Beat-to-beat aortic valve opening detection from impedance cardiography using machine learning

Accurate detection of the aortic valve opening, represented by the B-point in the impedance cardiogram (ICG), is a critical step in non-invasive hemodynamic monitoring. The B-point is essential for deriving key parameters like the pre-ejection period (PEP) and left ventricular ejection time (LVET). However, accurately identifying this fiducial point is challenging due to a high intra-individual difference in the waveform, interfering factors like body movement, and a historical lack of publicly available datasets for evaluation.

To overcome this research gap, we propose a novel machine learning-based approach for beat-to-beat B-point detection that leverages domain knowledge from 12 different established ICG-based fiducial point detection algorithms. Based on this feature set, we trained different machine learning pipelines on over 11,000 cardiac cycles from 39 young and healthy individuals included in our publicly available datasets.

The best-performing pipeline, based on a Random Forest Regressor, achieved a considerable improvement in B-point detection accuracy and robustness with a mean absolute error (MAE) of 8.13 ± 12.12 ms.

Compared to the best-performing traditional algorithm, our approach significantly reduced the MAE by 54.7% and its standard deviation by 47.4%. Furthermore, the achieved error is lower than the inter-rater error (8.35 ± 14.78 ms). Our novel B-point detection method is available for the research community through integration into our open-source framework PEPbench.