EUVIMEDEUVIMED entdecken
Open Access

A Hybrid ODE-NN Framework for Modeling Incomplete Physiological Systems

Ahmet Demirkaya, Kyle Lockwood, Georgios Stratis, Tales Imbiriba, Iulian Ilies, Sumientra Rampersad, Elie Alhajjar, Giovanna Guidoboni, Zachary C Danziger, Deniz Erdogmus

IEEE Transactions on Biomedical Engineering · 2025-04 · 72 (4) · 1377-1386

DOI: 10.1109/tbme.2024.3505796 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Biomedical Engineering, 2025-04. Verfasst von Ahmet Demirkaya und 9 weiteren Autor:innen. Hauptthemen laut MEDLINE: Neural Networks, Computer, Models, Biological. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 42 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NIH HHS, NIDDK NIH HHS, NEI NIH HHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper proposes a method to learn approximations of missing Ordinary Differential Equations (ODEs) and states in physiological models where knowledge of the system's relevant states and dynamics is incomplete. The proposed method augments known ODEs with neural networks (NN), then trains the hybrid ODE-NN model on a subset of available physiological measurements (i.e., states) to learn the NN parameters that approximate the unknown ODEs. Thus, this method can model an approximation of the original partially specified system subject to the constraints of known biophysics. This method also addresses the challenge of jointly estimating physiological states, NN parameters, and unknown initial conditions during training using recursive Bayesian estimation. We validate this method using two simulated physiological systems, where subsets of the ODEs are assumed to be unknown during the training and test processes. The proposed method almost perfectly tracks the ground truth in the case of a single missing ODE and state and performs well in other cases where more ODEs and states are missing. This performance is robust to input signal perturbations and noisy measurements. A critical advantage of the proposed hybrid methodology over purely data-driven methods is the incorporation of the ODE structure in the model, which allows one to infer unobserved physiological states. The ability to flexibly approximate missing or inaccurate components in ODE models improves a significant modeling bottleneck without sacrificing interpretability.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Neural Networks, ComputerModels, BiologicalHumansBayes TheoremAlgorithmsComputer Simulation

Förderung

  • NIH HHS
  • NIDDK NIH HHS
  • NEI NIH HHS
  • NINDS NIH HHS
  • NIH
  • NSF

Zitieren

APA

Demirkaya, A., Lockwood, K., Stratis, G., Imbiriba, T., Ilies, I., Rampersad, S., Alhajjar, E., Guidoboni, G., Danziger, Z. C., & Erdogmus, D. (2025). A Hybrid ODE-NN Framework for Modeling Incomplete Physiological Systems. IEEE Transactions on Biomedical Engineering, 72(4), 1377-1386. https://doi.org/10.1109/tbme.2024.3505796

Vancouver

Demirkaya A, Lockwood K, Stratis G, Imbiriba T, Ilies I, Rampersad S, et al. A Hybrid ODE-NN Framework for Modeling Incomplete Physiological Systems. IEEE Transactions on Biomedical Engineering. 2025;72(4):1377-1386. doi:10.1109/tbme.2024.3505796

BibTeX
@article{Demirkaya2025,
  author = {Demirkaya, Ahmet and Lockwood, Kyle and Stratis, Georgios and Imbiriba, Tales and Ilies, Iulian and Rampersad, Sumientra and Alhajjar, Elie and Guidoboni, Giovanna and Danziger, Zachary C and Erdogmus, Deniz},
  title = {A Hybrid ODE-NN Framework for Modeling Incomplete Physiological Systems},
  journal = {IEEE Transactions on Biomedical Engineering},
  year = {2025},
  volume = {72},
  number = {4},
  pages = {1377--1386},
  doi = {10.1109/tbme.2024.3505796}
}
Kostenlos starten →