MerMED-FM showed excellent diagnostic accuracy for optical coherence tomography.
MerMED-FM: Multimodal, Multi-Disease Medical Imaging Foundation Model.
Yang Zhou … Daniel Shu Wei Ting
The Lancet. Digital health · 2027
Background
Current artificial intelligence (artificial intelligence) models for medical imaging predominantly focus on a single imaging modality and a single disease.
Purpose
We aimed to train and evaluate an artificial intelligence model that can interpret diverse imaging modalities across specialties while maintaining robust performance within each modality.
Methods
We developed Multimodal, Multi-Disease Medical Imaging Foundation Model (MerMED-FM), a multi-specialty model trained using self-supervised learning and a memory module.
n = around 3·3 million images
Results
0·962 for OCT
MerMED-FM was very accurate at diagnosing diseases using eye scans
n = around 3·3 million images
More results
MerMED-FM was trained on around 3·3 million images from 53 publicly available, unlabelled datasets, comprising 713931 chest x-rays, 292353 CT slices, 389885 ultrasound frames, 1017712 pathology patches, 333099 colour fundus photography images, 176719 optical coherence tomography slices, and 401059 dermatoscopy images.
“
Conclusion
MerMED-FM has the potential to be a highly adaptable, versatile, cross-specialty foundation model that enables robust interpretation of medical imaging across diverse medical disciplines.