Related Experiment Video
Updated: Jun 13, 2025
11:54
Real-Time Proxy-Control of Re-Parameterized Peripheral Signals using a Close-Loop Interface
Published on: May 8, 2021
4.3K
A GENERATIVE APPROACH TO TESTING THE PERFORMANCE OF PHYSIOLOGICAL CONTROL ALGORITHMS.
Ali Tivay1, Ramin Bighamian2, Jin-Oh Hahn1
1Mechanical Engineering, University of Maryland, College Park, MD 20742, USA.
ASME Letters in Dynamic Systems and Control
|September 12, 2024
Summary
A new generative approach uses virtual subjects to test physiological closed-loop control algorithms. This method efficiently evaluates algorithm performance across diverse patient populations before clinical use.
Area of Science:
- Biomedical Engineering
- Computational Physiology
- Medical Systems Control
Background:
- Physiological closed-loop control algorithms are crucial for autonomous medical care systems.
- These algorithms aim to deliver personalized healthcare therapies.
- Computational approaches are needed to evaluate these algorithms considering patient variability.
Purpose of the Study:
- To present a generative approach for testing physiological closed-loop control algorithms.
- To estimate the distribution of performance metrics across a patient population.
- To demonstrate the approach's utility in a case study of hemodynamic management.
Main Methods:
- Developed a generative physiological model with stochastic and dynamic components.
- Generated virtual subjects representing diverse physiological behaviors.
- Tested a closed-loop fluid resuscitation algorithm against these virtual subjects.
Main Results:
- The approach successfully tested algorithms against virtual subjects with varied physiological characteristics.
- Test results enabled estimation of performance metric distributions in the population.
- Demonstrated applicability in a case study for hemodynamic management.
Conclusions:
- The generative testing approach offers a practical and efficient solution for pre-clinical evaluation.
- This method aids in assessing the robustness of control algorithms.
- Facilitates the development of reliable autonomous medical systems.
