Evaluating the Predictive Features of Person-Centric Knowledge Graph Embeddings: Unfolding Ablation Studies

TL;DR

This paper systematically analyzes how different features in person-centric knowledge graph embeddings influence the performance of GNN models in predicting patient readmissions, demonstrating robustness across various data types.

Contribution

It introduces a systematic ablation study approach to evaluate feature importance in person-centric knowledge graph embeddings for healthcare prediction tasks.

Findings

Identifies key predictive features in PKGs for readmission prediction

Shows robustness of GNN models across diverse clinical and social data

Provides insights into feature contributions through ablation studies

Abstract

Developing novel predictive models with complex biomedical information is challenging due to various idiosyncrasies related to heterogeneity, standardization or sparseness of the data. We previously introduced a person-centric ontology to organize information about individual patients, and a representation learning framework to extract person-centric knowledge graphs (PKGs) and to train Graph Neural Networks (GNNs). In this paper, we propose a systematic approach to examine the results of GNN models trained with both structured and unstructured information from the MIMIC-III dataset. Through ablation studies on different clinical, demographic, and social data, we show the robustness of this approach in identifying predictive features in PKGs for the task of readmission prediction.