Key result
Inter-subject configuration performance for blood pressure estimation from PPG was inferior to intra-subject configuration performance across multiple machine learning models.
Why the study?
Previous impressive results for estimating BP from PPG were hypothesized to be excessively optimistic due to train/test split configurations.
Does inter-subject data arrangement compared to intra-subject arrangement affect the performance of blood pressure estimation algorithms using photoplethysmography?
Does inter-subject data arrangement compared to intra-subject arrangement affect the performance of blood pressure estimation algorithms using photoplethysmography?
Machine learning models for PPG-based blood pressure estimation must be evaluated using inter-subject data splits to avoid overly optimistic results caused by the model learning patient identity instead of physiological signals.
Cardiovascular diseases are the leading causes of death, and blood pressure (BP) monitoring is essential for prevention, diagnosis, assessment, and treatment. Photoplethysmography (PPG) is a low-cost opto-electronic technique for BP measurement that allows the acquisition of a modulated light signal highly correlated with BP. There are several reports of methods to estimate BP from PPG with impressive results; in this study, we demonstrate that the previous results are excessively optimistic because of their train/test split configuration. To manage this limitation, we considered intra- and inter-subject data arrangements and demonstrated how they affect the results of feature-based BP estimation algorithms (i.e., XGBoost, LightGBM, and CatBoost) and signal-based algorithms (i.e., Residual U-Net, ResNet-18, and ResNet-LSTM). Inter-subject configuration performance is inferior to intra-subject configuration performance, regardless of the model. We also showed that, using only demographic attributes (i.e., age, sex, weight, and subject index number), a regression model achieved results comparable to those obtained in an intra-subject scenario. Although limited to a public clinical database, our findings suggest that algorithms that use an intra-subject setting without a calibration strategy may be learning to identify patients and not predict BP.
No takes yet. Share an insight, caveat, or question.
Costa et al. (2022) studied Blood pressure monitoring. Intra-subject vs inter-subject data arrangements for PPG-based BP estimation vs. Inter-subject configuration was evaluated on Blood pressure estimation performance. Inter-subject configuration performance for blood pressure estimation from PPG was inferior to intra-subject configuration performance across multiple machine learning models.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: