BioengineeringbioRxivPreprint
Compact Vision--Language Models Enable Efficient and Interpretable Automated OCT Analysis Through Layer Specific Multimodal Learning
Tania Haghighi, Sina Gholami, Jared T. Sokol…·Minhaj Nur Alam
This paper introduces LO-VLM, a compact vision-language model for OCT analysis that improves accuracy and interpretability in retinal disease classification and summary generation.