Dicrotic Notch Detection in PPG
The dicrotic notch in PPG corresponds to aortic valve closure and marks the transition from systole to diastole. Its detection enables computation of systolic and diastolic time intervals, reflection index, and arterial stiffness markers — but the notch is often absent or subtle in wrist PPG, challenging automated detection.
The dicrotic notch appears as a brief downward deflection between the systolic and diastolic components of the PPG pulse wave. In finger PPG with high perfusion, the notch is typically prominent and easily detected as a local minimum between the systolic peak and the diastolic peak. In wrist PPG, the notch is often dampened or absent due to greater tissue filtering of the pressure wave and lower perfusion, limiting dicrotic notch-dependent analyses.
Detection methods include second-derivative zero-crossing (the notch corresponds to a positive-to-negative transition in the second derivative between the systolic and diastolic peaks), curvature analysis (the notch is the point of maximum positive curvature on the descending limb), and template-based approaches that match the characteristic V-shaped notch pattern. Multi-scale Gaussian derivative filters at scales matching the typical notch width (20–50 ms) provide robust detection even in noisy signals.
The clinical significance of the dicrotic notch extends beyond timing. Notch depth (ratio of notch minimum to systolic peak) correlates with arterial compliance. Complete absence of the dicrotic notch indicates elevated peripheral vascular resistance and arterial stiffness, associated with aging, hypertension, and atherosclerosis. Progressive notch loss during acute illness can indicate hemodynamic deterioration, providing bedside monitoring value in ICU and perioperative settings.
Frequently Asked Questions
Why is the dicrotic notch absent in many wrist PPG signals?
Wrist tissue acts as a low-pass filter, damping high-frequency pressure wave components including the notch. Lower wrist perfusion reduces SNR. Sensor contact pressure variation further distorts the waveform. Only 40–60% of wrist PPG beats show a detectable dicrotic notch.
What clinical information does the dicrotic notch provide?
Notch timing gives systolic time (start to notch) and diastolic time (notch to next onset). Notch depth indicates arterial compliance. The systolic/diastolic time ratio correlates with cardiac contractility and arterial impedance.
Can machine learning detect the dicrotic notch more reliably?
CNN-based detectors trained on annotated PPG datasets achieve 85–92% notch detection accuracy, outperforming rule-based methods (70–80%). However, when the notch is truly absent due to arterial stiffness, even ML cannot detect what does not exist.