司 Endoscope moving time determining method and device and computer equipment
CN111000633B (en) * 2019-12-20 2020-11-03 山东大学齐鲁医院 Method and system for monitoring endoscope diagnosis and treatment operation process
CN111144271B (en) * 2019-12-23 2021-02-05 山东大学齐鲁医院 Method and system for automatically identifying biopsy parts and biopsy quantity under endoscope
CN110910992A (en) * 2019-12-25 2020-03-24 吉林大学 Automatic digestive endoscopy inspection quality evaluation system based on artificial intelligence
CN113689949A (en) * 2020-05-18 2021-11-23 日本电气株式会社 Information processing method, electronic device, and computer storage medium
CN112597981B (en) * 2021-03-04 2021-06-01 四川大学 Intelligent enteroscope withdrawal quality monitoring system and method based on deep neural network
CN113035329A (en) * 2021-03-22 2021-06-25 杭州联众医疗科技股份有限公司 Medical image quality control system
CN113240662B (en) * 2021-05-31 2022-05-31 萱闱(北京)生物科技有限公司 Endoscope inspection auxiliary system based on artificial intelligence
CN113052843B (en) * 2021-05-31 2021-09-28 萱闱(北京)生物科技有限公司 Method, apparatus, system, storage medium and computing device for assisting endoscopy
CN113763360A (en) * 2021-09-08 2021-12-07 山东大学 Digestive endoscopy simulator inspection quality assessment method and system
CN113706536B (en) * 2021-10-28 2022-01-18 武汉大学 Sliding mirror risk early warning method and device and computer readable storage medium
CN113962998A (en) * 2021-12-23 2022-01-21 天津御锦人工智能医疗科技有限公司 Method and device for evaluating effective endoscope withdrawal time of enteroscopy and storage medium
CN114359273B (en) * 2022-03-15 2022-06-21 武汉楚精灵医疗科技有限公司 Method and device for detecting abnormal digestive endoscopy video
CN114419521B (en) * 2022-03-28 2022-07-01 武汉楚精灵医疗科技有限公司 Method and device for monitoring intestinal endoscopy
CN115511885B (en) * 2022-11-16 2023-03-14 武汉楚精灵医疗科技有限公司 Method and device for detecting success rate of cecum intubation

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108665454A (en) * 2018-05-11 2018-10-16 复旦大学 A kind of endoscopic image intelligent classification and irregular lesion region detection method
CN109326343A (en) * 2017-10-25 2019-02-12 首都医科大学附属北京友谊医院 A kind of method, identification lesion method and the computer equipment of acquisition lesion data
CN109598716A (en) * 2018-12-05 2019-04-09 上海珍灵医疗科技有限公司 Colonoscopy based on computer vision moves back mirror speed method of real-time and system
CN109616195A (en) * 2018-11-28 2019-04-12 武汉大学人民医院（湖北省人民医院） The real-time assistant diagnosis system of mediastinum endoscopic ultrasonography image and method based on deep learning
CN110097105A (en) * 2019-04-22 2019-08-06 上海珍灵医疗科技有限公司 A kind of digestive endoscopy based on artificial intelligence is checked on the quality automatic evaluation method and system

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105718952B (en) * 2016-01-22 2018-10-30 武汉科恩斯医疗科技有限公司 The system that lesion classification is carried out to tomography medical image using deep learning network
CN107256552B (en) * 2017-06-14 2020-08-18 成都微识医疗设备有限公司 Polyp image recognition system and method
CN107887002A (en) * 2017-12-07 2018-04-06 卢乃吉 A kind of medical imaging diagnosis quality control system
CN108109681A (en) * 2017-12-21 2018-06-01 青岛美迪康数字工程有限公司 Digestive endoscopy structured report system and its method for building up
CN109544526B (en) * 2018-11-15 2022-04-26 首都医科大学附属北京友谊医院 Image recognition system, device and method for chronic atrophic gastritis

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109326343A (en) * 2017-10-25 2019-02-12 首都医科大学附属北京友谊医院 A kind of method, identification lesion method and the computer equipment of acquisition lesion data
CN108665454A (en) * 2018-05-11 2018-10-16 复旦大学 A kind of endoscopic image intelligent classification and irregular lesion region detection method
CN109616195A (en) * 2018-11-28 2019-04-12 武汉大学人民医院（湖北省人民医院） The real-time assistant diagnosis system of mediastinum endoscopic ultrasonography image and method based on deep learning
CN109598716A (en) * 2018-12-05 2019-04-09 上海珍灵医疗科技有限公司 Colonoscopy based on computer vision moves back mirror speed method of real-time and system
CN110097105A (en) * 2019-04-22 2019-08-06 上海珍灵医疗科技有限公司 A kind of digestive endoscopy based on artificial intelligence is checked on the quality automatic evaluation method and system

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113327238A (en) * 2021-06-10 2021-08-31 紫东信息科技(苏州)有限公司 Gastroscope image classification model construction method and gastroscope image classification method
CN117198473A (en) * 2023-11-03 2023-12-08 首都医科大学宣武医院 Intestinal tract preparation quality judging system
CN117198473B (en) * 2023-11-03 2024-02-23 首都医科大学宣武医院 Intestinal tract preparation quality judging system

Also Published As

Publication number Publication date
CN110097105A (en) 2019-08-06

Similar Documents

Publication Publication Date Title
WO2020215593A1 (en) Artificial intelligence-based automatic evaluation method and system on quality check of gastrointestinal endoscopy
Gong et al. Detection of colorectal adenomas with a real-time computer-aided system (ENDOANGEL): a randomised controlled study
EP4018909A1 (en) Deep learning-based diagnosis assistance system for early digestive tract cancer and examination apparatus
Xue et al. Modality alignment contrastive learning for severity assessment of COVID-19 from lung ultrasound and clinical information
WO2019088121A1 (en) Image diagnosis assistance apparatus, data collection method, image diagnosis assistance method, and image diagnosis assistance program
JPWO2018225448A1 (en) Diagnosis support method for diseases by endoscopic image of digestive organs, diagnosis support system, diagnosis support program, and computer-readable recording medium storing the diagnosis support program
CN110916606A (en) Real-time intestinal cleanliness scoring system and method based on artificial intelligence
Riegler et al. Eir—efficient computer aided diagnosis framework for gastrointestinal endoscopies
CN103975364B (en) Selection of images for optical examination of the cervix
Ali et al. Endoscopy disease detection challenge 2020
CN109447985A (en) Colonoscopic images analysis method, device and readable storage medium storing program for executing
Ghosh et al. Automatic bleeding detection in wireless capsule endoscopy based on RGB pixel intensity ratio
Keser et al. A deep learning algorithm for classification of oral lichen planus lesions from photographic images: A retrospective study
Xu et al. Upper gastrointestinal anatomy detection with multi‐task convolutional neural networks
Bejakovic et al. Analysis of Crohn's disease lesions in capsule endoscopy images
Huang et al. Evaluation of pulmonary edema using ultrasound imaging in patients with COVID-19 pneumonia based on a non-local Channel attention ResNet
Song et al. Use of artificial intelligence to improve the quality control of gastrointestinal endoscopy
Aliyi et al. Detection of gastrointestinal tract disorders using deep learning methods from colonoscopy images and videos
Wimpy et al. Classification of cervical type image using capsule networks
CN114359131A (en) Helicobacter pylori stomach video full-automatic intelligent analysis system and marking method thereof
Adelaja et al. Operating Artificial Intelligence to Assist Physicians Diagnose Medical Images: A Narrative Review
Mi et al. Diagnostic Accuracy of Wireless Capsule Endoscopy in Polyp Recognition Using Deep Learning: A Meta‐Analysis
Liu et al. A real-time system using deep learning to detect and track ureteral orifices during urinary endoscopy
Lu [Retracted] Image Aided Recognition of Wireless Capsule Endoscope Based on the Neural Network
Evans-Barns et al. Post-Operative Anorectal Manometry in Children following Anorectal Malformation Repair: A Systematic Review

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: 19925917

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: 19925917

Country of ref document: EP

Kind code of ref document: A1