Coach-BP: Computer Vision-Based Assessment and Correction of Blood Pressure Measurement Technique
- 1University of California, San Francisco
Abstract
Background: Accurate blood pressure (BP) measurement, essential for hypertension care, requires proper technique. Posture errors can raise systolic BP by more than 20 mm Hg, yet technique is rarely verified objectively. We aimed to develop and evaluate Coach-BP, a computer vision system that detects technique errors from video to support real-time coaching. Methods: We enrolled adults attending an academic cardiovascular prevention center between February and April 2026 and filmed BP measurements with front and side cameras. We analyzed 408 encounters from 225 participants. Coach-BP uses machine learning models trained on clinician labels to assess five American Heart Association criteria using pose, visual context, and audio-visual features. We evaluated its accuracy against expert clinician ratings by leave-one-participant-out cross-validation. We then examined robustness to simulated camera artifacts and interpreted the features important for each criterion. Findings: At baseline, 138 (62%) of 223 participants had at least one technique error and 54 (24%) two or more. Coach-BP achieved a mean area under the receiver operating characteristic curve (AUROC) of 0.93 (95% CI 0.90-0.94), with 0.97 (0.94-0.99) for legs uncrossed, 0.97 (0.93-1.00) for no talking, 0.92 (0.85-0.98) for back supported, 0.91 (0.86-0.95) for feet flat, and 0.86 (0.81-0.91) for arm supported at heart level. Mean balanced accuracy was 0.88 (0.85-0.90). Performance was robust to simulated camera shift, zoom, and rotation. Its models relied on features clinicians would assess, such as cuffed-hand height and foot position. One model classified back support similarly from either camera. Interpretation: Coach-BP accurately detected errors in the five guideline-recommended criteria of BP measurement technique from clinic video, enabling objective assessment of each criterion.
Related
- Office blood pressure measurement versus ambulatory blood pressure monitoring for detecting undiagnosed hypertension in an apparently healthy Mexican adult cohort: complementary central hemodynamic and arterial stiffness assessment
- The inwardly pointing knee angle: A novel computed tomography‐based measurement technique and validation study
- All-Inside Anterior Cruciate Ligament Reconstruction With Accurate Measurement Technique.
- Comparative Evaluation of Postoperative Corneal Astigmatic Correction Following Pterygium Excision by Bare Sclera Technique Versus Conjunctival Limbal Autograft: A Prospective Observational Study
- Surgical technique for correction of sagittal translation using V-shaped rod repositioning and lifting reduction in ankylosing spondylitis kyphosis
BibTeX
@misc{biswas2026coachbp,
title = {{Coach-BP: Computer Vision-Based Assessment and Correction of Blood Pressure Measurement Technique}},
author = {R. Biswas and A. J. Abad and J. Nugas and J. Shim and M. J. Pletcher and J. A. Lamprea-Montealegre and R. Abbasi-Asl},
howpublished = {medRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.29.26366628},
url = {https://www.medrxiv.org/content/10.64898/2026.10.29.26366628v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Coach-BP: Computer Vision-Based Assessment and Correction of Blood Pressure Measurement Technique
AU - R. Biswas
AU - A. J. Abad
AU - J. Nugas
AU - J. Shim
AU - M. J. Pletcher
AU - J. A. Lamprea-Montealegre
AU - R. Abbasi-Asl
PY - 2026
DA - 2026/10/29
PB - medRxiv
DO - 10.64898/2026.10.29.26366628
UR - https://www.medrxiv.org/content/10.64898/2026.10.29.26366628v1
AB - Background: Accurate blood pressure (BP) measurement, essential for hypertension care, requires proper technique. Posture errors can raise systolic BP by more than 20 mm Hg, yet technique is rarely verified objectively. We aimed to develop and evaluate Coach-BP, a computer vision system that detects technique errors from video to support real-time coaching. Methods: We enrolled adults attending an academic cardiovascular prevention center between February and April 2026 and filmed BP measurements with front and side cameras. We analyzed 408 encounters from 225 participants. Coach-BP uses machine learning models trained on clinician labels to assess five American Heart Association criteria using pose, visual context, and audio-visual features. We evaluated its accuracy against expert clinician ratings by leave-one-participant-out cross-validation. We then examined robustness to simulated camera artifacts and interpreted the features important for each criterion. Findings: At baseline, 138 (62%) of 223 participants had at least one technique error and 54 (24%) two or more. Coach-BP achieved a mean area under the receiver operating characteristic curve (AUROC) of 0.93 (95% CI 0.90-0.94), with 0.97 (0.94-0.99) for legs uncrossed, 0.97 (0.93-1.00) for no talking, 0.92 (0.85-0.98) for back supported, 0.91 (0.86-0.95) for feet flat, and 0.86 (0.81-0.91) for arm supported at heart level. Mean balanced accuracy was 0.88 (0.85-0.90). Performance was robust to simulated camera shift, zoom, and rotation. Its models relied on features clinicians would assess, such as cuffed-hand height and foot position. One model classified back support similarly from either camera. Interpretation: Coach-BP accurately detected errors in the five guideline-recommended criteria of BP measurement technique from clinic video, enabling objective assessment of each criterion.
ER -