Deep Supervised Information Bottleneck Hashing for Cross-modal Retrieval based Computer-aided Diagnosis

TL;DR

This paper introduces DSIBH, a deep supervised hashing method that enhances cross-modal medical data retrieval accuracy by reducing ambiguous semantics and superfluous information in hash codes.

Contribution

It extends the Deep Deterministic Information Bottleneck to cross-modal scenarios, improving discriminability of hash codes for medical data retrieval.

Findings

DSIBH outperforms state-of-the-art methods in accuracy

Reduces superfluous information in hash codes

Enhances discriminability of cross-modal medical data retrieval

Abstract

Mapping X-ray images, radiology reports, and other medical data as binary codes in the common space, which can assist clinicians to retrieve pathology-related data from heterogeneous modalities (i.e., hashing-based cross-modal medical data retrieval), provides a new view to promot computeraided diagnosis. Nevertheless, there remains a barrier to boost medical retrieval accuracy: how to reveal the ambiguous semantics of medical data without the distraction of superfluous information. To circumvent this drawback, we propose Deep Supervised Information Bottleneck Hashing (DSIBH), which effectively strengthens the discriminability of hash codes. Specifically, the Deep Deterministic Information Bottleneck (Yu, Yu, and Principe 2021) for single modality is extended to the cross-modal scenario. Benefiting from this, the superfluous information is reduced, which facilitates the discriminability of hash codes. Experimental results demonstrate the superior accuracy of the proposed DSIBH compared with state-of-the-arts in cross-modal medical data retrieval tasks.