MedVL-SAM2: A unified 3D medical vision-language model for multimodal reasoning and prompt-driven segmentation
MedVL-SAM2 is a unified 3D medical model combining image reasoning and pixel-level segmentation. It integrates a SAM2-based module with a vision-language model to perform tasks like report generation, VQA, and various types of segmentation. The model uses multi-modal prompting (language, point, box) to achieve superior performance in 3D medical imaging tasks.