Benchmarking Deep Learning Architectures for Predicting Readmission to the ICU and Describing Patients-at-Risk

TL;DR

This study compares various deep learning models, including attention mechanisms and neural ODEs, for predicting ICU readmission risk, highlighting the interpretability of attention-based models with comparable accuracy.

Contribution

It introduces a comprehensive benchmarking of deep learning architectures for ICU readmission prediction, emphasizing interpretability and model comparison using the MIMIC-III dataset.

Findings

Recurrent neural networks with neural ODEs achieved the highest precision.

Predictive accuracy was similar across different neural network architectures.

Patients at risk often had infectious complications or chronic conditions.

Abstract

Objective: To compare different deep learning architectures for predicting the risk of readmission within 30 days of discharge from the intensive care unit (ICU). The interpretability of attention-based models is leveraged to describe patients-at-risk. Methods: Several deep learning architectures making use of attention mechanisms, recurrent layers, neural ordinary differential equations (ODEs), and medical concept embeddings with time-aware attention were trained using publicly available electronic medical record data (MIMIC-III) associated with 45,298 ICU stays for 33,150 patients. Bayesian inference was used to compute the posterior over weights of an attention-based model. Odds ratios associated with an increased risk of readmission were computed for static variables. Diagnoses, procedures, medications, and vital signs were ranked according to the associated risk of readmission. Results: A recurrent neural network, with time dynamics of code embeddings computed by neural ODEs, achieved the highest average precision of 0.331 (AUROC: 0.739, F1-Score: 0.372). Predictive accuracy was comparable across neural network architectures. Groups of patients at risk included those suffering from infectious complications, with chronic or progressive conditions, and for whom standard medical care was not suitable. Conclusions: Attention-based networks may be preferable to recurrent networks if an interpretable model is required, at only marginal cost in predictive accuracy.