内容来源：https://research.google/blog/earth-ais-planetary-geospatial-foundation-models-for-global-public-health/

内容总结：

谷歌研究团队发布公共卫生领域新成果，地理空间基础模型助力弥合数据鸿沟

近日，谷歌研究（Google Research）软件工程师阿尔巴兹·穆斯利姆与高塔姆·普拉萨德撰文介绍，谷歌地球人工智能（Google Earth AI）旗下的人口动态基础模型（PDFM）在公共卫生领域取得重要进展。该模型通过自监督学习，将隐私保护搜索趋势、人口流动、建成环境密度及环境决定因素等多元信号压缩为地理位置嵌入向量，无需针对特定任务进行微调即可直接接入现有流行病学工作流程，在全球五项公共卫生挑战中展现出显著应用价值。

传统流行病学监测长期面临报告滞后、地缘政治边界造成数据孤岛以及数据稀疏等瓶颈。即便克服上述限制，传统建模方法仍需大量任务专属数据采集和定制化数据工程管道，在疫情快速暴发或资源受限地区难以部署。PDFM以即插即用方式为流行病学家现有统计与机器学习模型提供地理空间上下文，在多种疾病领域、地理环境和流行病学任务中，其现成的位置嵌入向量表现达到或优于传统输入。

在五项由全球卫生研究合作伙伴主导的独立评估中，PDFM展现出跨领域适用性：

一、跨境疫苗覆盖率预测。 流行病学模型受主权边界制约，但疾病传播并不受限。在美国距加拿大边境150公里内的146个县，仅使用美国数据的模型难以准确预测当地疫苗接种率。西奈山卫生系统与波士顿儿童医院的研究人员将美国县级嵌入向量与加拿大前瞻分拣区嵌入向量相结合，捕捉跨境行为与人口流动溢出效应，使模型对MMR疫苗接种覆盖率变异的解释率从16%提升至22%（提高36%），并为470万边境居民将覆盖率估计精度提升至少3个百分点，揭示了仅靠国内模型可能遗漏的局部模式。

二、心血管疾病死亡实时估算。 美国每年因心血管疾病死亡人数超过91.6万。然而，国家生命统计系统（NVSS）的县级死亡率数据通常滞后1至2年，美国社区调查（ACS）协变量反映的是2至3年前的情况。纽约大学格罗斯曼医学院的研究人员测试了PDFM能否在估算全美约3100个县当年心血管疾病死亡人数时替代基于人口普查的输入。结果显示，PDFM与传统普查数据在实时估算方面无统计学显著差异，且PDFM数据更新、覆盖范围更广，有助于卫生部门依据当前状况而非等待多年调查周期来指导预防资源配置。

三、登革热疫情预测。 在应对登革热等虫媒传染病暴发时，时间至关重要。谷歌与牛津大学及蒙特雷理工学院公共卫生研究人员合作，将PDFM与TimesFM 2.0时间序列基础模型相结合，预测墨西哥约2450个市镇2020年至2025年间的登革热病例数。模型在提前一个月预测时性能提升最为显著，在多达72%的活跃登革热传播市镇中提高了预测精度，总误差减少量是误差增加量的3.4倍，恰好在及时开展病媒控制和临床人员配置决策最为关键的区域发挥作用。

四、产后抑郁症筛查。 产后抑郁症等孕产妇心理健康筛查通常依赖临床入院信息，难以反映塑造母亲风险与就医渠道的更广泛社区条件。华盛顿大学研究人员在CDC PRAMS全国调查的332，970名参与者中评估了PDFM嵌入向量。PDFM不仅能捕捉社区层面社会经济状况（R²=0.45），加入嵌入向量后还持续、统计显著地提升了对高风险母亲的预测能力（已见州AUC提升0.0020；未见州提升0.0038）。在模型训练中从未见过的州，该信号依然有效，可弥补约15%的母亲个人收入与保险记录缺失时的预测信号。在模拟中，若卫生系统可对风险最高的20%母亲进行随访，加入PDFM后每年可多覆盖5640名农村产后抑郁症母亲；若系统目标为覆盖80%的病例，PDFM每年可减少17723次误报。

