FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3207126487> ?p ?o ?g. }

• W3207126487 endingPage "135" @default.
• W3207126487 startingPage "132" @default.
• W3207126487 abstract "This paper outlines what the Japan Gastroenterological Endoscopy Society (JGES) has done in terms of regulatory science for the development and spreading of gastrointestinal endoscopy and related medical devices. Since 2015, JGES had established an industry–government–academia social gathering and held regular meetings for the purpose of promoting research and development and commercialization of gastrointestinal endoscopes and related devices. At this social gathering, industry, regulatory agencies, and academia were able to understand their respective positions and share their problems. This became one of the contributions to the development of gastrointestinal endoscopic medicine. This social gathering completed in 2018, and then JGES focused on the issues of endoscopic device development using artificial intelligence (AI). In recent years, social implementation of AI technology has progressed, AI research has been activated in medical care, and many JGES members have participated in such research. By 2018 more than 13 Japanese companies were developing AI software to support endoscopic diagnosis. Utilization of AI in the medical field is highly expected, but on the other hand, there have been the following problems: consent and personal information protection in collecting medical information, evaluation of device performance, quality control, compensation for misjudgment. Therefore, JGES established the “AI Promotion Review Committee” in October 2018, to provide certain guidelines for the development of AI endoscopes and related devices and to promote appropriate development, so that they would not be distributed in a disorderly manner. Immediately after the establishment of committee, discussions were held with the advisors from Pharmaceuticals and Medical Devices Agency (PMDA) and the Ministry of Health, Labour and Welfare, and a notification was issued to members on June 1, 2019. This notice provided regulatory information. The accuracy of endoscopic diagnosis depends on the skill of the endoscopist, and there is difference in diagnostic ability between expert endoscopists and non-expert endoscopists. There are two differences. One is the difference in lesion identification. The other is the difference in the qualitative diagnosis of the lesion, such as the differentiation of benign and malignant tumors. By using computer-aided diagnosis/detection (CAD) which has learned the answers of experts using AI technology, it is expected that the diagnostic ability of non-experts will be improved. In addition, experts can be also expected to prevent oversight of lesions and assist in qualitative diagnosis. These will lead to the provision of appropriate medical care and contribution to patients, and can be expected to be effective in the medical economy. As a project commissioned by the Japan Agency for Medical Research and Development (AMED), JGES collected images including diagnostic data and built a database in cooperation with the National Institute of Informatics for 2 years from 2017 (https://www.jges.net/news/2018/05/18/33064). From 2019, the JGES performed the following research, “Development of information and communications technology infrastructure linked to integrated database for prototype AI device development in the field of gastrointestinal endoscopy” and “Formulation of guidelines for empirical research for public cloud of AI device development infrastructure and its problem extraction” (https://www.jges.net/medical/newslist/clinical_research). The following research derived from above projects are still ongoing: “Research on AI-CAD accuracy improvement of gastric cancer”, “Disease classification by AI-CAD for inflammatory bowel disease”, “Morphological classification of duodenal papilla by AI-CAD” and “AI-CAD to prevent oversight”. In addition, for the purpose of accelerating the development of AI-CAD, JGES is working on the automation of structuring and parameterizing pathological diagnosis description by AI language analysis. Furthermore, for the purpose of continuous development of AI research and out-licensing to companies, “Research on ethical aspects for data collection and data utilization” and “Research on datasets for AI device development and AI medical device evaluation” have also started. In the gastrointestinal field, on December 6, 2018, colonoscopic computer-aided diagnosis (CADx) software “EndoBRAIN” co-developed by Professor Shinei Kudo's group of Showa University Northern Yokohama Hospital, Professor Kensaku Mori of Nagoya University Informatics, and Cybernet Systems Co Ltd (Tokyo, Japan) was approved and marketed by Olympus Co., Ltd. (Tokyo, Japan) for the first time in the world. “EndoBRAIN” was developed by learning about 69,000 endoscopic images based on the AI machine learning method (support vector machine). It has been shown to distinguish between neoplastic and non-neoplastic colorectal polyps with 96.9% sensitivity, 100% specificity, and 98% accuracy. These values were all significantly higher than those of the endoscopy trainees and was comparable to the accuracy rate of specialists.1 Since then, they released computer-aided detection (CADe) software “EndoBRAIN-EYE” and CADx software which added diagnostic capabilities for invasive cancer to EndoBRAIN “EndoBRAIN-Plus”. In October 2020, “CAD EYE” developed by Fujifilm, which combines CADe and CADx, was approved. Subsequently, in November 2020, the CADx “WISE VISION Endoscopy” co-developed by the National Cancer Center and NEC Corporation was