Development and Integration of the Digital Laboratory Module CardioSimLab for Teaching Computational Cardiology and Scientific Machine Learning
DOI:
https://doi.org/10.32603/2071-2340-2025-4-68-80Keywords:
digital laboratory module, computational cardiology, neural operators, scientific machine learning, educational technologies, interactive visualization, containerizationAbstract
The paper presents the design, software implementation and methodological validation of the interactive digital laboratory module CardioSimLab intended for teaching computational cardiology, scientific machine learning and hybrid modeling of partial differential equations. The computational core is based on the AR-FNO+CG hybrid scheme in which an autoregressive Fourier neural operator approximates nonlinear parabolic evolution of the transmembrane potential, while the conjugate-gradient method enforces physical consistency of the elliptic coupling. In contrast to the original algorithm-oriented study, this article focuses on system architecture, containerized deployment, interface design, LMS integration and pedagogical validation. A client-server architecture with an asynchronous backend, task queue and web interface for interactive visualization of computational experiments has been implemented. The inference time of one step with Δt=2 ms does not exceed 0.3 s, while GPU memory consumption stays below 2.2 GB. The module has been deployed on 15 educational workstations and integrated with Moodle via REST API. Pilot use in a course on scientific machine learning (n=18) showed improved understanding of spectral bias and error accumulation in autoregressive schemes. The module can be used as a reproducible digital laboratory tool in courses on computational mathematics, biomedical modeling and scientific machine learning.
References
M. Holonicki et al., "AI literacy in higher education: frameworks and competencies," Computers & Education: AI, vol. 6, Art. 100142, 2024.
FGOS VO 3++ in the field 09.03.01 "Computer Science and Engineering," Moscow, 2023. (In Russian)
G. Plank et al., "The openCARP simulation environment for cardiac electrophysiology," Computer Methods and Programs in Biomedicine, vol. 208, Art. 106223, 2021.
V. V. Zakharov and E. S. Petrova, "Digital laboratory workshops in a technical university: methodology and practice," Higher Education in Russia, no. 3, pp. 45–58, 2024. (In Russian)
European Commission, "AI in Education: Guidelines for Higher Education Institutions," Brussels, 2025.
D. Brook et al., "Teaching uncertainty quantification in deep learning: pedagogical strategies and tools," Journal of Machine Learning Research. Education Track, vol. 25, pp. 1–18, 2024.
V. V. Rubtsov, "Design of a digital educational environment: a systems approach," Pedagogy, no. 5, pp. 12–24, 2024. (In Russian)
C. S. Henriquez, "Simulating the electrical behavior of cardiac tissue using the bidomain model," Critical Reviews in Biomedical Engineering, vol. 21, pp. 1–77, 1993.
N. A. Trayanova, "Whole-heart modeling: applications to cardiac electrophysiology," Circulation Research, vol. 108, pp. 113–128, 2011.
S. A. Niederer et al., "Computational models in cardiology," Nature Reviews Cardiology, vol. 16, pp. 100–111, 2019.
Z. Li et al., "Fourier neural operator for parametric partial differential equations," in International Conference on Learning Representations, 2021.
N. Rahaman et al., "On the spectral bias of neural networks," in Proceedings of the 36th International Conference on Machine Learning, vol. 97, pp. 5301–5310, 2019.
W. M. Czarnecki et al., "Sobolev training for neural networks," in Advances in Neural Information Processing Systems, vol. 30, pp. 1–11, 2017.
E. Yu. Shchetinin, "Study of a hybrid approach to cardiac electrophysiology modeling based on spectral neural operators," Computer Tools in Education, no. 1, pp. 39–55, 2026. DOI: 10.32603/2071-2340-2026-1-39-55. (In Russian)
Published
Issue
Section
License
Copyright (c) 2026 Evgeniy Shchetinin, Anna Pestryakova, Andrey Shevchuk
This work is licensed under a Creative Commons Attribution 4.0 International License.
This work is licensed under a Creative Commons Attribution 4.0 International License.
