跳到正文
原文
Nature 综合主刊· Paolo Di Achille·· 15 天前AI 评分55

利用智能手表上的AI算法检测高血压

Opportunistically Detecting Signs of Hypertension on a Consumer Smartwatch

AI 导读

研究团队开发了一种名为AI-PPG-ACC-HTN的算法,通过智能手表的多模态波形记录,无需袖带校准即可检测高血压。在一项涉及196名参与者的前瞻性多中心研究中,该算法的敏感性为65.8%,特异性为90.0%,阳性预测值为80.6%,优于传统办公室血压筛查。

正文
  • Article
  • Published: 05 November 2026

Nature (2026) Cite this article

We’re sharing this article early to provide faster access to peer-reviewed, accepted research. It is citable and carries a permanent DOI. This version is subject to further edits and will be replaced automatically by the final Version of Record. All legal disclaimers apply.

Abstract

Hypertension is a silent killer, with over half of affected adults unaware of their condition1,2. This lack of awareness is a major concern, as early intervention is critical for preventing major adverse cardiovascular events3,4. Here we show that applying artificial intelligence (AI) to multimodal waveforms (photoplethysmography and accelerometry) recorded using a widely available consumer smartwatch (AI-PPG-ACC-HTN), without any cuff calibration, can detect hypertension with accuracy levels comparable to traditional cuffed blood pressure (BP) devices in the existing clinical framework. We validated AI-PPG-ACC-HTN in a prospective, multicenter study of 196 diverse participants free from known cardiovascular disease and antihypertensive medication against gold-standard 24-hour ambulatory BP monitoring. Over seven days of real-world monitoring, AI-PPG-ACC-HTN detected hypertension with a sensitivity of 65.8% (95% CI, 54.0%-76.3%), specificity of 90.0% (95% CI, 83.2%-94.7%), and positive predictive value (PPV) of 80.6% (95% CI, 68.6%-89.6%). In comparison, initial office BP screening achieved a sensitivity of 55.3% (95% CI, 43.4%-66.7%), specificity of 90.0% (95% CI, 83.2%-94.7%) and PPV of 77.8% (95% CI, 64.4%-88.0%). These results highlight an opportunity for consumer smartwatches to facilitate population-level opportunistic hypertension screening by prompting at-risk individuals toward clinical assessment and confirmatory diagnosis using established cuff-based methodologies, offering an accessible tool to address this major public health challenge.

This is a preview of subscription content, access via your institution

Access options

Access through your institution

Access Nature and 54 other Nature Portfolio journals

Get Nature+, our best-value online-access subscription

32.70 € / 30 days cancel any time

Learn more

Rent or buy this article

Prices vary by article type

$1.95 to $39.95

Learn more

Prices may be subject to local taxes which are calculated during checkout

Additional access options:

Author information

Authors and Affiliations

  1. Google Research, Cambridge, MA, USA

    Paolo Di Achille, Mingwu Gao, Pramod Rudrapatna & Ming-Zher Poh

  2. Google Research, Mountain View, CA, USA

    Lawrence Cai, Jiang Wu, Bhavna Daryani, Jonathan Wang, Utkarsh Khanna, Ho Ko, Anupam Pathak, Mark Malhotra & Jacqueline B. Shreibati

  3. Google Research, Seattle, WA, USA

    Shwetak Patel & Matthew Thompson

  4. University of Washington, Seattle, WA, USA

    Shwetak Patel

  5. Division of Medicine, Beth Israel Deaconess Medical Center, Harvard Medical School, <City>, MA, USA

    Stephen P. Juraschek

Authors

  1. Paolo Di Achille
  2. Lawrence Cai
  3. Jiang Wu
  4. Mingwu Gao
  5. Bhavna Daryani
  6. Jonathan Wang
  7. Utkarsh Khanna
  8. Ho Ko
  9. Anupam Pathak
  10. Mark Malhotra
  11. Shwetak Patel
  12. Jacqueline B. Shreibati
  13. Stephen P. Juraschek
  14. Pramod Rudrapatna
  15. Matthew Thompson
  16. Ming-Zher Poh

Corresponding author

Correspondence to Ming-Zher Poh.

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Di Achille, P., Cai, L., Wu, J. et al. Opportunistically Detecting Signs of Hypertension on a Consumer Smartwatch. Nature (2026). https://doi.org/10.1038/s41586-026-11336-3

Download citation

  • Received: 01 December 2025

  • Accepted: 20 October 2026

  • Published: 05 November 2026

  • DOI: https://doi.org/10.1038/s41586-026-11336-3

来源:Nature 综合主刊 · nature.com