Search papers, labs, and topics across Lattice.
This study introduces a Transformer-based network for cuffless blood pressure estimation from photoplethysmography (PPG) signals, addressing limitations in existing models that often rely on amplitude shortcuts and late fusion of demographic data. By incorporating demographic conditioning through FiLM-style feature modulation and adding an auxiliary morphology head, the model effectively captures long-range dependencies and subject-specific vascular characteristics. The proposed method achieves significant improvements in accuracy, with a mean absolute error (MAE) of 4.56 mmHg for systolic and 2.62 mmHg for diastolic blood pressure, outperforming previous demographic-enhanced PPG models by 47% and 50%, respectively.
Achieving a 50% reduction in blood pressure estimation errors, this model integrates demographic conditioning and morphology analysis for unprecedented accuracy in cuffless monitoring.
Blood pressure (BP) is a key marker for cardiovascular risk assessment and therapeutic decision-making, and Photoplethysmography (PPG) enables low-cost, wearable-friendly cuffless BP estimation. However, even with recent progress, many PPG-based models are trained with BP regression alone and may rely on amplitude-dominated shortcuts. In addition, demographic covariates that systematically modulate vascular compliance are often incorporated only via late fusion, limiting subject-specific representation learning. We propose a Transformer-based network for cuffless BP estimation from PPG signal, leveraging self-attention to capture long-range dependencies across multiple cardiac cycles. To account for subject-specific vascular differences, the model is conditioned on demographics via FiLM-style feature modulation applied through the attention and feed-forward sublayers of Transformer blocks. In addition, we add an auxiliary morphology head to guide the model to attend to BP-relevant waveform morphology associated with arterial stiffness and wave reflection. Under calibration-based evaluation protocols on the large-scale PulseDB dataset, the proposed method achieves MAE of 4.56 mmHg for systolic BP and 2.62 mmHg for diastolic BP, reducing errors by 47% and 50% compared with prior demographic-enhanced PPG baselines. The resulting lightweight, single-sensor model supports scalable and clinically grounded cuffless BP estimation in calibration-enabled deployment settings.
