我们提出 SensorFM,这是一个面向可穿戴健康的基座模型,基于来自 500 万人的超过一万亿分钟传感器数据进行预训练。通过模型规模与数据的协同扩展,SensorFM 学习到一种通用的人体生理表征,可迁移至 35 项健康预测任务,支持标签高效的自适应和数据填补,并可作为个人健康智能体的基础工具。
快速链接
-
- 复制链接 ×
据估计,目前有数十亿台可穿戴设备正在使用中,持续数天、数周乃至数月精确追踪心率、运动、皮肤温度、血氧水平和睡眠。这一连续、纵向的生理与行为数据流,为预防性、个性化健康提供了最有前景的原材料之一。然而,将这些低层级信号转化为有意义的洞察仍然困难重重。首先,基线生理、生活方式和健康状况因人而异,差异巨大,因此在一个人身上预示风险的某种模式,在另一个人身上未必如此。其次,训练模型所需的标签——确诊诊断、化验结果、经过验证的问卷——获取成本高昂、收集缓慢,而且基本上不可能回溯性收集。因此,大多数可穿戴健康模型都是针对单一结局逐一构建的,采用定制化的监督式流程,瞄准狭窄的终点,难以在人类健康的整体广度上泛化。
在《迈向可穿戴健康数据的通用智能与接口》一文中,我们采取了不同的方法。我们推出了 SensorFM,这是一个大型传感器基础模型,能够直接在人群规模的无标注可穿戴数据上进行学习。SensorFM 在超过一万亿分钟的多模态传感器信号上进行了预训练,这些信号来自五百万名已获知情同意的参与者,它学习到了一种单一的、可复用的、对人体生理感知的表示——这种表示能够跨心血管、代谢、睡眠和心理健康,以及生活方式和人口统计学因素进行迁移。据我们所知,这是迄今为止用于训练模型的最大、最多样化的可穿戴数据集。
从一万亿分钟的传感器数据中学习
为了构建预训练语料库,我们从五百万名已同意将其数据用于健康与保健研究的人员中抽取了去标识化数据,采集时间为 2024 年 9 月至 2025 年 9 月。该数据集覆盖 100 多个国家、美国全部 50 个州,以及 20 多种 Fitbit 和 Pixel Watch 设备型号。我们从每个人身上抽取了数周的数据,产生了超过二十亿小时——也就是超过一万亿分钟——的分钟级分辨率信号。
SensorFM 接收来自五种传感器模态的 34 个一分钟聚合特征:光电容积描记法(PPG)、加速度测量、皮肤电活动(EDA)、皮肤温度和高度测量。这些特征共同捕捉了完整 24 小时窗口内的心率和心率变异性、血氧饱和度、睡眠阶段、运动与步数、皮肤电导以及温度。
SensorFM 不依赖标签,而是通过自监督重建进行学习,其基础是LSM-2方法及其自适应与继承掩码(AIM)框架。这是一个至关重要的设计选择,因为缺失和碎片化的数据(例如某些时间段内数据不可用)在可穿戴设备中是常态,其原因多种多样,包括传感器电源循环、设备从手腕上取下、省电运行模式以及传感器反复开关。传统的自监督方法假设输入是完整、不间断的,因此被迫要么填补空缺(可能引入偏差),要么丢弃不完整的窗口(从而丢弃宝贵的数据)。AIM 两条路都不走:它将现实世界中的缺失视为一种自然产物,直接从不完整的记录中学习,将来自真实空缺的继承 token 与为重建目标而人为掩码的 token 结合起来,并将两者视为等价。其结果是一种在构造上就具备缺失感知能力的表征。SensorFM 不只是容忍碎片化数据,它还富有成效地利用这些数据,如下方的生成结果所示。

SensorFM 通过缺失感知掩码重建,在超过一万亿分钟的多模态传感器数据上进行了预训练。
模型与数据同步扩展带来回报
对于任何基础模型而言,一个核心问题是规模能否转化为能力。我们开展了一组系统性的扩展实验,覆盖了预训练数据量(从约 200 万到 20 亿传感器小时)和模型规模(从 100K 到 100M 参数)各四个数量级的范围。
结果是一个清晰而令人鼓舞的信号:随着数据和容量的增长,预训练损失可预测地下降,而且——关键在于——这些收益能够迁移到下游健康任务上。在完整的五百万人语料库上训练的最大模型(SensorFM-B)相比最小变体将重建损失降低了 31%,并在分类任务上平均提升 9% 的性能(AUC),在回归任务上提升 21% 的性能(Pearson Coefficient)。
最大的提升来自同时扩展这两个维度。按比例增加数据和容量,会在生成式预训练和判别式下游性能上产生近乎线性的收益——而且曲线没有显示出饱和的迹象。在所有模型变体中,SensorFM-B 在 35 项任务中的 33 项上胜出。

