Adaptive tumor growth forecasting via neural & universal ODEs

TL;DR

This paper introduces adaptive neural and universal ODE-based models for tumor growth forecasting, enhancing prediction accuracy and interpretability over classical models, especially with limited data.

Contribution

It develops novel neural and universal differential equation models that adapt to patient-specific tumor dynamics and enable symbolic interpretation.

Findings

Improved tumor growth forecasting accuracy.

Ability to recover explicit mathematical dynamics.

Effective modeling with limited data.

Abstract

Forecasting tumor growth is critical for optimizing treatment. Classical growth models such as the Gompertz and Bertalanffy equations capture general tumor dynamics but may fail to adapt to patient-specific variability, particularly with limited data available. In this study, we leverage Neural Ordinary Differential Equations (Neural ODEs) and Universal Differential Equations (UDEs), two pillars of Scientific Machine Learning (SciML), to construct adaptive tumor growth models capable of learning from experimental data. Using the Gompertz model as a baseline, we replace rigid terms with adaptive neural networks to capture hidden dynamics through robust modeling in the Julia programming language. We use our models to perform forecasting under data constraints and symbolic recovery to transform the learned dynamics into explicit mathematical expressions. Our approach has the potential to improve predictive accuracy, guiding dynamic and effective treatment strategies for improved clinical outcomes.