Alzheimer's Disease Classification Using Retinal OCT: TransnetOCT and Swin Transformer Models

2025-03-14

AI Summary

This study addresses the challenge of noninvasive early classification of Alzheimer’s disease (AD) versus healthy controls (CO) using retinal optical coherence tomography (OCT) images. We propose TransNetOCT, a lightweight, task-specific network that uniquely integrates OCT image segmentation-based data augmentation with a deep feature disentanglement mechanism—marking the first such combination for small-sample medical imaging tasks. This design significantly enhances model robustness and generalization under limited-data conditions. Under five-fold cross-validation, TransNetOCT achieves 98.18% accuracy on raw OCT images and improves to 98.91% with segmentation-enhanced inputs, substantially outperforming Swin Transformer (93.54%). These results empirically validate retinal OCT as a clinically viable biomarker for AD and establish a novel paradigm for intelligent, OCT-based辅助 diagnosis of neurodegenerative disorders.

Technology Category

Computer Vision: SegmentationMachine Learning: Multi-instance/Multi-view LearningIntelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

Security and Privacy: Data transparency and provenanceResponsible Web: Algorithmic accountability and transparency on the webEconomics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Retinal optical coherence tomography (OCT) images are the biomarkers for neurodegenerative diseases, which are rising in prevalence. Early detection of Alzheimer's disease using retinal OCT is a primary challenging task. This work utilizes advanced deep learning techniques to classify retinal OCT images of subjects with Alzheimer's disease (AD) and healthy controls (CO). The goal is to enhance diagnostic capabilities through efficient image analysis. In the proposed model, Raw OCT images have been preprocessed with ImageJ and given to various deep-learning models to evaluate the accuracy. The best classification architecture is TransNetOCT, which has an average accuracy of 98.18% for input OCT images and 98.91% for segmented OCT images for five-fold cross-validation compared to other models, and the Swin Transformer model has achieved an accuracy of 93.54%. The evaluation accuracy metric demonstrated TransNetOCT and Swin transformer models capability to classify AD and CO subjects reliably, contributing to the potential for improved diagnostic processes in clinical settings.

Problem

• Early detection of Alzheimer's disease using retinal OCT images.
• Classification of Alzheimer's and healthy subjects via deep learning.
• Improving diagnostic accuracy with TransNetOCT and Swin Transformer models.

Innovation

• Utilizes TransNetOCT for high-accuracy Alzheimer's classification
• Employs Swin Transformer for reliable OCT image analysis
• Preprocesses retinal OCT images using ImageJ for enhanced diagnostics