Patient-Specific Cardiac Magnetic Resonance Feature Tracking Approach for Scar Detection in Concomitant Ischemic and...

Area of Science:

• Cardiology
• Medical Imaging
• Biomedical Engineering

Background:

• Cardiac magnetic resonance (CMR) is crucial for evaluating heart disease.
• Feature tracking (FT) analysis of CMR cine images offers insights into myocardial function.
• Differentiating scar tissue in heterogeneous heart conditions remains a challenge.

Purpose of the Study:

• To investigate a patient-specific CMR feature tracking approach for scar detection.
• To assess the utility of global circumferential strain (GCS) in identifying myocardial scar.
• To evaluate this method in a mixed cohort of ischemic and non-ischemic heart disease patients.

Main Methods:

• Retrospective analysis of CMR exams from 89 patients with chronic ischemic and non-ischemic heart disease (IHD+) and 65 with ischemic heart disease (IHD) only.
• Global and segmental circumferential strain (GCS/SCS) derived from native cine images using Segment CMR software.
• Correlation of patient-specific median GCS (GCSmedian) with late gadolinium enhancement (LGE) findings.

Main Results:

• Overall GCS ranged from -3.5% to -19.8%, with lower average GCS in IHD+ patients compared to IHD patients.
• Infarcted segments constituted 19% in IHD and 16.6% in IHD+ groups; non-ischemic LGE was found in 6.7% of segments in IHD+.
• A GCSmedian cutoff of 39.5% demonstrated high accuracy (87.5% sensitivity, 86.3% specificity) for detecting ischemic scar.

Conclusions:

• Patient-specific GCSmedian percentage plots from native cine CMR images can indicate ischemic scar.
• A threshold of approximately 40% GCSmedian suggests the presence of scar tissue.
• This CMR feature tracking method shows promise for scar detection in heterogeneous heart disease populations.