Deep Learning for Disease Outbreak Prediction: A Robust Early Warning Signal for Transcritical Bifurcations

TL;DR

This work addresses robust early warning signals for disease outbreaks by predicting $transcritical$ bifurcations in noisy dynamics using deep learning. The authors evaluate three architectures (par-LSTM-CNN, seq-LSTM-CNN, Conv1d+SE) trained on two simulated datasets, RAPO and NISIR, and test generalization on SEIR/SEIRx models and real influenza and COVID-19 data. Their best model, par-LSTM-CNN trained on a combined RAPO+NISIR corpus, achieves strong performance on censored data and competitive results on real-world datasets, outperforming at least one prior method on RAPO and matching or exceeding others in several settings. The study highlights the value of diverse, noise-augmented simulations to improve robustness, while noting limitations in transfer to some real-world outbreak scenarios and suggesting avenues like transformers and ensemble strategies for future improvements.

Abstract

Early Warning Signals (EWSs) are vital for implementing preventive measures before a disease turns into a pandemic. While new diseases exhibit unique behaviors, they often share fundamental characteristics from a dynamical systems perspective. Moreover, measurements during disease outbreaks are often corrupted by different noise sources, posing challenges for Time Series Classification (TSC) tasks. In this study, we address the problem of having a robust EWS for disease outbreak prediction using a best-performing deep learning model in the domain of TSC. We employed two simulated datasets to train the model: one representing generated dynamical systems with randomly selected polynomial terms to model new disease behaviors, and another simulating noise-induced disease dynamics to account for noisy measurements. The model's performance was analyzed using both simulated data from different disease models and real-world data, including influenza and COVID-19. Results demonstrate that the proposed model outperforms previous models, effectively providing EWSs of impending outbreaks across various scenarios. This study bridges advancements in deep learning with the ability to provide robust early warning signals in noisy environments, making it highly applicable to real-world crises involving emerging disease outbreaks.

Figures (1)

• Figure 1: Area under Receiver Operator Characteristics curve (AUC) of our model (par-LSTM-CNN), Bury et al.'s model, and Chakraborty et al.'s model in tests with (A) SEIR model, (B) SEIRX model, (C) influenza data, and (D) COVID-19 data.