FLICU: A Federated Learning Workflow for Intensive Care Unit Mortality Prediction

TL;DR

This paper introduces a federated learning workflow for ICU mortality prediction using sensitive healthcare data, demonstrating comparable performance to centralized models and superiority over local models, while preserving patient privacy.

Contribution

It proposes a novel federated learning workflow for ICU mortality prediction and benchmarks its performance against centralized and local models using multivariate time series data.

Findings

Federated learning performs comparably to centralized models in ICU mortality prediction.

Federated learning outperforms local machine learning models.

The approach preserves privacy while maintaining high predictive accuracy.

Abstract

Although Machine Learning (ML) can be seen as a promising tool to improve clinical decision-making for supporting the improvement of medication plans, clinical procedures, diagnoses, or medication prescriptions, it remains limited by access to healthcare data. Healthcare data is sensitive, requiring strict privacy practices, and typically stored in data silos, making traditional machine learning challenging. Federated learning can counteract those limitations by training machine learning models over data silos while keeping the sensitive data localized. This study proposes a federated learning workflow for ICU mortality prediction. Hereby, the applicability of federated learning as an alternative to centralized machine learning and local machine learning is investigated by introducing federated learning to the binary classification problem of predicting ICU mortality. We extract multivariate time series data from the MIMIC-III database (lab values and vital signs), and benchmark the predictive performance of four deep sequential classifiers (FRNN, LSTM, GRU, and 1DCNN) varying the patient history window lengths (8h, 16h, 24h, 48h) and the number of FL clients (2, 4, 8). The experiments demonstrate that both centralized machine learning and federated learning are comparable in terms of AUPRC and F1-score. Furthermore, the federated approach shows superior performance over local machine learning. Thus, the federated approach can be seen as a valid and privacy-preserving alternative to centralized machine learning for classifying ICU mortality when sharing sensitive patient data between hospitals is not possible.