2023 Fiscal Year Final Research Report
Development of Endoscopic Diagnosis of Ulcerative Colitis-Associated Neoplasia Using Artificial Intelligence
Project/Area Number |
20K17002
|
Research Category |
Grant-in-Aid for Early-Career Scientists
|
Allocation Type | Multi-year Fund |
Review Section |
Basic Section 53010:Gastroenterology-related
|
Research Institution | Showa University |
Principal Investigator |
|
Project Period (FY) |
2020-04-01 – 2024-03-31
|
Keywords | 潰瘍性大腸炎 / 大腸癌 / 大腸内視鏡 / 人工知能 / 光学生検 |
Outline of Final Research Achievements |
The specific goals of this study were to (1) annotate images for machine learning and (2) evaluate the constructed artificial intelligence. We focused on developing an ultra-expanded endoscopic diagnostic method for differentiating between tumor and non-tumor in ulcerative colitis-related tumors, which underpins the annotation of images for machine learning. The results were summarized in an English paper and accepted for publication. Kudo SE, Maeda Y, et al. developed a prototype by annotating images for machine learning based on this ultra-expanded endoscopy diagnostic method. We also evaluated the utility of the currently available CADe for monitoring patients with ulcerative colitis, aiming to develop software to assist in lesion detection. This work was reported by Maeda Y, Kudo SE, et al. Endoscopy. 2021. Additionally, we identified the challenge of misdiagnosing highly inflamed mucosa and inflammatory polyps as neoplastic lesions (Maeda, et al., UEGW 2022).
|
Free Research Field |
消化器内視鏡学
|
Academic Significance and Societal Importance of the Research Achievements |
本研究は人工知能(AI)を利用することで、潰瘍性大腸炎(UC)関連腫瘍の内視鏡診断の確立を目標とした。UCは、本邦で22万人の罹患者がいる疾患である。UC患者では年3%と高率でUC関連腫瘍が発生する。UC関連腫瘍を早期に正確に診断し治療することで大腸癌の抑制や大腸全摘術の回避が可能である。しかしながら、周囲粘膜との境界が不明瞭であり、既存の内視鏡モダリティでは十分な精度での診断が困難であった。今後、本研究で開発したシステムが実用化を目指す。
|