2023 Fiscal Year Final Research Report
Development of an endoscopic diagnosis system for colorectal cancer using artificial intelligence
| Project/Area Number |
21K07885
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Review Section |
Basic Section 53010:Gastroenterology-related
|
|---|
| Research Institution | Chiba University |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
千葉 哲博 千葉大学, 大学院医学研究院, 講師 (00381583)
徳長 鎮 千葉大学, 大学院医学研究院, 特任助教 (30866992)
沖元 謙一郎 千葉大学, 医学部附属病院, 助教 (30770739)
藤江 舞 千葉大学, 医学部附属病院, 臨床工学技士 (90899337)
野村 行弘 千葉大学, フロンティア医工学センター, 准教授 (60436491)
|
|---|
| Project Period (FY) |
2021-04-01 – 2024-03-31
|
| Keywords | 大腸癌 / AI |
|---|
| Outline of Final Research Achievements |
Endoscopic detection and resection of early-stage colorectal cancer is associated with reduced mortality from colorectal cancer. There is almost no risk of lymph node metastasis if the invasion of colorectal cancer is confined to the submucosa. Therefore, endoscopic treatment is indicated. However, the diagnosis of colorectal cancer depth is based on the judgment made at the time of endoscopy, and there are many cases of misdiagnosis.
In this study, we developed a computer-assisted diagnosis (CAD) system that uses artificial intelligence (AI) to diagnose the depth of colorectal cancer. This CAD system is useful for improving the diagnostic accuracy of non-expert endoscopists, and in cases diagnosed with high confidence, it was shown to be equivalent to the accuracy of expert endoscopists, suggesting that it could be a useful device in actual clinical practice.
|
| Free Research Field |
消化器内科
|
| Academic Significance and Societal Importance of the Research Achievements |
本邦における大腸癌死亡率は増加している。今回我々は、AIを用いた大腸癌の発見と、治療方針決定に影響する深達度診断をサポートするコンピュータ診断支援システムを開発し、その有用性を実証した。このシステムが臨床に使用可能となれば、大腸癌患者に対する最適な治療方針の提供に貢献するばかりか、過度の侵襲を伴う治療の回避にもつながると考えられ、よって社会的意義は高いと考えられる。
|
