Robustness

ROBUST GLAUCOMA DETECTION AT MULtIPLE LOCATIONS & AMD Detection from multiple imaging modalities

We developed end-to-end deep learning models which enable robust glaucoma detection from data collected at multiple locations by harnessing the power of natural-image pre-trained neural networks followed by fine-tuning on optical coherence tomography images.  

We achieved state-of-the-art 3-class Age-Related Macular Degeneration (AMD) detection from multiple imaging modalities (OCT, OCTA, high-definition 5-line 2D b-scans, and low-resolution 2D b-scans). We also achieved interpretability via Grad-CAMs and via comparing odds ratios of features of importance both for AI and for human experts. Variation in feature rank will help guide us as to how to improve AI and may help elucidate novel clinical features for accurate AMD detection.

ROBUST GLAUCOMA DETECTION AT MULIPLE LOCATIONS

Glaucoma Progression Detection and 

Visual Field Prediction 

We have developed two vision transformer (ViT)-based networks to detect glaucoma progression and predict future visual field appearance. We are now leveraging prior knowledge from OCT images to enhance the performance of these models using multimodal information (VF + OCT).

Glaucoma Progression Detection and Humphrey Visual Field Prediction Using Discriminative and Generative Vision Transformers

Read more here: 

Tian, Y., Zang, M., Sharma, A., Gu, S., Leshno, A., Thakoor, K.A. “Glaucoma Progression Detection and Humphrey Visual Field Prediction Using Discriminative and Generative Vision Transformers.” OMIA-X Workshop in conjunction with MICCAI 2023, October 2023.