将数据和模型容量协同扩展数个数量级,会在预训练和下游性能上同时带来收益,且没有饱和的迹象。
一种表示,多个健康领域
为了测试所学表示到底有多通用,我们在来自三项独立、经机构审查委员会批准的前瞻性研究、共计 13,985 名参与者的 35 项判别式健康任务上评估了 SensorFM。这些任务涵盖六个类别:心血管健康、代谢风险、心理健康、睡眠、人口统计特征和生活方式。
为了直接检验嵌入向量的质量,我们保持 SensorFM 编码器冻结,仅在其之上训练一个轻量级线性头——然后与基于工程化特征训练的监督基线进行比较。以下是我们发现的结果:
- 广泛的泛化能力: 在 SensorFM 嵌入向量上进行的线性探测在 35 项任务中的 34 项上优于基于工程化特征的监督基线,且无需任何任务特定的架构。
- 隐式学习生理特征: 加入人口统计学特征(年龄、性别等)会带来适度提升——但随着模型规模扩大,这一提升逐渐缩小,表明更大的模型在预训练过程中隐式地捕捉到了与生理相关的特征。
- 在难以测量的状况上的优势: 规模化预训练对于抑郁症和焦虑症等状况尤其有价值,这些状况因人而异,在传感器数据中只留下微弱的痕迹。SensorFM 似乎能够学习到那些通常会掩埋此类信号的个体差异,并提取出可在人群中普遍适用的模式。
- 标签效率: 仅使用一小部分标注样本,SensorFM 就迅速超越了仅使用人口统计学特征和基于工程化特征的基线——这在医疗健康领域是一项重要特性,因为高质量的标注数据十分稀缺。

在各项判别性健康任务中,冻结 SensorFM 嵌入向量上线性探测的相对性能提升。
一个由智能体组成的“课堂”,用来构建预测头
一个通用嵌入向量的价值,取决于将其适配到具体任务所需付出的努力。传统上,要把嵌入向量变成针对每个新端点的强预测器,需要人工特征工程、架构选择和超参数调优——这些繁琐的工作会随着任务数量的增长而变得越来越困难。
为了实现这一过程的自动化,我们构建了一个智能体“课堂”:一组相互协作又相互竞争的 LLM 智能体,它们迭代地生成、测试并优化可执行代码,从而在 SensorFM 嵌入向量之上构建预测头。在我们的实验中,该系统探索了超过 30,000 个候选方案。
由智能体设计的预测头在 20 个分类任务中的 16 个、以及 15 个回归任务中的 12 个上,都击败了简单的线性探针。有两个规律尤为突出:方案质量随着搜索过程单调提升,并且会随底层 LLM 的能力而扩展——能力更强的模型(例如更新版本的 Gemini)能产出更好的方案,而智能体之间的协作则能帮助能力较弱的模型缩小差距。

一个由 LLM 智能体组成的“课堂”迭代地编写、测试并优化代码,从而在 SensorFM 嵌入向量之上构建预测头,并随时间推移不断改进方案。
为个人健康智能体奠定基础
最后,我们探究了 SensorFM 是否在端到端场景中具有实用价值——即作为一种将 AI 健康教练建立在个人自身生理数据之上的工具。我们将 SensorFM 集成到一个 Personal Health Agent 中,并比较了基于 31 份真实参与者档案生成健康摘要的三种条件:
- 人口统计数据 + 每日可穿戴设备指标 + SensorFM 预测
- 人口统计数据 + 每日可穿戴设备指标 + 真实测量值
- 人口统计数据 + 仅每日可穿戴设备指标(基线)
一个由临床医生组成的小组在不知晓分组条件的情况下,按照五个评分维度——上下文、相关性、可辩护性、个性化程度和潜在危害——对生成的摘要进行评分,在超过 40 小时的专家评估中产生了 1,860 条评分。结果令人瞩目。加入 SensorFM 预测后,在每一个评分维度上都显著优于基线。
而将智能体建立在 SensorFM 预测之上与建立在真实测量值之上,两者之间没有统计学显著差异——这意味着模型的推理结果对智能体的帮助程度,与真实标签几乎相当。

