• 3D Deep-Learning Model: Develop a novel 3D deep-learning model based on residual networks (ResNet) to evaluate glaucomatous optic neuropathy comprehensively with a 3D view
• Clinical Copilot: Develop and validate the AI-assisted system to efficiently support large-scale screening and clinical evaluation based on optical coherence tomography (OCT) scans

• Multitask Deep Learning Algorithm: Use a 3D ResNet model to analyze OCT volumetric scans, distinguishing center-involved DME (CI-DME) and non-CI-DME for better triage
• Cross-Device Compatibility: Enables automated DME classification across multiple types of OCT devices

• Advanced Deep Learning Solution: Integrate 3D CNN, attention mechanisms, and multi-instance learning to address limitations on traditional ophthalmic imaging quality metrics
• Image Quality Control: Incorporate pre-diagnosis image assessment to improve clinical efficiency by filtering ungradable retinal photographs and OCT scans automatically

• Privacy-Preserving Deep Learning Models: Develop deep learning models based on federated learning frameworks to ensure data privacy and enhance multi-center collaboration without data exchanges
• Domain Adaptation Technique: Improve model adaptability across diverse datasets and enhance model generalizability using test-time batch normalization

• "Safety Net" Mechanism: Identify uncertain cases by dual probability thresholds to improve the reliability of deep learning models in ophthalmic disease diagnosis
• AI copilot with Expert in Decision-Making: Ensure sensitive or uncertain cases receive proper medical attention

Detection of glaucomatous optic neuropathy with spectral-domain optical coherence tomography: a retrospective training and validation deep-learning analysis

A Multitask Deep-Learning System to Classify Diabetic Macular Edema for Different Optical Coherence Tomography Devices: A Multicenter Analysis

Artificial intelligence deep learning algorithm for discriminating ungradable optical coherence tomography three-dimensional volumetric optic disc scans

Deep learning-based image quality assessment for optical coherence tomography macular scans: a multicentre study

Deep-Learning–Based Pre-Diagnosis Assessment Module for Retinal Photographs: A Multicenter Study

Advancing Diabetic Macular Edema Detection from 3D Optical Coherence Tomography Scans: Integrating Privacy-Preserving AI and Generalizability Techniques — A Prospective Validation in Vietnam

Developing a privacy-preserving deep learning model for glaucoma detection: a multicentre study with federated learning