Related Experiment Video
Updated: Jun 16, 2026
11:09
High-Resolution Cardiac Positron Emission Tomography/Computed Tomography for Small Animals
Published on: December 16, 2022
Unique parameter identification for cardiac diagnosis in critical care using minimal data sets.
C E Hann1, J G Chase, T Desaive
1Department of Mechanical Engineering, Centre for Bio-Engineering, University of Canterbury, Christchurch, New Zealand. Chris.Hann@canterbury.ac.nz
Computer Methods and Programs in Biomedicine
|January 26, 2010
Summary
Simplified cardiovascular models improve parameter identification by creating uniquely identifiable structures. A novel feedback control approach ensures global minimum convergence, potentially reducing data requirements for better patient-specific cardiac modeling and critical care management.
Area of Science:
- Biomedical Engineering
- Computational Physiology
- Cardiovascular System Modeling
Background:
- Lumped parameter models of the cardiovascular system often suffer from parameter non-identifiability due to limited data.
- This complexity hinders accurate data fitting and increases computational burden in parameter identification.
- Existing models frequently require ventricle volume data, limiting applicability.
Purpose of the Study:
- To develop uniquely identifiable simplified cardiovascular models from complex systems.
- To introduce a novel parameter identification method using feedback control for guaranteed global minimum convergence.
- To reduce data requirements for cardiovascular model parameterization, potentially excluding ventricle volume.
Main Methods:
- Decomposition of complex cardiovascular models into uniquely identifiable substructures.
- Application of a feedback control system where parameter changes act as actuation forces.
- Utilizing continuous arterial/pulmonary pressure waveforms and end-diastolic time for parameter estimation.
Main Results:
- Demonstrated creation of simplified, uniquely identifiable cardiovascular models.
- The novel parameter identification method converges to the global minimum, ensuring optimal data fitting.
- Potential to exclude ventricle volume from required datasets, simplifying data acquisition.
Conclusions:
- Simplified models provide patient-specific parameters, facilitating the development of more sophisticated cardiac models.
- These models can characterize population trends and un-modeled dynamics from retrospective data.
- Improved cardiovascular management in critical care through enhanced prediction of patient responses to interventions.