Key result
Novel cuffless laser interferometry biosensor tracks blood pressure within ~1 mmHg of mean absolute error.
Why the study?
Continuous and precise cuffless blood pressure monitoring methods with high sensitivity and computational efficiency are needed to improve BP assessment accuracy.
Does a wearable cuffless biosensor using laser self-mixing interferometry and an SA-CNN model accurately measure blood pressure in volunteers?
Does a wearable cuffless biosensor using laser self-mixing interferometry and an SA-CNN model accurately measure blood pressure in volunteers?
A novel cuffless wearable blood pressure monitoring system using laser self-mixing interferometry and deep learning demonstrates high accuracy that meets clinical standards in a controlled setting.
Answers are grounded in this paper and cite their sources.
Encourages cuffless BP device refinement; leaves open prospective validation in clinical populations before practice change.
This paper presents a wearable cuffless biosensor for continuous and precise blood pressure (BP) monitoring, leveraging the high sensitivity of laser self-mixing interferometry (LSMI) to detect subtle pulsatile motions from the radial artery. To avoid the complex pulse wave reconstruction preprocess, we introduce a fast time–frequency analytical (TFA) methodology based on frequency synchrosqueezed transform (FSST), which characterizes superior time–frequency profile resolution for non-stationary LSMI signals and greater computational efficiency compared to the traditional tools such as continuous wavelet transform (CWT). We further develop a hybrid self-attention- Convolutional Neural Network (SA-CNN) model that effectively links the extracted time–frequency features to BP values by capturing long-range dependencies through self-attention and local patterns via convolutional operations. The system was evaluated in a controlled setting with 180 volunteers, achieving mean absolute error and standard deviation of 1.31 ± 1.55 mmHg for systolic BP and 1.04 ± 1.45 mmHg for diastolic BP, outperforming existing models and meeting the AAMI SP-10 standard. This study demonstrates the significant potential of combining advanced optical sensing with deep learning for high-accuracy BP assessment in clinical and stationary settings.
No takes yet. Share an insight, caveat, or question.
Zhao et al. (2026) studied Blood pressure monitoring (n=180). Wearable cuffless biosensor using laser self-mixing interferometry and SA-CNN model was evaluated on Mean absolute error and standard deviation for systolic and diastolic blood pressure. A cuffless biosensor using laser self-mixing interferometry and a SA-CNN model measured BP with a mean absolute error of 1.31 ± 1.55 mmHg (systolic) and 1.04 ± 1.45 mmHg (diastolic).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: