Physiological Stratified Models for Personalized Cuffless Blood Pressure Estimation Using Wearable Multi-Modal Sensors

Cuffless blood pressure (BP) estimation from wearable sensors is challenged by significant inter-subject physiological variability. To resolve the technical bottlenecks constraining the clinical translation of cuffless BP monitoring, including sensor modality limitations, dataset generalizability issues, and computational complexity, we propose a two-stage framework that integrates established calibration concepts with a physiologically stratified pre-training stage. First, a one-time, recording with infrared photoplethysmogram (PPG) and pulse pressure wave (PPW) is used to assign subjects to a physiological stratum. Subsequently, cuffless BP estimation is performed using the signals with efficient stratified models. By training distinct models for each stratum, BP estimation accuracy is significantly enhanced compared to one model trained on the entire cohort. The analysis was conducted on the CAS-BP dataset which contains 1272 subjects’ recordings (age: 18-70). Transformer-based models were trained to perform estimation and personalized for test subjects using a fine-tuning process on 30-second data. This process required updating 129 parameters. Our method trains stratified models for each stratum, a baseline model was trained on all subjects’ data for comparison. The baseline model yielded a Mean Absolute Error (MAE) of 4.10 mmHg (Mean Error (ME)±Standard Deviation (SD): -0.15±5.77) for systolic BP (SBP) and 2.69 mmHg (ME±SD: -0.05±3.93) for diastolic BP (DBP). For stratified models, SBP MAE was reduced to 3.95 mmHg, 3.02 mmHg and 2.95 mmHg, and DBP MAE was reduced to 2.59 mmHg, 2.11 mmHg and 2.15 mmHg. We improved estimation accuracy over a generalized baseline model, demonstrating that physiological stratification provides an effective initialization space for computationally lightweight model personalization on wearable devices, requiring minimal calibration data. © 2026 IEEE.