Automatic Analysis of Heart Blood Flow Images
Author Information
Author(s): Jolly Marie-Pierre, Xue Hui, Lu Xiaoguang, Guetter Christoph, Kellman Peter, Hsu Li-Yueh, Arai Andrew E, Zuehlsdorff Sven, Guehring Jens
Hypothesis
To demonstrate feasibility of automatic semi-quantitative per-segment analysis of myocardial perfusion time series according to the AHA 17-segment model.
Conclusion
The study successfully showed that automatic semi-quantitative analysis of heart blood flow images is feasible with data from five subjects.
Supporting Evidence
- The method was validated on all 20 slices with a mean error for landmark detection of 4.5±2.6mm.
- The Dice ratio for myocardial segmentation was 0.93±0.025.
- Median minimal distance between segmentation and reference was 1.13/1.02mm for endo/epi.
Takeaway
Researchers found a way to automatically analyze heart images to see how well blood flows, and it worked well in a small group of people.
Methodology
Five subjects were scanned using two perfusion sequence techniques, and image noise was suppressed while segment boundaries were determined using a landmark detection algorithm.
Limitations
The study involved a small sample size, which may not represent the broader population.
Participant Demographics
Subjects had suspected coronary artery disease (CAD).
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website