将 SensorFM 作为 Personal Health Agent 的工具,可提升临床医生评分的回复质量。
结论
SensorFM 指向了可穿戴健康研究的一次转变:从众多定制的、单一结果的模型,转向一个单一的、通用的人类生理表征,它能够灵活、高效且大规模地适配。通过从超过一万亿分钟的无标注传感器数据中学习,它能够泛化到心血管、代谢、睡眠和心理健康领域;支持标签高效的适配和稳健的每日指标估计;能够由智能体课堂自动进行专业化;并能够将个人健康智能体建立在一个人自身的信号之上。
致谢
我们要感谢来自 Google Research、Google DeepMind 和学术界的共同作者与合作者对本项工作所做的贡献。
We present SensorFM, a foundation model for wearable health pre-trained on more than one trillion minutes of sensor data from five million people. By co-scaling model size and data, SensorFM learns a general-purpose representation of human physiology that transfers to 35 health prediction tasks, supports label-efficient adaptation and data infilling, and can serve as a grounding tool for a Personal Health Agent.
Quick links
-
- Copy link ×
Estimates suggest that billions of wearable devices are now in use, precisely tracking heart rate, movement, skin temperature, blood-oxygen levels, and sleep across days, weeks, and months. This continuous, longitudinal stream of physiology and behavior provides one of the most promising raw materials for preventive, personalized health. Yet turning those low-level signals into meaningful insights remains hard. First, baseline physiology, lifestyle, and health vary enormously from person to person, so a pattern that signals risk in one individual may not in another. Second, the labels needed to train models — confirmed diagnoses, lab results, validated questionnaires — are expensive, slow to collect, and essentially impossible to gather retrospectively. As a result, most wearable health models have been built one outcome at a time, with bespoke, supervised pipelines that target a narrow endpoint and struggle to generalize across the full breadth of human health.
In “Towards a General Intelligence and Interface for Wearable Health Data”, we take a different approach. We introduce SensorFM, a Large Sensor Foundation Model that learns directly from unlabeled wearable data at population scale. Pre-trained on over one trillion minutes of multimodal sensor signals drawn from five million consented participants, SensorFM learns a single, reusable representation of sensed human physiology — one that transfers across cardiovascular, metabolic, sleep, and mental health, as well as lifestyle and demographic factors. To our knowledge, this is the largest and most diverse wearable dataset used to train a model to date.
Learning from a trillion minutes of sensor data
To build the pre-training corpus, we sampled de-identified data from five million people who had consented to the use of their data for health and wellness research, captured between September 2024 and September 2025. The dataset spans more than 100 countries, all 50 U.S. states, and over 20 Fitbit and Pixel Watch device models. From each person we drew several weeks of data, yielding over two billion hours — more than a trillion minutes — of minute-resolution signals.
SensorFM ingests 34 one-minute aggregate features derived from five sensor modalities: photoplethysmography (PPG), accelerometry, electrodermal activity (EDA), skin temperature, and altimetry. Together these capture heart rate and heart-rate variability, blood-oxygen saturation, sleep stages, motion and steps, skin conductance, and temperature over a full 24-hour window.
Rather than relying on labels, SensorFM learns through self-supervised reconstruction, building on theLSM-2 approach and its Adaptive and Inherited Masking (AIM) framework. This is a crucial design choice, because missing and fragmented data (e.g., stretches of time where data is not available) is the norm with wearable devices, caused by a variety of factors such as sensors’ power-cycle, devices coming off the wrist, power saving modes of operation, and sensors switching on and off. Conventional self-supervised methods assume complete, uninterrupted inputs and so are forced to either impute the gaps (which can introduce bias) or discard incomplete windows (which throws away valuable data). AIM takes neither path: it treats real-world missingness as a natural artifact and learns directly from incomplete recordings, combining the tokens inherited from genuine gaps with those artificially masked for the reconstruction objective and treating the two as equivalent. The result is a representation that is missingness-aware by construction. SensorFM does not just tolerate fragmented data, it uses it productively, as the generative results below show.

SensorFM is pre-trained on over a trillion minutes of multimodal sensor data via missing-aware masked reconstruction.
Scaling model and data together pays off
A central question for any foundation model is whether scale translates into capability. We ran a systematic set of scaling experiments, spanning four orders of magnitude in both pre-training data volume (from roughly 2 million to 2 billion sensor-hours) and model size (from 100K to 100M parameters).
The result is a clean, encouraging signal: pre-training loss falls predictably as data and capacity grow, and — crucially — those gains carry over to downstream health tasks. The largest model (SensorFM-B) trained on the full five-million-person corpus reduces reconstruction loss by 31% over the smallest variant, and improves downstream performance by an average 9% performance gain (AUC) on classification tasks and 21% of performance gain (Pearson Coefficient) on regression tasks.
The biggest improvements come from scaling both dimensions at once. Increasing data and capacity in proportion produces near-linear gains in both generative pre-training and discriminative downstream performance — and the curve shows no sign of saturating. Across model variants, SensorFM-B wins on 33 of 35 tasks.

