Ensembling Neural Networks for Improved Prediction and Privacy in Early Diagnosis of Sepsis

TL;DR

This paper demonstrates that ensembling neural networks tailored for individual patients enhances early sepsis prediction accuracy and privacy, leveraging parallel training and differential privacy techniques on ICU data.

Contribution

It introduces a novel framework combining patient-specific neural network ensembles with privacy guarantees for early sepsis diagnosis.

Findings

Ensembles outperform single models trained on pooled data.

Parallel and asynchronous training improves efficiency.

Differential privacy is effectively applied to patient-specific models.

Abstract

Ensembling neural networks is a long-standing technique for improving the generalization error of neural networks by combining networks with orthogonal properties via a committee decision. We show that this technique is an ideal fit for machine learning on medical data: First, ensembles are amenable to parallel and asynchronous learning, thus enabling efficient training of patient-specific component neural networks. Second, building on the idea of minimizing generalization error by selecting uncorrelated patient-specific networks, we show that one can build an ensemble of a few selected patient-specific models that outperforms a single model trained on much larger pooled datasets. Third, the non-iterative ensemble combination step is an optimal low-dimensional entry point to apply output perturbation to guarantee the privacy of the patient-specific networks. We exemplify our framework of differentially private ensembles on the task of early prediction of sepsis, using real-life intensive care unit data labeled by clinical experts.