The role that ‘Fitbit-style cough-tracking’ could play in the future of digital health

Joe Brew

Joe Brew

In the realm of digital healthcare, it’s frustrating to see the most sophisticated and cutting-edge technology used to increase click rates on social media, while the medical industry relies on decades-old processes, writes Joe Brew, CEO and Co-Founder of Hyfe.

What is Hyfe, in a nutshell?

Hyfe is an artificial intelligence platform that is dedicated to pioneering the emerging field of “acoustic epidemiology” – that is, using sound to identify and quantify symptoms, detect and track disease, and predict health outcomes.

Hyfe’s main focus is on coughing since it is medically important, and acoustically unique. We are most well-known for our “cough tracker” apps (which we recently launched in the UK on both iOS and Android). The apps are freely available to the general public who want to track their coughs, but our main focus is on our researcher suite of tools, which encompasses our researcher phone app, dashboard and data analysis platform.

How do researchers use Hyfe?

Researchers use Hyfe’s apps in a wide variety of ways. The most common research applications we have seen are:

  • Cough as a biomarker: When participants come to clinics for medical tests, their coughs are collected and paired with their medical tests. When participants return for subsequent tests (e.g., after the onset of treatment), cough is collected again to produce a longitudinal cough dataset.
  • Remote cough monitoring: Between initial and follow-up visits to a clinic, participants are given a study phone with the Hyfe Research app installed. This phone captures all coughs during the monitoring period. They are asked to carry this phone with them until they return.
  • Inpatient monitoring: Phones are placed next to hospitalised patients to track their coughs over the course of their stay.
  • Cough screening: Cough collection is integrated into a medical screening centre (e.g., COVID-19 testing site) using either the Hyfe Research app or the Hyfe Screening website, allowing researchers to pair participants’ coughs with other test results.
  • Syndromic surveillance in a population: Participants download Hyfe Cough Tracker on their own phones and go about their lives. De-identified locations are tracked by the app, if users opt in.
  • Veterinary medicine and livestock monitoring: The Hyfe Veterinary application (not yet released) is placed in an area with one or more animals. Coughs are quantified over space and time.

The studies enabled by these forms of data collection also seem to be limitless. We have some examples of what it has been used for on our website.

In simple terms, how does the technology work?

The Hyfe platform consists of the phone apps, which capture coughs; the server, which categorizes and analyzes coughs; and the researcher/practitioner dashboard, where patient-level or population-level data can be aggregated and analyzed.

The phone app works by detecting “explosive” sounds – short, loud, abrupt sounds. We call this “peak detection”. We do peak detection – rather than simply recording everything – because we want to protect users’ privacy and we want to limit bandwidth usage, particularly in low-resource contexts. Since peak sounds are so short (<0.5 seconds), they don’t contain identifiable information.

After a peak is detected, it gets sent securely to a server where a machine learning model analyzes it. Analysis consists of classifying the sound as a cough, sneeze, other, etc. and then – depending on the type of sound – running downstream algorithms such as disease-specific classifiers.

The disease classifier models are not publicly deployed yet, so the main experience for the user consists of getting their cough frequency quantified (a Fitbit-style dashboard for coughs). The app shows cough frequency over time, and allows for sound playback, sharing with a doctor or friend, etc. It might sound trivial, but it’s not – nearly none of us know intuitively how much we cough, and just knowing this information is helpful in monitoring our own health, identifying triggers, being aware of improvements and deteriorations, etc.

For researchers, doctors, and public health practitioners, a dashboard exists where they can monitor the coughing of individual patients, or their entire “cohort”. This allows them to monitor and analyze coughing among an entire group of people, rather than just individuals.

Is it possible for an app like Hyfe to detect the early stages of an respiratory disease outbreak in the future?

All forms of syndromic surveillance are useful for early outbreak detection, and the most useful forms are those which are rapidly scalable, low-cost and capable of tracking symptoms among those who do not present severe illness (and therefore don’t go to the doctor).

We think that acoustic syndromic surveillance – cough tracking in particular – checks all these boxes. And we think that it will be useful not only for rapidly identifying (and therefore responding to) emerging disease outbreaks, but also for monitoring and better controlling those ongoing pandemics like tuberculosis.

Finally, how was the idea for Hyfe conceived?

As someone operating at the intersection of innovation and public health I have always been bothered by the fact that we are using the most sophisticated, cutting-edge technology to increase click rates on instagram posts, while the interface of healthcare is still using decades-old technology like pen and paper.

This frustration has brought us three co-founders together and, accelerated by what was then the emergence of a global pandemic, so we pulled the trigger and launched Hyfe. Our timing couldn’t have been better, as the pandemic has brought respiratory disease, remote healthcare, coughing, and digital health to the front of mind for many people who found our idea resonating.