1. Identification of Sex and Age from Macular Optical Coherence Tomography and Feature Analysis Using Deep Learning
- Author
-
Homer H. Chen, Yi-Ting Hsieh, I-Hsin Ma, Kuan-Ming Chueh, and Sheng-Lung Huang
- Subjects
medicine.medical_specialty ,genetic structures ,Fundus Oculi ,Age prediction ,Deep Learning ,Optical coherence tomography ,Foveal ,Ophthalmology ,medicine ,Humans ,Macula Lutea ,Child ,Retina ,medicine.diagnostic_test ,business.industry ,Deep learning ,Fundus photography ,eye diseases ,Cross-Sectional Studies ,medicine.anatomical_structure ,Child, Preschool ,Population study ,sense organs ,Artificial intelligence ,Choroid ,business ,Tomography, Optical Coherence - Abstract
PURPOSE To develop deep learning models for identification of sex and age from macular optical coherence tomography (OCT), and to analyze the features for differentiation of sex and age. DESIGN Algorithm development using database of macular OCT. SETTING One eye center in Taiwan. STUDY POPULATION 6147 sets of macular optical coherence tomography (OCT) images from the healthy eyes of 3134 persons. MAIN OUTCOME MEASURES Deep learning based algorithms were used to develop models for identification of sex and age, and 10-fold cross-validation was applied. Gradient-weighted class activation mapping (Grad-CAM) was used for feature analysis. RESULTS The accuracy for sex prediction using deep learning from macular OCT was 85.6±2.1%, compared to the accuracy of 61.9% by using macular thickness and 61.4±4.0% by using deep learning from infrared fundus photography (P
- Published
- 2022
- Full Text
- View/download PDF