Forecasting Black Sigatoka Infection Risks with Latent Neural ODEs

TL;DR

This paper introduces MR. NODE, a neural network model using Neural ODEs to predict black Sigatoka infection risks by incorporating climate data, demonstrating superior short-term forecasting and generalization capabilities.

Contribution

The paper presents a novel neural ODE-based model that integrates climate predictors into the latent space for disease risk forecasting, improving adaptability over traditional models.

Findings

Superior generalization performance up to one month ahead

Effective incorporation of climate factors into disease modeling

Potential to aid in disease control strategies

Abstract

Black Sigatoka disease severely decreases global banana production, and climate change aggravates the problem by altering fungal species distributions. Due to the heavy financial burden of managing this infectious disease, farmers in developing countries face significant banana crop losses. Though scientists have produced mathematical models of infectious diseases, adapting these models to incorporate climate effects is difficult. We present MR. NODE (Multiple predictoR Neural ODE), a neural network that models the dynamics of black Sigatoka infection learnt directly from data via Neural Ordinary Differential Equations. Our method encodes external predictor factors into the latent space in addition to the variable that we infer, and it can also predict the infection risk at an arbitrary point in time. Empirically, we demonstrate on historical climate data that our method has superior generalization performance on time points up to one month in the future and unseen irregularities. We believe that our method can be a useful tool to control the spread of black Sigatoka.