Co-scaling data and model capacity by orders of magnitude yields gains in both pre-training and downstream performance, with no sign of saturation.
One representation, many health domains
To test how general the learned representation really is, we evaluated SensorFM across 35 discriminative health tasks drawn from three independent, Institutional Review Board Approved prospective studies with a total of 13,985 participants. The tasks span six categories: cardiovascular health, metabolic risk, mental health, sleep, demographics, and lifestyle.
To probe the quality of the embeddings directly, we kept the SensorFM encoder frozen and trained only a lightweight linear head on top of it — then compared against supervised baselines trained on engineered features. Here’s what we found:
- Broad generalization: Linear probes on SensorFM embeddings outperform the feature-engineered supervised baseline on 34 of 35 tasks, without any task-specific architecture.
- Physiology learned implicitly: Adding demographic features (age, sex, etc.) gives a modest boost — but that boost shrinks as the model scales, suggesting larger models implicitly capture physiologically relevant traits during pre-training.
- Strength in hard-to-measure conditions: Scaled pre-training is especially valuable for conditions like depression and anxiety, which vary from person to person and leave only faint traces in sensor data. SensorFM appears to learn the individual differences that usually bury such signals and pick out the patterns that carry across people.
- Label efficiency: With only a small fraction of labeled examples, SensorFM quickly surpasses both demographic-only and feature-engineered baselines — an important property in healthcare, where high-quality labels are scarce.

Relative performance improvement of linear probing on frozen SensorFM embeddings across discriminative health tasks.
A “classroom” of agents to build prediction heads
A general embedding is only as useful as the effort required to adapt it. Traditionally, turning embeddings into a strong predictor for each new endpoint demands manual feature engineering, architecture selection, and hyperparameter tuning — tedious work that gets harder as the number of tasks grows.
To automate this, we built an agentic “classroom”: a set of collaborating and competing LLM agents that iteratively generate, test, and refine executable code to build prediction heads on the SensorFM embeddings. Across our experiments the system explored more than 30,000 candidate solutions.
The agent-designed heads beat a simple linear probe on 16 of 20 classification tasks and 12 of 15 regression tasks. Two patterns stood out: solution quality improves monotonically over the search, and it scales with the capability of the underlying LLM — more capable models (such as more recent versions of Gemini) produce better solutions, while collaboration between agents helps less capable models close the gap.

A “classroom” of LLM agents iteratively writes, tests, and refines code to build prediction heads on SensorFM embeddings, improving solutions over time.
Grounding a Personal Health Agent
Finally, we asked whether SensorFM is useful end-to-end — as a tool that grounds an AI health coach in a person’s own physiology. We integrated SensorFM into a Personal Health Agent and compared three conditions for generating health summaries from 31 real participant profiles:
- Demographics + daily wearable metrics + SensorFM predictions
- Demographics + daily wearable metrics + ground-truth measurements
- Demographics + daily wearable metrics only (baseline)
A panel of clinicians, blinded to condition, rated the resulting summaries across five rubric dimensions — context, relevance, justifiability, personalization, and potential for harm — producing 1,860 ratings over more than 40 hours of expert evaluation. The findings are striking. Adding SensorFM predictions significantly improved responses over the baseline on every rubric dimension. And there was no statistically significant difference between grounding the agent in SensorFM predictions versus actual ground-truth measurements — meaning the model’s inferences served the agent about as well as the real labels would have.

Using SensorFM as a tool for a Personal Health Agent improves clinician-rated responses.
Conclusion
SensorFM points toward a shift in wearable health research: away from many bespoke, single-outcome models and toward a single, generalist representation of human physiology that can be adapted flexibly, efficiently, and at scale. By learning from over a trillion minutes of unlabeled sensor data, it generalizes across cardiovascular, metabolic, sleep, and mental health; supports label-efficient adaptation and robust daily-metric estimation; can be specialized automatically by an agentic classroom; and can ground a Personal Health Agent in a person’s own signals.
Acknowledgements
We would like to thank our co-authors and collaborators from Google Research, Google DeepMind, and academia for their contributions to this work.