五、霍乱疫情早期预警。 霍乱等水源性流行病早期预警对口服霍乱疫苗和清洁水源预置至关重要。但疫情暴发罕见——该国403个卫生区中任何一周仅有不到1%出现疫情起始，这种罕见性恰恰使其难以预判。针对世界卫生组织非洲区域办事处基于刚果民主共和国综合疾病监测与应对（IDSR）国家监测数据设定的应用场景，研究团队测试了轻量级、低资源版本的PDFM能否帮助预测霍乱热点。结果显示，提前1至2周预测时，近期病例数已能说明大部分情况，PDFM未显著提升精度；提前4至8周预测时，在仍有时间调配物资的窗口期内，PDFM帮助实现了更准确的预测。这表明短期追踪可依赖近期临床数据，而基础模型嵌入向量捕捉了补充历史数据的潜在环境、连通性和人口决定因素，支持提前一至两个月进行主动规划。

研究团队指出，地理空间基础模型使公共卫生领域得以从被动、局部建模转向行星尺度的主动、时间敏感型健康情报。当前静态快照等局限性正推动时序动态嵌入和面向连通性不足地区的地理迁移学习等前沿研究。通过将行星级上下文数据整合到现有公共卫生工作流程中，有助于改善医疗资源、干预措施和疫情警报向最需要社区的可及性。

PDFM嵌入向量已以“人口动态洞察”（Population Dynamics Insights）形式在Google Maps Platform上预览版商用。学术和公共卫生研究人员还可申请免费访问权限，用于选定的非操作性研究用例。

中文翻译：

2026年10月6日
Arbaaz Muslim，软件工程师，Google Research；Gautam Prasad，软件工程师，Google Research

借助 Google Earth AI 的人口动态基础模型（PDFM），我们可以弥补现有流行病学工作流程中的数据缺口和时间报告滞后问题。在我们最新的研究中，我们展示了五个由合作伙伴驱动的案例研究，说明该模型如何体现了面向全球公共卫生的行星级地理空间基础模型范式。

公共卫生决策高度依赖及时、细粒度的数据。对于心血管疾病或产后抑郁等健康状况，这些数据能告诉我们应将资源和支持集中在何处。对于登革热或霍乱等急性疾病暴发，数据有助于指导紧急操作方案和资源分配。然而，传统的流行病学监测往往受限于数据方面的不足：多年报告滞后、数据被僵硬的地缘政治边界所割裂，以及数据稀疏。即便没有这些限制，传统建模方法也需要大量针对特定任务的数据收集和定制化数据工程流水线，这在疫情快速暴发期间或资源受限的环境中难以部署。

为解决这些系统性瓶颈，我们引入了一种利用行星级地理空间基础模型的公共卫生新范式。我们以 Google Earth AI 的人口动态基础模型（PDFM）作为概念验证，展示如何将“地点”的自监督预训练表征作为即插即用的输入直接整合到现有的健康科学和流行病学工作流程中——从而增强流行病学家已经在使用的统计和机器学习（ML）模型，而非从零构建新的流水线。PDFM 将隐私保护的搜索趋势、人类流动性、建成环境密度和环境决定因素压缩为位置嵌入。无需针对特定任务进行微调，这些开箱即用的位置嵌入在多种疾病领域、地理环境和流行病学任务中，达到或超越了传统输入的效果。

作为 Google Earth AI 的一部分——这是我们的一套地理空间模型套件，连接卫星影像、天气、匿名搜索趋势、人类流动性及其他人口动态——PDFM 使用自监督学习，将以下多样化的隐私保护信号综合为紧凑、通用的嵌入，作为以月度频率刷新的位置“指纹”：

PDFM 无需研究人员自行收集和处理这些原始数据流，其嵌入可以轻松接入现有的机器学习工作流程，提供即用型的地理空间上下文。

此前的工作表明，这些嵌入与任务无关。由于这些日常信号捕捉了健康的潜在社会、行为和环境决定因素，相同的嵌入在填补多种 CDC 健康指标的缺口方面表现出强劲性能。

然而，建立全球健康的新范式需要更高的举证标准。为达到这一标准，我们着手在更广泛的健康挑战中、在全球不同的环境中展示 PDFM 嵌入的价值。

为此，我们的全球健康研究合作伙伴在五个不同的公共卫生挑战中推动了独立评估——涵盖多样化的流行病学任务、资源环境和疾病类型：

我们的评估表明，虽然流行病学模型受主权边界所限，但它们所追踪的结果并不受限。例如，在距加拿大边境 150 公里以内的 146 个美国县中，仅基于国内数据的模型往往难以预测当地的疫苗接种情况。

