Related Experiment Video
Updated: Mar 12, 2026
Quantification of Mouse Heart Left Ventricular Function, Myocardial Strain, and Hemodynamic Forces by Cardiovascular Magnetic Resonance Imaging
Published on: May 24, 2021
Towards fully automated synthetic ECV quantification: an open-access machine learning-based approach for fast blood
Rebecca Elisabeth Beyer1,2,3, Markus Hüllebrand2,3,4,5, Patrick Doeblin1,2,3
1Department of Cardiology, Angiology and Intensive Care Medicine, Deutsches Herzzentrum der Charité, Augustenburger Platz 1, 13353, Berlin, Germany.
This study introduces a novel, automated machine learning method for extracellular volume (ECV) assessment, eliminating the need for blood draws. The synthetic ECV shows strong agreement with conventional methods, paving the way for efficient myocardial fibrosis evaluation.
Area of Science:
- Cardiovascular Imaging and Diagnostics
- Artificial Intelligence in Medical Imaging
- Biomarker Quantification
Background:
- Conventional extracellular volume (ECV) quantification requires time-consuming manual post-processing and a blood draw for hematocrit measurement.
- Diffuse myocardial fibrosis assessment is crucial for diagnosing and managing various cardiac conditions.
- There is a need for more efficient and non-invasive methods for ECV assessment.
Purpose of the Study:
- To develop and validate a fully automated, blood draw-free, machine learning-based approach for synthetic ECV assessment.
- To evaluate the feasibility of this automated method for non-invasive assessment of diffuse myocardial fibrosis.
Main Methods:
- Retrospective analysis of 1092 patients undergoing cardiac magnetic resonance (CMR) and ECV measurement at 1.5T or 3T.
- Development of a neural network segmentation model using manual contouring of T1 maps for automated analysis.
- Calculation of fully-automated synthetic ECV using validated sex- and field strength-specific models; agreement assessed via correlation, t-tests, and Bland-Altman analysis.
Main Results:
- Fully-automated synthetic ECV demonstrated strong correlation with conventional ECV (r=0.79, p<0.001).
- No significant differences were observed between synthetic and conventional ECV values (26.9% ± 4.9% vs. 27.3% ± 6.4%, p=0.056).
- Bland-Altman analysis showed minimal mean difference (0.4%) with moderate limits of agreement, exhibiting good agreement for ECV values up to 35%.
Conclusions:
- Fully automated synthetic ECV provides a blood-free proof-of-concept for large-scale post-processing of CMR data.
- This method supports consistent and efficient assessment of myocardial fibrosis in research settings.
- Further validation is required for clinical application, particularly at higher ECV ranges.