approved. All of these are intended for colorectal tumors. So far, AI-CAD for inflammatory bowel disease, gastric lesions, and biliary-pancreatic lesions has not been approved in Japan. However, the “WISE VISION Endoscopy” has developed a technology to detect tumors in Barrett's esophagus and has obtained the CE mark, which is the European safety standard, for the first time in the world. For gastric cancer, Tada et al. are actively working on the development of AI-CAD.2, 3 In addition, research and development of AI-CAD for not only diagnosis of gastrointestinal tract by endoscopy and also diagnosis of pancreas cancer by endoscopic ultrasound are also underway by Japanese endoscopists.4 AI technology is becoming more and more common in our daily lives. Incorporating these technologies into medical care is natural and necessary. Also, it is expected to contribute to the progress of medical care. In the field of gastrointestinal endoscopy, AI-CAD was first put into practical use as a diagnostic aid for neoplastic diseases of the colon. In the near future, AI-CAD will be widespread, not only in neoplastic lesion but also in inflammatory disease, and not only for colon but also for esophagus, stomach, duodenum, small intestine, and pancreatobiliary disease.5-7 The evolution of these AI devices requires a large amount of accurately annotated teacher data. And how quickly and in large quantities these teacher data can be collected will determine the speed of development. To do so, it is necessary to build a data bank that complies with the regulations regarding the protection of personal information. It is also important to recognize regulatory approval from the early stages of development. In order to be approved by the regulatory affairs, it is necessary to prove the effectiveness and safety of AI-CAD devices as scientific evidence. Approval is the first step in the spread of AI-CAD. The next step is to list up insurance. If the use of AI-CAD is covered by insurance, it will be adopted by many hospitals and will become widespread in the medical field. In order to be covered by insurance, it is necessary to present the impact of AI-CAD on the medical economy as scientific evidence. It is convinced that the widespread use of AI-CAD in the medical field will lead to the best medical care for patients and the improvement of quality of life in the endoscopist's workplace. However, if the quality of those AI-CAD is not guaranteed, it will cause confusion in the medical field. JGES is a group of experts in the field of gastrointestinal endoscopy. All of us JGES members will need to build an annotated image database that anyone can use freely in order to develop a better AI-CAD. Also, it is also expected to build a data infrastructure that can prove the economic benefits of using AI-CAD. Author declares no conflict of interest for this article. None." @default.
• W3207126487 created "2021-10-25" @default.
• W3207126487 creator A5086844751 @default.
• W3207126487 date "2021-10-15" @default.
• W3207126487 modified "2023-09-23" @default.
• W3207126487 title "Contribution of the Japan Gastroenterological Endoscopy Society to promote computer‐aided diagnosis/detection system development using artificial intelligence technology" @default.
• W3207126487 cites W2903886348 @default.
• W3207126487 cites W2909293781 @default.
• W3207126487 cites W2973140425 @default.
• W3207126487 cites W3024896321 @default.
• W3207126487 cites W3042541362 @default.
• W3207126487 cites W3085806799 @default.
• W3207126487 cites W3087297434 @default.
• W3207126487 doi "https://doi.org/10.1111/den.14146" @default.
• W3207126487 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/34652003" @default.
• W3207126487 hasPublicationYear "2021" @default.
• W3207126487 type Work @default.
• W3207126487 sameAs 3207126487 @default.
• W3207126487 citedByCount "0" @default.
• W3207126487 crossrefType "journal-article" @default.
• W3207126487 hasAuthorship W3207126487A5086844751 @default.
• W3207126487 hasConcept C126838900 @default.
• W3207126487 hasConcept C154945302 @default.
• W3207126487 hasConcept C19527891 @default.
• W3207126487 hasConcept C2778451229 @default.
• W3207126487 hasConcept C41008148 @default.
• W3207126487 hasConcept C61434518 @default.
• W3207126487 hasConcept C71924100 @default.
• W3207126487 hasConceptScore W3207126487C126838900 @default.
• W3207126487 hasConceptScore W3207126487C154945302 @default.
• W3207126487 hasConceptScore W3207126487C19527891 @default.
• W3207126487 hasConceptScore W3207126487C2778451229 @default.
• W3207126487 hasConceptScore W3207126487C41008148 @default.
• W3207126487 hasConceptScore W3207126487C61434518 @default.
• W3207126487 hasConceptScore W3207126487C71924100 @default.
• W3207126487 hasIssue "S2" @default.
• W3207126487 hasLocation W32071264871 @default.
• W3207126487 hasLocation W32071264872 @default.
• W3207126487 hasOpenAccess W3207126487 @default.
• W3207126487 hasPrimaryLocation W32071264871 @default.
• W3207126487 hasRelatedWork W2313802994 @default.
• W3207126487 hasRelatedWork W2403726136 @default.
• W3207126487 hasRelatedWork W2404146888 @default.
• W3207126487 hasRelatedWork W2405232492 @default.
• W3207126487 hasRelatedWork W2413394245 @default.
• W3207126487 hasRelatedWork W2474093687 @default.
• W3207126487 hasRelatedWork W3030951041 @default.
• W3207126487 hasRelatedWork W3208701539 @default.
• W3207126487 hasRelatedWork W4313346385 @default.
• W3207126487 hasRelatedWork W4317816533 @default.
• W3207126487 hasVolume "34" @default.
• W3207126487 isParatext "false" @default.
• W3207126487 isRetracted "false" @default.
• W3207126487 magId "3207126487" @default.
• W3207126487 workType "article" @default.