MedCoDi-M: A Multi-Prompt Foundation Model for Multimodal Medical Data Generation
Research Unit of Computer Systems and Bioinformatics, Department of Engineering, Università Campus Bio-Medico di Roma, Roma, Italy.
Department of Diagnostics and Intervention, Radiation Physics, Biomedical Engineering, Umeå University, Umeå, Sweden.
Department of Diagnostic Imaging and Stereotactic Radiosurgery, Centro Diagnostico Italiano S.p.A., Milano, Italy.
Department of Radiology and Interventional Radiology, Fondazione Policlinico Universitario Campus Bio-Medico, Rome, Italy.
Research Unit of Radiology and Interventional Radiology, Department of Medicine and Surgery, Università Campus Bio-Medico di Roma, Rome, Italy.
College of Computer Science and Software Engineering, Shenzhen University, Shenzhen, China.
Abstract
MedCoDi-M is an innovative multimodal framework for medical data generation. By leveraging contrastive learning and modular training, it facilitates any-to-any generation across medical data modalities, effectively tackling challenges related to data scarcity, privacy preservation, and multimodal integration.