通过将美国县级嵌入与加拿大前置分拣区（Forward Sortation Area）嵌入相结合，西奈山医疗系统和波士顿儿童医院的研究人员开发了能够捕捉跨境行为和流动性溢出效应的模型。为帮助公共卫生官员精准定位面临麻疹暴发风险的社区，研究人员利用这些新增的跨境上下文，将模型对 MMR 疫苗接种覆盖率变异的解释比例从 16% 提升至 22%（提升 36%），并为 470 万边境居民将覆盖率估计值精细化至少 3 个百分点，揭示了仅基于国内数据的模型可能遗漏的局部模式。

心血管疾病（CVD）在美国每年夺去超过 916,000 人的生命。然而，来自国家生命统计系统（NVSS）的未经抑制的官方县级死亡率数据通常滞后 1 至 2 年，而美国社区调查（ACS）的协变量反映的是 2 至 3 年前的情况。PDFM 覆盖更多地区且滞后要小得多，从而提升了我们建模疾病的能力。

我们在纽约大学格罗斯曼医学院的合作伙伴测试了 PDFM 能否在估算全美约 3,100 个县的当年 CVD 死亡人数时替代这些基于人口普查的输入。他们发现，PDFM 与传统普查数据在现时预测方面没有统计学上的显著差异，而 PDFM 的数据更新得多，且在更多地方可用：

这些结果表明，PDFM 可以帮助卫生部门依据当前状况来指导预防资源的配置，而不必等待多年度的调查周期。

在应对登革热等病媒传播疾病暴发时，时间尤为关键。我们与牛津大学和蒙特雷理工学院（Tecnológico de Monterrey）的公共卫生研究人员合作，将 PDFM 与 TimesFM 2.0（我们的开源时间序列基础模型）相结合，预测 2020 年至 2025 年间墨西哥约 2,450 个市镇的登革热病例数。

该模型最大的性能提升出现在提前一个月的预测中，在多达 72% 的登革热活跃传播市镇中提高了预测准确性，总误差减少量是误差增加量的 3.4 倍。这一性能恰好体现在及时的病媒控制和临床人员配置决策最为关键的地方，尤其考虑到短期预测已被证明的实用性。

筛查产后抑郁（PPD）等孕产妇心理健康状况通常依赖临床接诊信息，而这些信息很少能捕捉到影响母亲风险和就医机会的更广泛社区条件。为测试地理空间基础模型能否帮助弥合这一差距，华盛顿大学的研究人员在 CDC PRAMS 全国调查的 332,970 名参与者中评估了 PDFM 嵌入。

除了 PDFM 能够捕捉社区层面的社会经济状况（R²=0.45）之外，加入这些嵌入还一致且具有统计学显著性地提升了对哪些母亲处于风险中的预测能力（在已见过的州中 AUC +0.0020；在未见过的州中 +0.0038）。至关重要的是，这一信号在模型训练期间从未见过的州中依然成立——作为补充性的本地上下文，在无法获取母亲自身收入和保险记录时，可恢复约 15% 的预测信号。

这种可迁移的上下文在新州中如何调配筛查资源方面产生了最大差异。在模拟中，当卫生系统能够对风险最高的 20% 母亲进行随访时，加入 PDFM 帮助每年多覆盖 5,640 名农村产后抑郁母亲。反之，在以覆盖 80% 病例为目标的系统中，PDFM 每年减少了 17,723 次误报——表明地理空间嵌入如何帮助卫生系统要么扩大农村外展覆盖，要么提高随访效率。

霍乱等水源性疫情的早期预警对于预先部署口服霍乱疫苗和清洁水源至关重要。所幸疫情较为罕见：在任何一周内，全国 403 个卫生区中启动疫情的不超过百分之一。但不幸的是，正因为如此罕见，它们难以被预判。

为应对世界卫生组织非洲区域办事处利用刚果民主共和国综合疾病监测与应对（IDSR）报告系统中的国家监测数据所定义的一个应用场景，我们测试了 PDFM 的轻量级、低资源版本（针对互联网连接稀疏的地区进行了适配）能否帮助预测霍乱热点。

其效益取决于我们预判的时间跨度。提前一到两周时，近期病例数已能说明大部分情况，PDFM 未能显著提升准确性。提前四到八周时，在仍有时间调运物资的情况下，它有助于产生：

