Search papers, labs, and topics across Lattice.
This paper introduces a semantic-aware multimodal pre-training framework designed to enhance the representation of medical tabular data by addressing the limitations of existing methods that treat tabular inputs as flat vectors. By implementing Importance-Aware Adaptive Masking and a Soft-Label Discretized Module, the authors effectively prioritize salient features and stabilize the modeling of ordinal relationships, respectively. Extensive evaluations on large-scale dermatology and ophthalmology datasets reveal that this approach achieves state-of-the-art performance, showcasing significant robustness and generalizability across domains.
Medical tabular data representation can be revolutionized by a framework that prioritizes feature importance and stabilizes ordinal relationships, setting a new state-of-the-art in multimodal pre-training.
While vision-language models dominate medical representation learning, unstructured text lacks the dense, quantitative diagnostic phenotypes inherent in structured clinical tables. However, existing multimodal pre-training methods underutilize this potential due to semantic-agnostic designs that treat tabular inputs as flat vectors and employ unstable continuous regression objectives. To overcome this, we propose a novel semantic-aware framework explicitly modeling the intrinsic two-dimensional structure of tabular data. First, addressing the inter-feature hierarchy of varying diagnostic importance, we introduce Importance-Aware Adaptive Masking to construct a label-free curriculum prioritizing salient features. Second, addressing the intra-feature continuity-discreteness duality, we propose a Soft-Label Discretized Module that replaces unstable numerical regression with stable distribution matching, thereby mathematically preserving ordinal relationships. Extensive experiments across large-scale dermatology (SLICE-3D, HOP) and ophthalmology (EyePACS) datasets establish a new state-of-the-art (SOTA), demonstrating exceptional robustness and cross-domain generalizability.
