Establish a pipeline for using wearables’ data for detection, triage, and contact tracing during a pandemic.

## **Stakes**
In a pandemic scenario, costs compound exponentially with the spread of the pathogen. Every day that passes before countermeasure deployment could mean tens of thousands of lives lost.

Earlier detection buys time for every other defense layer. The current detection system is fully reactive: a novel pathogen has to make enough people extremely sick to read as an unusual signal in hospital, lab, or wastewater data. In practice, we generally detect a new pandemic by some threshold of patients seeking care or clinicians noticing out of the ordinary patterns, meaning one or more transmission generations have likely already elapsed.

A pathogen that initially causes very few symptoms or lethal cases, also known as a “stealth” pathogen, may not be detected by our current systems until it is too late to effectively prevent spread or distribute vaccines.

Building a pipeline that takes advantage of data from common wearables like Oura rings and Apple watches to detect pathogens could save tens to hundreds of thousands of lives through early warning.

## Context
[Roughly a third of US adults](https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2805753) wear devices like Oura rings and Apple watches that regularly track minor physiological signals. Infections often perturb those signals before a person even begins to feel sick. These small changes would not be enough to detect a novel pathogen in one person; however, with population-level anomaly tracking, we could get warning weeks before the current system does. The relevant data is plausibly obtainable. Many companies sell real-time continuous feeds, others give research access, and there are even some existing third party data aggregators.

The underlying science is already relatively established for respiratory illness. During COVID, studies from UCSF/UCSD, Stanford, and Scripps all showed wearable signals could flag infection before symptoms. The [UCSF/UCSD study](https://www.nature.com/articles/s41598-022-07314-0) was Oura-based, and showed they could catch cases an average of ~2.75 days before participants sought testing.

Government is interested. For instance, DARPA has [funded work](https://www.rti.org/announcements/rti-international-and-collaborators-receive-darpa-award-develop-system-early-warning) in this direction through its SIGMA+ program. Modeling for this program by Johns Hopkins found that [wearable detection for as low as 5% of the population could lead to the detection of seasonal-flu 5–14 days earlier than current systems](https://www.jhuapl.edu/technical-digest/issues/vol-37-no-4-2025/wearables-based-disease-surveillance). The DoW also funded work here, via the [RATE program](https://www.usa.philips.com/healthcare/customer-story/ai-infection-prediction-aiding-military-readiness) (Rapid Analysis of Threat Exposure), run by the Defense Threat Reduction Agency, the Defense Innovation Unit, and Philips. Their findings even support that detection can be made for individual-level cases. Using Garmin watches and Oura rings, the method in [this study](https://www.nature.com/articles/s41598-022-07764-6) was able to predict infection ~48 hours before cases became symptomatic. Risk scores sometimes increased as much as 6 days in advance of symptom onset.

The case for building a pipeline to analyze this data for detection is getting stronger over time. More people wear these devices each year, and anomaly-detection methods will likely continue to improve. [Industry surveys](https://rockhealth.com/insights/whats-your-score-insights-on-wearables-and-connected-devices-from-rock-healths-2025-consumer-adoption-survey/) estimate device ownership have increased by 33 percentage points in the last ten years. Most importantly, the threat from engineered pathogens is rising as the tools to build one get cheaper and more accessible. Building a pipeline now, ahead of the increase in threat level, is critical.

## **Target End-State**
A standing pipeline exists, within ~12–24 months of a founder starting, that acquires wearable data and turns it into an early diagnosis and contact tracing tool, with an always-on population-level anomaly-detection layer for pathogen detection. This is tested and validated by continual use for normal pathogens like flu and COVID.

## **Owner Profile**
This can be led by a driven generalist with the ability to hire out expertise.
The ideal owner profile:
- Someone who can raise capital.
- Comfort negotiating data access with wearable companies, governments, and platforms (HealthKit / Google Fit / Fitbit).
- ML / anomaly-detection capability, or the ability to recruit it.
- An existing network/path into at least one beachhead e.g. a sector with critical or high-paid employees, a hospital system, or a public health authority.

## **Initial Scoping**
**First 3 Months**
- Talk to wearables companies and research who has tried to do this before.
- Talk to Jonah Weinbaum and Sandeep Patel.
- Consider how to integrate with government programs (some Medicare Advantage plans already cover Apple Watches; similar hooks in VA/DoD).
- Start trying to secure data access: buy from providers or third parties that sell, build an opt-in app pulling from HealthKit/Google Fit/Fitbit APIs, and/or set up government data-sharing.

## **Key Questions**
- Do you need this to be opt-in by users, or can you just buy the data?
- How do you validate the detection layer? Can you do backtesting against real signals from 2019 and 2020 (Covid)? Do they keep data for that long?
- How do you mitigate risk of misuse?

## **Funders and Partners**
**How much money is behind this?**
A [wave of philanthropic capital](https://nanransohoff.substack.com/p/the-third-wave-of-american-philanthropy) is soon to be available for founders looking to work on important problems like biosecurity. The OpenAI Foundation, for instance, is part of an estimated $37–100B/year in new AI-lab-driven giving. Existing philanthropic capital is also significant. Coefficient Giving alone deploys over $100M per year to biosecurity.

**Potential Funders**
[Coefficient Giving](http://coefficientgiving.org)
[BlueDot Impact](http://bluedot.org)

**Potential Partners and Stakeholders**
[SecureBio](https://securebio.org)
[Blueprint Biosecurity](https://blueprintbiosecurity.org)

## Learn More
[DARPA — SIGMA+ and wearables for infectious-disease detection](https://www.darpa.mil/news/2021/covid-19)
[RTI / Garmin — DARPA-sponsored Navy wearable study](https://www.rti.org/news/garmin-wearable-devices-covid-19-detection)
[RTI — how wearable sensors can change the course of a pandemic](https://www.rti.org/insights/how-wearable-sensors-can-change-course-covid-19-pandemic)
[JHU APL — SIGMA+ Human Sentinel Networks](https://www.jhuapl.edu/technical-digest/issues/vol-37-no-4-2025/wearables-based-disease-surveillance)
[Philips — RATE (Rapid Analysis of Threat Exposure)](https://www.usa.philips.com/healthcare/customer-story/ai-infection-prediction-aiding-military-readiness)
[RATE active-duty COVID study, Scientific Reports 2022](https://www.nature.com/articles/s41598-022-07764-6)
[Bloomberg Law — Congress adds $10M to expand RATE](https://news.bloomberglaw.com/health-law-and-business/dod-advances-program-using-wearables-for-early-disease-detection)
[Scripps DETECT study, wearable data to flag illnesses](https://blog.google/products-and-platforms/devices/fitbit/fitbit-detect/)
[Snyder Lab — Real-time alerting system for COVID-19 using wearable data](https://www.nature.com/articles/s41591-021-01593-2)
[SensifAI.Health](https://sensifai.health/)

## Interested in building this?
Contact: Olivia Scharfman, [olivia@bluedot.org](mailto:olivia@bluedot.org).


Wearables Data Pipeline for Detection