Pulse wave-driven machine learning for the non-invasive assessment of coronary artery calcification in patients with end-stage renal disease undergoing hemodialysis
Observational analysis shows significant variations in pulse waveforms linked to coronary artery calcification severity in hemodialysis patients, suggesting a novel assessment tool.
Key Points
Pulse waveform analysis can distinguish between varying severities of coronary artery calcification in patients with end-stage renal disease.
The gradient boosting decision tree model achieved an accuracy of 84.1% and a macro-AUC of 0.962 in classifying CAC severity.
Recorded pulse waveforms displayed clear morphological differences corresponding to four levels of CAC severity as identified by low-dose computed tomography.
This innovative approach points to a potential non-invasive method for monitoring cardiovascular risk in patients undergoing hemodialysis.