A Multimodal Biomedical Foundation Model Trained from Fifteen Million Image–Text PairsSheng Zhang, Hoifung Poon, Naoto Usuyama et al.|NEJM AI|2024Cited by 218