这些发现表明，虽然短期追踪可以依赖近期临床数据，但基础模型嵌入能够捕捉补充历史数据的潜在环境、连通性和人口决定因素，从而支持提前一到两个月的主动规划。

地理空间基础模型使得从被动的局部建模转向行星尺度上主动的、时间敏感的健康情报成为可能。当前的局限性，如静态快照，正推动着对时间动态嵌入和面向连接不足地区的地理迁移学习的积极研究。

通过将行星级上下文数据整合到现有的公共卫生工作流程中，我们可以帮助改进医疗资源、干预措施和疫情警报送达最需要它们的社区的方式。阅读论文了解更多详情。

PDFM 嵌入已作为 Population Dynamics Insights 以预览版商业提供，这是来自 Google Maps Platform 的地理空间嵌入数据集。学术和公共卫生研究人员还可以申请免费访问，用于选定的非操作性研究用例。

英文来源：

October 6, 2026
Arbaaz Muslim, Software Engineer, Google Research, and Gautam Prasad, Software Engineer, Google Research
With Google Earth AI’s Population Dynamics Foundation Model (PDFM), we can address the data gaps and temporal reporting lags of existing epidemiological workflows. In our latest work, we present five partner-driven case studies demonstrating how this model exemplifies the planetary geospatial foundation model paradigm for global public health.
Public health decisions rely heavily on timely, granular data. For health conditions, such as cardiovascular disease or postpartum depression, that data shows us where to focus resources and support. For acute disease outbreaks, such as dengue or cholera, it can help inform urgent operational protocols and resource allocation. However, conventional epidemiological surveillance is often hindered by limitations in data: multi-year reporting lags, data siloed by rigid geopolitical boundaries, and data sparsity. Even without these limitations, traditional modeling approaches require extensive task-specific data collection and custom data engineering pipelines that are difficult to deploy during rapid outbreaks or in resource-constrained settings.
To address these systemic bottlenecks, we introduce a new paradigm in public health leveraging planetary geospatial foundation models. Using Google Earth AI’s Population Dynamics Foundation Model (PDFM) as a proof-of-concept, we demonstrate how self-supervised, pre-trained representations of "place" can be integrated directly into existing health sciences and epidemiological workflows as plug-and-play inputs — enhancing the statistical and machine learning (ML) models epidemiologists already use, rather than building new pipelines from scratch. PDFM compresses privacy-preserving search trends, human mobility, built-environment density, and environmental determinants into location embeddings. Without requiring task-specific fine-tuning, these off-the-shelf location embeddings matched or improved on conventional inputs across a wide variety of disease domains, geographic settings, and epidemiological tasks.
Part of Google Earth AI — our suite of geospatial models connecting satellite imagery, weather, anonymous search trends, human mobility, and other population dynamics — PDFM uses self-supervised learning to synthesize the following diverse, privacy-preserving signals into compact, versatile embeddings that serve as “fingerprints” for locations refreshed at a monthly cadence:
Rather than requiring researchers to collect and process these raw data streams themselves, PDFM embeddings can be easily plugged into existing ML workflows to provide ready-to-use geospatial context.
Prior work showed that these embeddings are task-agnostic. Because these everyday signals capture the underlying social, behavioral, and environmental determinants of health, the same embeddings showed strong performance on filling gaps in a wide variety of CDC health metrics.
However, establishing a new paradigm for global health carries a higher burden of proof. To meet this standard, we set out to demonstrate the value of PDFM embeddings across a wider variety of health challenges, and in different environments across the globe.
To achieve this, our global health research partners facilitated independent evaluations across five distinct public health challenges — spanning diverse epidemiological tasks, resource settings, and disease types:
Our evaluations show that while epidemiological models are constrained by sovereign borders, the outcomes they track are not. For example, in the 146 U.S. counties situated within 150 km of the Canadian border, domestic-only models often struggle to predict local vaccine uptake.
By supplementing U.S. county embeddings with Canadian Forward Sortation Area embeddings, researchers at the Mount Sinai Health System and Boston Children’s Hospital developed models that captured cross-border behavioral and mobility spillovers. To help public health officials pinpoint communities at risk of measles outbreaks, the researchers used this added cross-border context to increase the share of variation in MMR vaccination coverage explained by the models from 16% to 22% (a 36% increase) and refine coverage estimates by at least 3 percentage points for 4.7 million border residents, revealing local patterns that domestic-only models may miss.
Cardiovascular disease (CVD) claims over 916,000 lives annually in the U.S. However, unsuppressed official county-level mortality data from the National Vital Statistics System (NVSS) typically lags by 1–2 years, and American Community Survey (ACS) covariates reflect conditions up to 2–3 years in the past. PDFM is available in more places and has much less of a lag, allowing for the ability to improve our capacity to model disease.
Our partners at NYU Grossman School of Medicine tested whether PDFM could stand in for these census-based inputs when estimating current-year CVD deaths across roughly 3,100 U.S. counties. They found no statistically significant differences between PDFM and traditional census data on nowcasting, with PDFM being much fresher and available in far more places:
These results suggest that PDFM can enable health departments to guide prevention resources using current conditions rather than waiting on multi-year survey cycles.
Time is especially of the essence when addressing outbreaks of vector-borne diseases like dengue. In partnership with public health researchers at the University of Oxford and Tecnológico de Monterrey, we coupled PDFM with TimesFM 2.0 (our open time-series foundation model) to forecast dengue case counts across ~2,450 Mexican municipalities between 2020 and 2025.
The model’s greatest performance gains occurred when predicting one month ahead, improving forecast accuracy in up to 72% of active dengue transmission municipalities, yielding total error reductions 3.4X larger than degradations. This performance occurred precisely where timely vector control and clinical staffing decisions matter most, especially considering the demonstrated utility of short term forecasts.
Screening for maternal mental health conditions like postpartum depression (PPD) typically relies on clinical intake information, which rarely captures the broader community conditions that shape a mother’s risk and access to care. To test whether geospatial foundation models can help bridge this gap, researchers at the University of Washington evaluated PDFM embeddings across 332,970 participants in the CDC PRAMS national survey.
In addition to PDFM capturing community-level socioeconomic conditions (R2=0.45), adding the embeddings provided a consistent, statistically significant boost to predicting which mothers were at risk (AUC +0.0020 in seen states; +0.0038 in unseen states). Crucially, this signal held up in states the model had never seen during training—acting as complementary local context that recovers about 15% of the predictive signal of a mother’s own income and insurance records when those details are unavailable.
This transferable context made the biggest difference in how screening resources are directed in new states. In simulations where a health system can follow up with the 20% highest-risk mothers, adding PDFM helped reach 5,640 more rural mothers with postpartum depression each year. Alternatively, in systems aiming to catch 80% of all cases, PDFM cut 17,723 false alarms annually—showing how geospatial embeddings can help health systems either broaden rural outreach or improve follow-up efficiency.
Early warning for waterborne epidemics like cholera is critical for prepositioning oral cholera vaccines and clean water supplies. Fortunately, outbreaks are rare: in any given week, fewer than 1 in 100 of the country's 403 health zones sees one begin. Unfortunately, this very same rarity makes them hard to anticipate.
To address a use case defined by the World Health Organization Regional Office for Africa using national surveillance data from the Democratic Republic of the Congo’s Integrated Disease Surveillance and Response (IDSR) reporting, we tested whether a lightweight, low-resource version of PDFM (adapted for regions with sparse internet connectivity) could help forecast cholera hotspots.
The benefit depended on how far ahead we looked. One or two weeks out, recent case counts told most of the story and PDFM did not significantly enhance the accuracy. Four to eight weeks out, when there is still time to move supplies, it helped produce:
These findings demonstrate that while short-term tracking can rely on recent clinical data, foundation model embeddings capture underlying environmental, connectivity, and population determinants that supplement historical data, allowing for proactive planning one to two months in advance.
Geospatial foundation models enable moving from reactive, localized modeling to proactive and time-sensitive health intelligence at planetary scale. Current limitations, such as static snapshots, are driving active research into temporally dynamic embeddings and geographic transfer learning for under-connected regions.
By integrating planetary contextual data into existing public health workflows, we can help take steps to improve how healthcare resources, interventions, and outbreak alerts reach the communities that need them most. Read the paper for more details.
PDFM embeddings are commercially available in Preview as Population Dynamics Insights, a geospatial embeddings dataset from Google Maps Platform. Academics and public health researchers can also request no-cost access for select, non-operational research use cases.