Physiological Waveform Features Based Aging Clocks Associate with Cardiovascular Disease

LightGBM-based aging clocks from high-fidelity PPG and ABP waveform features predict age with r=0.92 and R-squared 0.81–0.82 in 17,327 perioperative surgical patients.

• Why it matters: Age-related changes in waveform morphology may reflect cardiovascular health, but it is unclear whether ML-derived age discrepancy can flag cardiovascular disease risk.
• What they did: Using over 840 per–cardiac-cycle time-, frequency-, and statistical-domain features from PPG or ABP, the study trained PPGAge and ABPAge and computed Age-Gap categories versus model-specific MAE.
• The result: Accelerated aging (Age-Gap ≥ +MAE) is linked to higher CVD odds for PPGAge (OR 1.87, p<0.001) and ABPAge (OR 2.28, p<0.001), while delayed aging is linked to lower CVD odds for PPGAge (OR 0.68, p<0.001).

AI summary of the abstract · may contain errors · not medical advice · read the paper.