Learning Inter-Modal Correspondence and Phenotypes from Multi-Modal Electronic Health Records

TL;DR

This paper introduces cHITF, a novel tensor factorization method that jointly infers inter-modal correspondence and discovers clinically meaningful phenotypes from multi-modal EHR data, improving predictive accuracy.

Contribution

The proposed cHITF method addresses the challenge of unobserved inter-modal correspondence in EHR data, enhancing phenotype discovery without prior correspondence assumptions.

Findings

cHITF accurately infers inter-modal correspondence.

Discovered phenotypes are more clinically relevant and diverse.

Achieves superior predictive performance on MIMIC-III dataset.

Abstract

Non-negative tensor factorization has been shown a practical solution to automatically discover phenotypes from the electronic health records (EHR) with minimal human supervision. Such methods generally require an input tensor describing the inter-modal interactions to be pre-established; however, the correspondence between different modalities (e.g., correspondence between medications and diagnoses) can often be missing in practice. Although heuristic methods can be applied to estimate them, they inevitably introduce errors, and leads to sub-optimal phenotype quality. This is particularly important for patients with complex health conditions (e.g., in critical care) as multiple diagnoses and medications are simultaneously present in the records. To alleviate this problem and discover phenotypes from EHR with unobserved inter-modal correspondence, we propose the collective hidden interaction tensor factorization (cHITF) to infer the correspondence between multiple modalities jointly with the phenotype discovery. We assume that the observed matrix for each modality is marginalization of the unobserved inter-modal correspondence, which are reconstructed by maximizing the likelihood of the observed matrices. Extensive experiments conducted on the real-world MIMIC-III dataset demonstrate that cHITF effectively infers clinically meaningful inter-modal correspondence, discovers phenotypes that are more clinically relevant and diverse, and achieves better predictive performance compared with a number of state-of-the-art computational phenotyping models.