| Project/Area Number |
17K00425
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Life / Health / Medical informatics
|
|---|
| Research Institution | Kitami Institute of Technology (2018-2019)
National Institute of Public Health (2017) |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2017-04-01 – 2020-03-31
|
| Project Status |
Completed (Fiscal Year 2019)
|
| Budget Amount *help |
¥4,550,000 (Direct Cost: ¥3,500,000、Indirect Cost: ¥1,050,000)
Fiscal Year 2019: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Fiscal Year 2018: ¥1,950,000 (Direct Cost: ¥1,500,000、Indirect Cost: ¥450,000)
Fiscal Year 2017: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
|
|---|
| Keywords | 医療用人工知能 / ユーザーインターフェース / 疾患類似度 / 診断支援システム |
|---|
| Outline of Final Research Achievements |
The Medical AI research has developed mainly focusing diagnostic algorithms. Accordingly, there has been a lack of research that addresses user interface issues that realize efficient input of patient information, effective output of diagnostic results, and methods that integrate multiple diagnostic algorithms. Our research project addressed the user interface of clinical decision support systems, and performed an exploratory study of user interfaces that facilitate physicians input and produce inspiring outputs, together with an investigation of efficient integration of the input and the output toward improved diagnostic performance. Our achievements include an estimation method of physicians' disease knowledge volume, two-dimensional representations of relationships between dieases, and contributions in medical natural language processing, which enable efficient interaction between a compueter and a physician.
|
| Academic Significance and Societal Importance of the Research Achievements |
診断支援用AI研究は、診断アルゴリズムの研究開発を中心に発展してきたことで、その入力や出力に求められる基礎的な知見の蓄積を欠きがちであった。我々の研究成果により、医師それぞれの知識において欠けている箇所を補うように機能する、医師と補完的な役割を担う医療用人工知能の研究開発が実現する。また、患者情報の入力に際した非効率は、この分野において30年以上も指摘されている問題であり、自然言語処理による解決に向けた貢献を果たすことができた。さらに、診断結果の出力と再計算においても、効率的な手法を提案することができた。
|
