原标题：Google's AI detects high blood pressure from a regular smartwatch without a cuff
Google Research 的 AI 无需血压袖带，仅靠智能手表内置的脉搏传感器和运动传感器检测高血压迹象，研究发表于 Nature。神经网络从近 50 万小时传感器数据中无监督学习脉搏与心跳模式，在覆盖五家美国研究机构的 196 人研究中，检出 65.8% 的高血压患者，高于标准院内测量的 55.3%。该方法已纳入 Google Health 应用的血压趋势功能，但仍是研究项目，不能替代诊断。
原文给出 Nature 研究的检出率数字和与袖带测量的对比，读者可据此评估无袖带血压检测的可用程度。
本页的摘要、推荐理由与解读出自 AIsa Studio 的 AI 风向智能体，原文版权归原作者所有。
A Google Research AI can detect signs of high blood pressure using a regular smartwatch, with no blood pressure cuff needed for calibration. The AI relies on two sensors already built into modern watches—a pulse sensor and a motion sensor—according to a Google blog post. The study was published in Nature.
A neural network learned from nearly 500,000 hours of sensor data by comparing short segments and figuring out which ones belonged to the same person, picking up typical pulse and heartbeat patterns without any diagnoses to guide it. The system then averages these patterns over several days and calculates a probability for high blood pressure. Google's health model SensorFM also learns from smartwatch data.
In a study of 196 people across five US research sites, the watch spotted hypertension about as well as a traditional cuff, identifying 65.8 percent of those with high blood pressure compared to 55.3 percent for the standard in-office measurement.
Google says the findings have been folded into the blood pressure trends feature in the Google Health app, but the method is still a research project and doesn't replace a diagnosis. The goal is to nudge people toward getting checked. More than half of people with high blood pressure don't know they have it, according to Google.
Subscribe to THE DECODER for ad-free reading, a weekly AI newsletter, our exclusive "AI Radar" frontier report six times a year, full archive access, and access to our comment section.
原文版权归 The Decoder 及原作者所有，本站转载仅供阅读。如需删除，请联系 support@aisa.one。
2026年11月5日，Google Research 联合 Beth Israel Deaconess Medical Center 与 Harvard Medical School 在 Nature 发表 AI-PPG-ACC-HTN 系统，利用消费级智能手表的 PPG（光电容积脉搏波）与加速度传感器被动检测高血压迹象，无需袖带校准。方法分两阶段：先在 2109 名参与者共 492,347 小时的传感器数据上进行自监督预训练，再按人聚合嵌入并结合身高做分类。在美国五家中心开展的前瞻性研究中，初始筛查场景（N=196）敏感度为 65.8%、特异度 90.0%、PPV 80.6%；确诊场景（N=48）敏感度 78.4%、特异度 90.9%、PPV 96.7%；与 ABPM 金标准对比表现接近传统袖带流程。11月7日，The Decoder 报道称该方法检出 65.8% 的高血压患者，高于标准院内测量的 55.3%，与主报道数字一致。研究结果支持了 Google Health 应用中的血压趋势 Health Guardian 功能。团队强调这是早期可行性研究，不代表任何现有或即将推出的产品；跟进报道同样指出该功能仍是研究项目，不能替代诊断。
Studio