Spatial-aware Transformer-GRU Framework for Enhanced Glaucoma Diagnosis from 3D OCT Imaging

2024-03-08

arXiv.org

AI Summary

To address insufficient spatial structural modeling of 3D optical coherence tomography (OCT) volumes in early glaucoma detection, this paper proposes a spatially aware ViT-BiGRU collaborative framework. Specifically, a fine-tuned Vision Transformer (ViT) extracts intra-slice local semantic features, while a bidirectional gated recurrent unit (Bi-GRU) explicitly captures inter-slice spatial topological dependencies—enabling, for the first time, joint modeling of intra-layer anatomical details and inter-layer volumetric structure. This end-to-end approach overcomes the depth-agnostic limitation of conventional 2D models. Evaluated on a large-scale clinical dataset, the method achieves an F1-score of 93.58%, Matthews Correlation Coefficient (MCC) of 73.54%, and AUC of 95.24%, significantly outperforming existing state-of-the-art methods. The framework delivers a novel, interpretable, and robust paradigm for accurate early screening of glaucoma.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multi-instance/Multi-view Learning
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Glaucoma, a leading cause of irreversible blindness, necessitates early detection for accurate and timely intervention to prevent irreversible vision loss. In this study, we present a novel deep learning framework that leverages the diagnostic value of 3D Optical Coherence Tomography (OCT) imaging for automated glaucoma detection. In this framework, we integrate a pre-trained Vision Transformer on retinal data for rich slice-wise feature extraction and a bidirectional Gated Recurrent Unit for capturing inter-slice spatial dependencies. This dual-component approach enables comprehensive analysis of local nuances and global structural integrity, crucial for accurate glaucoma diagnosis. Experimental results on a large dataset demonstrate the superior performance of the proposed method over state-of-the-art ones, achieving an F1-score of 93.58%, Matthews Correlation Coefficient (MCC) of 73.54%, and AUC of 95.24%. The framework's ability to leverage the valuable information in 3D OCT data holds significant potential for enhancing clinical decision support systems and improving patient outcomes in glaucoma management.

Problem

• Automated glaucoma detection from 3D OCT imaging
• Integrating Vision Transformer and GRU for spatial dependencies
• Enhancing clinical decision support for glaucoma management

Innovation

• Vision Transformer for slice-wise feature extraction
• Bidirectional GRU captures inter-slice spatial dependencies
• Dual-component analysis of local and global structures