2022 Fiscal Year Final Research Report
AI diagnostic imaging for prognosis and chemosensitivity of pancreatic cancer
Project/Area Number |
21K20919
|
Research Category |
Grant-in-Aid for Research Activity Start-up
|
Allocation Type | Multi-year Fund |
Review Section |
0903:Organ-based internal medicine and related fields
|
Research Institution | Kobe University |
Principal Investigator |
|
Project Period (FY) |
2021-08-30 – 2023-03-31
|
Keywords | 膵癌 / 線維化 / AI画像診断技術 |
Outline of Final Research Achievements |
Pancreatic cancer is the most aggressive in all cancer types; however, recent reports have indicated a correlation between the fibrosis within the tumor and the prognosis. The objective of this study is to develop an imaging diagnostic method using AI to predict the extent of fibrosis within the tumor and contribute to the selection of chemotherapy treatments. In an examination using resected pancreatic cancer specimens, it was revealed that the group with a high degree of fibrosis had significantly better prognosis, and they showed high sensitivity to postoperative chemotherapy. Next, an AI diagnostic system was first developed to automatically identify pancreatic cancer. Tumor detection AI was created to detect pancreatic tumors, along with pancreatic duct extension detection to identify indirect signs of pancreatic cancer, and pancreatic atrophy detection AI, which achieved good sensitivity and specificity. The future plan is to focus on developing fibrosis prediction AI.
|
Free Research Field |
消化器内科学分野
|
Academic Significance and Societal Importance of the Research Achievements |
膵癌は未だ予後が悪く、5年生存率は10%に満たない。予後改善のためには治療前に化学療法の感受性を予測することが重要であるが、未だ予測する手段は存在しない。本研究により、腫瘍内部の線維化を測定することは予後予測に有用であり、術後化学療法感受性予測マーカーとして有用である可能性が示唆された。また、線維化予測AIができれば手術例のみならず非手術例においても化学療法感受性予測、予後予測が可能となり、膵癌診療に大いに役立つことが期待される。
|