2023 Fiscal Year Final Research Report
Sex and disease judgment using color fundus photography
| Project/Area Number |
21K09704
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Review Section |
Basic Section 56060:Ophthalmology-related
|
|---|
| Research Institution | Kagoshima University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2021-04-01 – 2024-03-31
|
| Keywords | 眼球の個人差 / 眼底写真の個人差 / 眼底写真による推定 / 眼底の性差 / 眼底の加齢変化 |
|---|
| Outline of Final Research Achievements |
Similar to face photographs, fundus photographs have large individual differences to the extent that individuals can be identified. In other words, there are characteristics hidden in the fundus of the eye that can identify an individual. Artificial intelligence was able to predict the gender with 97% accuracy and the age with an error of 3.26 years just from the fundus photo. However, artificial intelligence cannot determine the specific characteristics of the fundus. Therefore, we quantified various features of fundus photographs and conducted statistical analysis. As a result, we were able to estimate the gender at 80.4%, and the specific gender difference was optic disc shape, angle of reinal vessels, and the green color tone. Age could also be estimated with an error of 3.64 years, and color tone changes with age. These results will lead to the elucidation of the pathology of eye diseases that occur due to gender differences and aging.
|
| Free Research Field |
眼球の個人差
|
| Academic Significance and Societal Importance of the Research Achievements |
眼底写真だけから年齢や性別がある程度推定できることが証明された。性差や加齢変化が影響する加齢黄斑変性や緑内障、黄斑円孔などの眼疾患のリスクが眼底写真から推定できる可能性が示唆される。眼疾患の病態解明に本研究の結果が役立つことに加え、健診で痛み無く取得できる眼底写真から眼疾患や性格傾向などが推定できれば国民の健康維持にも役立つ。実年齢と推定年齢の差から眼底が実年齢よりも加齢変化が進んでいるかどうかも分かる。今後は眼底写真に限らず生体から得られるひとつの画像から様々な要因を推定することが可能となるため、社会的な意義は大きいが、受け入れる側の十分な理解も必要である。
|
