AUTHOR=Parker Christopher , Nelson Erik , Zhang Tongli TITLE=Applying neural ordinary differential equations for analysis of hormone dynamics in Trier Social Stress Tests JOURNAL=Frontiers in Genetics VOLUME=Volume 15 - 2024 YEAR=2024 URL=https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2024.1375468 DOI=10.3389/fgene.2024.1375468 ISSN=1664-8021 ABSTRACT=The application of Machine Learning (ML) and Artificial Intelligence (AI) has revolutionized data analysis and pattern recognition, leading to significant advances in various academic disciplines. In this study, we explore the use of Neural Ordinary Differential Equations (NODEs) for analyzing hormone dynamics in the hypothalamic-pituitary-adrenal (HPA) axis during Trier Social Stress Tests (TSST). This research aims to understand the HPA axis response in both healthy individuals and patients with Major Depressive Disorder (MDD). Using NODE models, we replicated hormone changes without incorporating any prior knowledge of the control system. The dynamic analysis revealed that stress effects are embedded in the nonautonomous vector fields derived from the NODE model, which were subsequently used as inputs for a Convolutional Neural Network (CNN) for patient classification. Our results demonstrate the potential of combining NODEs and CNNs to classify patients based on disease state, providing a preliminary step towards further research using the HPA axis stress response as an objective biomarker for MDD.