Search papers, labs, and topics across Lattice.
This paper introduces UniCon, a novel open-linguistic unified concept learning framework designed to enhance cross-site interpretability in dermatology image diagnosis. By creating a shared semantic representation space and employing open-linguistic specifications, UniCon effectively navigates the challenges posed by heterogeneous concept taxonomies across diverse modalities without the need for dataset-specific retraining. The extensive experiments reveal that UniCon not only achieves superior diagnostic accuracy but also facilitates unprecedented cross-site intervention capabilities, significantly improving clinical decision-making processes.
UniCon bridges disparate clinical taxonomies, enabling seamless cross-site interventions in dermatology diagnosis without costly retraining.
Human-interpretable computer-aided diagnosis is crucial for clinical decision making. Concept-based models excel by providing transparent reasoning and enabling post-hoc, clinician-in-the-loop interventions. However, their rigid dataset-specific adaptation inherently restricts cross-site generalization. Applying them across diverse modalities, such as dermoscopic and clinical photographs, is challenging due to heterogeneous concept taxonomies varying in availability, granularity, and semantics across cohorts. Consequently, adapting Foundation Vision-Language Models (FVLMs) demands costly label engineering and repeated post-training. Existing intervention mechanisms remain rigidly tied to predefined concepts, lacking adaptability and hindering scalable dermatology CAD deployment. To address these bottlenecks, we propose UniCon, an open-linguistic unified concept learning framework for multimodal interpretable vision-language diagnosis. UniCon resolves these challenges through three contributions: (1) A shared semantic representation space via a unified concept prototype codebook, seamlessly coordinating heterogeneous concept systems across modalities without dataset-specific retraining. (2) Open-linguistic based multi-faceted semantic specifications to overcome sparse textual label limitations, improving boundary sensitivity in uncertain clinical contexts. (3) A robust, cross-site adjustable intervention interface powered by reliability-gated bottleneck aggregation, enabling consistent reasoning and transferable clinician corrections. Extensive experiments demonstrate that beyond securing top-tier diagnostic accuracy, UniCon successfully bridges disparate clinical taxonomies, unlocking unprecedented cross-site intervention capabilities. Code is available at https://github.com/wuchengyu123/UniCon.
