CAT: Coordinating Anatomical-Textual Prompts for Multi-Organ and Tumor Segmentation

arXiv:2406.07085v1 Announce Type: new
Abstract: Existing promptable segmentation methods in the medical imaging field primarily consider either textual or visual prompts to segment relevant objects, yet they often fall short when addressing anomalies in medical images, like tumors, which may vary greatly in shape, size, and appearance. Recognizing the complexity of medical scenarios and the limitations of textual or visual prompts, we propose a novel dual-prompt schema that leverages the complementary strengths of visual and textual prompts for segmenting various …

abstract arxiv cat complexity cs.cv images imaging medical medical imaging objects prompts segment segmentation shape textual tumors type visual