IVCM-Insight: automated interactive interpretation of in vivo confocal microscopy.
Artificial intelligence system IVCM-Insight classified corneal microscopy findings with high accuracy.
IVCM-Insight: automated interactive interpretation of in vivo confocal microscopy.
In vivo confocal microscopy (IVCM) is a critical ophthalmic examination that provides in vivo cytological and neurological information essential for diagnosing corneal and certain systemic diseases, but its clinical utility is limited by time-consuming interpretation and the need for subspecialty expertise.
Based on 30,368 IVCM images and 4155 paired clinical reports, the model was trained with contrastive alignment, image-conditioned language modeling, and multi-image consistency loss to produce structured diagnostic reports while a domain-adapted large language models supported patient-centered question answering.
Automated evaluation showed strong agreement with the reference reports: Bilingual Evaluation Understudy (Bilingual Evaluation Understudy)-1 to Bilingual Evaluation Understudy-4 scores were 0.69, 0.58, 0.47, and 0.41, Recall-Oriented Understudy for Gisting Evaluation (Recall-Oriented Understudy for Gisting Evaluation) was 0.67, Consensus-based Image Description Evaluation (CIDEr) was 1.85, and Metric for Evaluation of Translation with Explicit Ordering (Metric for Evaluation of Translation with Explicit Ordering) was 0.66.
To our knowledge, IVCM-Insight is the first dedicated artificial intelligence system for comprehensive IVCM interpretation, with potential to enhance diagnostic efficiency, strengthen physician-patient communication, and broaden access to advanced corneal imaging across care settings.