| Project/Area Number |
18K18121
|
|---|
| Research Category |
Grant-in-Aid for Early-Career Scientists
|
| Allocation Type | Multi-year Fund |
|---|
| Review Section |
Basic Section 61040:Soft computing-related
|
|---|
| Research Institution | The University of Tokyo |
|---|
Principal Investigator |
Kiwaki Taichi 東京大学, 大学院情報理工学系研究科, 特任助教 (70786011)
|
|---|
| Project Period (FY) |
2018-04-01 – 2022-03-31
|
| Project Status |
Completed (Fiscal Year 2021)
|
| Budget Amount *help |
¥4,160,000 (Direct Cost: ¥3,200,000、Indirect Cost: ¥960,000)
Fiscal Year 2020: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2019: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000)
Fiscal Year 2018: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
|
|---|
| Keywords | 機械学習 / 眼科学 / ニューラルネットワーク / 緑内障 / マルチタスク学習 / 深層ニューラルネットワーク / 医療画像分析 / 医療データマイニング / バッチ正規化 / PCA |
|---|
| Outline of Final Research Achievements |
In this research project, we improved training of neural networks, which are commonly used machine learning models, for ophthalmology dataset on glaucoma, which is one of the most serious eye disease. In this application field, we need to (1) tackle a problem of limited availability of data, and (2) explain the result in a medical perspective. Upon these points, this project developed a method to train a model so that it can attain meaningful knowledge from small sized datasets, explained the behavior of a trained model from a medical point of view, and verified the effectiveness of the developed methods in a form which can be accepted by medical communities.
|
| Academic Significance and Societal Importance of the Research Achievements |
緑内障は失明の可能性もある進行性の眼病であるが、治療によって進行を遅らせることが出来るため、その進行予測ならびに症状診断は非常に重要である。しかし症状診断と進行予測の両方に不可欠とされて来た眼の視野感度試験は、計測コストが高くまた計測毎のばらつきから来る信頼性にも問題がある。本研究で開発した手法は機械学習手法を利用してこの問題に対処するものであり、社会的な意義は非常に高い。
|
