Phenotypical Ontology Driven Framework for Multi-Task Learning

TL;DR

This paper introduces OMTL, an ontology-driven multi-task learning framework that leverages medical ontologies to improve phenotype outcome predictions from EHR data, especially when data is limited or imbalanced.

Contribution

The novel contribution is the design of a deep learning architecture that mirrors a medical ontology, enabling effective multi-task learning across related phenotypes.

Findings

OMTL outperforms existing multi-task learning methods.

It effectively leverages medical ontologies for better shared representations.

Demonstrated improved prediction accuracy on MIMIC-III data.

Abstract

Despite the large number of patients in Electronic Health Records (EHRs), the subset of usable data for modeling outcomes of specific phenotypes are often imbalanced and of modest size. This can be attributed to the uneven coverage of medical concepts in EHRs. In this paper, we propose OMTL, an Ontology-driven Multi-Task Learning framework, that is designed to overcome such data limitations. The key contribution of our work is the effective use of knowledge from a predefined well-established medical relationship graph (ontology) to construct a novel deep learning network architecture that mirrors this ontology. It can effectively leverage knowledge from a well-established medical relationship graph (ontology) by constructing a deep learning network architecture that mirrors this graph. This enables common representations to be shared across related phenotypes, and was found to improve the learning performance. The proposed OMTL naturally allows for multitask learning of different phenotypes on distinct predictive tasks. These phenotypes are tied together by their semantic distance according to the external medical ontology. Using the publicly available MIMIC-III database, we evaluate OMTL and demonstrate its efficacy on several real patient outcome predictions over state-of-the-art multi-task learning schemes.