Project/Area Number |
16K10266
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Multi-year Fund |
Section | 一般 |
Research Field |
Radiation science
|
Research Institution | Tohoku University |
Principal Investigator |
|
Co-Investigator(Kenkyū-buntansha) |
本間 経康 東北大学, 医学系研究科, 教授 (30282023)
森 菜緒子 東北大学, 大学病院, 助教 (90535064)
|
Project Period (FY) |
2016-04-01 – 2019-03-31
|
Project Status |
Completed (Fiscal Year 2018)
|
Budget Amount *help |
¥4,680,000 (Direct Cost: ¥3,600,000、Indirect Cost: ¥1,080,000)
Fiscal Year 2018: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2017: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
Fiscal Year 2016: ¥1,950,000 (Direct Cost: ¥1,500,000、Indirect Cost: ¥450,000)
|
Keywords | デジタルマンモグラフィ / 乳房画像診断 |
Outline of Final Research Achievements |
Mammographic breast cancer screening is a cost-effective way to improve survival. However, diagnostic accuracy greatly varies depending on experience of the doctor. CAD using AI technology is attracting attention as a diagnostic support method for doctors. We constructed a database of over 20,000 normal breast and cancer cases and succeeded in developing CAD using deep learning. We made a diagnostic workstation equipped with this software, and confirmed that the detection rate of calcified lesions and mass lesions was superior to existing CAD. At the same time, we developed a report management support software that can accurately measure breast tissue and calculate breast cancer risk factors from past medical history and family history, with a view to future personalized medicine.
|
Academic Significance and Societal Importance of the Research Achievements |
日本のマンモグラフィ検診では精度管理のために医師2名による読影を義務化している。医師の負担増、経費増などで日本では検診率が低く、目標に達していない。精度の悪い検診では要精査率を高めてしまい、医療機関での精密検査などの医療費負担増も問題となっている。そのためにも経験豊富な専門医と同等のCADの開発、普及が社会的ニーズとなっている。近年の深層学習法を用いたAICADに新たに期待されるようになってきた。我々が開発したCAD搭載の読影支援システムは、既存のCADより優れた感度、特異度を有し、これらの社会的ニーズの答えることができると思われる。
|