TriPAH: Imbalance-Aware Tri-Prompt Affinity Hashing for Cross-Modal Medical Retrieval

2026-06-25

AI Summary

This work addresses semantic fragmentation in cross-modal medical retrieval caused by clinical text noise, long-tailed label distributions, and quantization vulnerability. To tackle these challenges, the authors propose TriPAH, a novel framework that integrates an ontology-guided tri-level prompting mechanism to generate low-noise textual representations and introduces a lightweight prompt-token mixer for multi-granularity alignment and progressive quantization. The approach further incorporates patient-level prompt synthesis, an asymmetric multi-task objective, multi-positive contrastive learning, and cross-view consistency constraints to mitigate data imbalance and enhance semantic coherence of hash codes. Extensive experiments on three public medical datasets demonstrate that TriPAH significantly outperforms state-of-the-art methods, achieving substantial improvements in both retrieval accuracy and robustness.

Abstract

In the era of big medical data, efficient cross-modal retrieval is pivotal for evidence-based diagnosis and large-scale case management. Cross-modal medical hashing retrieval aims to enable efficient image-text search and support downstream tasks such as case-based reasoning and decision support by learning compact, semantically aligned binary codes. However, current methods suffer from semantic fragmentation due to noisy clinical language, long-tailed labels, and brittle quantization that weakens alignment. We propose TriPAH, a Tri-Prompt Affinity Hashing framework. TriPAH synthesizes ontology-grounded, patient-level prompts conditioned on normalized clinical cues to yield low-noise textual representations for initial alignment. A lightweight prompt-token mixer performs hierarchical, multi-granularity alignment and produces quantization-ready features under an asymmetric multi-task objective coupling multi-positive contrastive alignment, imbalance-aware classification, and progressive quantization regularization. A patient-level consistency module further stabilizes codes across complementary views. Extensive experiments on three public datasets demonstrate that TriPAH significantly outperforms state-of-the-art methods.

Problem

• cross-modal retrieval
• semantic fragmentation
• medical hashing
• long-tailed labels
• quantization alignment

Innovation

• Tri-Prompt Affinity Hashing
• cross-modal medical retrieval
• imbalance-aware learning
• ontology-grounded prompting
• progressive quantization