Why the study?
Automated machine learning analysis of cardiac structure and function is hindered by poor generalizability, and comparison against clinicians ignores human observer error while preventing demonstration of superiority.
Does automated machine learning analysis provide similar precision to human clinicians for measuring left ventricular parameters on scan-rescan CMR?
Comparison
Fully automated convolutional neural network vs expert clinician vs trained junior clinician
Design
Multicenter, scan-rescan study across 5 institutions, 2 scanner manufacturers, and 2 field strengths
Follow-up
96% rescan within one week
Authors
Loading...
Automated CMR LVEF analysis matches clinician precision at far greater speed; leaves open prospective validation before routine adoption.
Does automated machine learning analysis provide similar precision to human clinicians for measuring left ventricular parameters on scan-rescan CMR?
Automated machine learning analysis of CMR provides similar precision to human experts but is significantly faster, supporting its translation to clinical practice.
Bhuva et al. (2019) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: