Surgical-LVLM: Learning to Adapt Large Vision-Language Model for Grounded Visual Question Answering in Robotic Surgery
This work addresses the need for automated, context-aware solutions in personalized surgical mentorship, representing a domain-specific advancement.
The paper tackles the problem of complex visual question answering and region grounding in robotic surgery by introducing Surgical-LVLM, which adapts a large vision-language model with specialized modules, achieving new state-of-the-art performance on benchmarks like EndoVis-17-VQLA and EndoVis-18-VQLA.
Recent advancements in Surgical Visual Question Answering (Surgical-VQA) and related region grounding have shown great promise for robotic and medical applications, addressing the critical need for automated methods in personalized surgical mentorship. However, existing models primarily provide simple structured answers and struggle with complex scenarios due to their limited capability in recognizing long-range dependencies and aligning multimodal information. In this paper, we introduce Surgical-LVLM, a novel personalized large vision-language model tailored for complex surgical scenarios. Leveraging the pre-trained large vision-language model and specialized Visual Perception LoRA (VP-LoRA) blocks, our model excels in understanding complex visual-language tasks within surgical contexts. In addressing the visual grounding task, we propose the Token-Interaction (TIT) module, which strengthens the interaction between the grounding module and the language responses of the Large Visual Language Model (LVLM) after projecting them into the latent space. We demonstrate the effectiveness of Surgical-LVLM on several benchmarks, including EndoVis-17-VQLA, EndoVis-18-VQLA, and a newly introduced EndoVis Conversations dataset, which sets new performance standards. Our work contributes to advancing the field of automated surgical mentorship by providing a context-aware solution.