Multi-task deep learning for predicting metabolic syndrome from retinal fundus images in a Japanese health checkup dataset

Abstract

Background Retinal fundus images provide a noninvasive window into systemic health, offering opportunities for early detection of metabolic disorders such as metabolic syndrome (METS). Objective This study aimed to develop a deep learning model to predict METS from fundus images obtained during routine health checkups, leveraging a multi-task learning approach. Methods We retrospectively analyzed 5,000 fundus images from Japanese health checkup participants. Convolutional neural network (CNN) models were trained to classify METS status, incorporating fundus-specific data augmentation strategies and auxiliary regression tasks targeting clinical parameters such as abdominal circumference (AC). Model performance was evaluated using validation accuracy, test accuracy, and the area under the receiver operating characteristic curve (AUC). Results Models employing fundus-specific augmentation demonstrated more stable convergence and superior validation accuracy compared to general-purpose augmentation. Incorporating AC as an auxiliary task further enhanced performance across architectures. The final ensemble model with test-time augmentation achieved a test accuracy of 0.696 and an AUC of 0.73178. Conclusion Combining multi-task learning, fundus-specific data augmentation, and ensemble prediction substantially improves deep learning-based METS classification from fundus images. This approach may offer a practical, noninvasive screening tool for metabolic syndrome in general health checkup settings.