徳島大学 教育・研究者情報データベース(EDB)

1. (英) Sogawa Takahiro

   役割	任意
   貢献度	任意
   学籍番号	推奨

2. (英) Tabuchi Hitoshi

   役割	任意
   貢献度	任意
   学籍番号	推奨

3. (英) Nagasato Daisuke

   役割	任意
   貢献度	任意
   学籍番号	推奨

4. (英) Masumoto Hiroki

   役割	任意
   貢献度	任意
   学籍番号	推奨

5. (英) Ikuno Yasushi

   役割	任意
   貢献度	任意
   学籍番号	推奨

6. (英) Ohsugi Hideharu

   役割	任意
   貢献度	任意
   学籍番号	推奨

7. (英) Ishitobi Naofumi

   役割	任意
   貢献度	任意
   学籍番号	推奨

8. 三田村 佳典([徳島大学.大学院医歯薬学研究部.医学域.医科学部門.外科系.眼科学])

   役割	任意
   貢献度	任意
   学籍番号	推奨

(英) Accuracy of a deep convolutional neural network in the detection of myopic macular diseases using swept-source optical coherence tomography.

(英) This study examined and compared outcomes of deep learning (DL) in identifying swept-source optical coherence tomography (OCT) images without myopic macular lesions [i.e., no high myopia (nHM) vs. high myopia (HM)], and OCT images with myopic macular lesions [e.g., myopic choroidal neovascularization (mCNV) and retinoschisis (RS)]. A total of 910 SS-OCT images were included in the study as follows and analyzed by k-fold cross-validation (k = 5) using DL's renowned model, Visual Geometry Group-16: nHM, 146 images; HM, 531 images; mCNV, 122 images; and RS, 111 images (n = 910). The binary classification of OCT images with or without myopic macular lesions; the binary classification of HM images and images with myopic macular lesions (i.e., mCNV and RS images); and the ternary classification of HM, mCNV, and RS images were examined. Additionally, sensitivity, specificity, and the area under the curve (AUC) for the binary classifications as well as the correct answer rate for ternary classification were examined. The classification results of OCT images with or without myopic macular lesions were as follows: AUC, 0.970; sensitivity, 90.6%; specificity, 94.2%. The classification results of HM images and images with myopic macular lesions were as follows: AUC, 1.000; sensitivity, 100.0%; specificity, 100.0%. The correct answer rate in the ternary classification of HM images, mCNV images, and RS images were as follows: HM images, 96.5%; mCNV images, 77.9%; and RS, 67.6% with mean, 88.9%.Using noninvasive, easy-to-obtain swept-source OCT images, the DL model was able to classify OCT images without myopic macular lesions and OCT images with myopic macular lesions such as mCNV and RS with high accuracy. The study results suggest the possibility of conducting highly accurate screening of ocular diseases using artificial intelligence, which may improve the prevention of blindness and reduce workloads for ophthalmologists.

1. (英) Adult
2. (英) Aged
3. (英) Blindness
4. (英) Choroid
5. (英) Choroidal Neovascularization
6. (英) Datasets as Topic
7. (英) Deep Learning
8. (英) Diagnosis, Differential
9. (英) Female
10. (英) Humans
11. (英) Image Interpretation, Computer-Assisted
12. (英) Macula Lutea
13. (英) Male
14. (英) Mass Screening
15. (英) Middle Aged
16. (英) Myopia
17. (英) ROC Curve
18. (英) Retinoschisis
19. (英) Severity of Illness Index
20. (英) Tomography, Optical Coherence

1. (英) PublicationType: Evaluation Study

2. (英) PublicationType: Journal Article