Automated Heart Volume and Mass Measurement Using Non-Contrast Chest CT
Author Information
Author(s): Schmitt Ramona, Schlett Christopher L., Sperl Jonathan I., Rapaka Saikiran, Jacob Athira J., Hein Manuel, Hagar Muhammad Taha, Ruile Philipp, Westermann Dirk, Soschynski Martin, Bamberg Fabian, Schuppert Christopher, El-Baz Ayman
Primary Institution: University Heart Center Freiburg—Bad Krozingen, Medical Center—University of Freiburg
Hypothesis
To validate the automated quantification of cardiac chamber volumes and myocardial mass on non-contrast chest CT using cardiac MR as a reference.
Conclusion
The study found that automated cardiac chamber volume and myocardial mass quantification on non-contrast chest CT produced viable measurements compared to cardiac MR.
Supporting Evidence
- The deep learning model generated CT measurements for 52 of the 53 patients (98%).
- CT measurements correlated well with CMR measurements for LVEDV, LVESV, and LVM (ICC = 0.85, 0.84, and 0.91; all p < 0.001).
- Major qualitative issues were found in three cases (6%).
- The CT results correctly identified 7/15 LV and 1/1 RV as dilated.
Takeaway
This study shows that a computer program can help measure heart sizes and weights using regular chest scans without contrast, which can help doctors find heart problems early.
Methodology
The study included 53 patients who underwent both non-contrast chest CT and cardiac MR within three weeks, using a deep learning model for cardiac segmentation.
Potential Biases
Potential bias due to the retrospective nature of the study and the exclusion criteria that may limit the diversity of cardiac conditions assessed.
Limitations
The limited sample size restricts the study's generalizability and may not represent all cardiac pathologies.
Participant Demographics
The sample comprised 19 female and 34 male patients with a mean age of 64 years.
Statistical Information
P-Value
p<0.001
Confidence Interval
95% CI: −90 mL, 47 mL
Statistical Significance
p<0.001
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website