GERO LAB: AN API TO USE MOVEMENTS TO PREDICT RISK OF AGE-RELATED DISEASES: INTERVIEW
Locomotome, as coined by the Human Locomotome Project is a set of human locomotive data that can be analyzed to predict human stress levels and proclivity of age-related metabolic or degenerative disorders.
Gero Lab, a new and burgeoning company in this space, has been collecting locomotome data to discover markers of age-related diseases and evaluate the clinical importance of these markers. They have an app that collects initial answers to health questions and then uses activity data from devices like FitBit, Jawbone, and Bodymedia to further cement their locomotome models. Users are then sent metrics on their neurological state and potential health conditions, increasing their awareness of various health factors important for early prevention and lifestyle changes.
Gero co-founder Vera Kozyr answers some of my questions below.
What was the driving force to create Gero? What are the company’s goals?
We were originally studying different biological signals including transcriptome and genome signals, looking for signatures of aging and associated chronic deceases. Then we realized that the locomotome signal is extremely rich and much more convenient to gather, so we adjusted all our mathematical models and algorithms for it. The goal of our company is to create a convenient (non-invasive and seamless) and reliable tool for the early stage diagnosis of different diseases.
How can data collected and used in Gero models be translated into action items for users?
Awareness is very important when it comes to health. Early warnings can be impactful, especially for slowly developing health conditions. For example, life-style changes during the early stages of diabetes type 2 can significantly slow down the development of the disease or even reverse it. In the future, after passing FDA approval, GERO technology could also be used by doctors for preventative measures.
What are some of the most interesting bits of data that you have gathered so far? What is to come?
The key takeaways of our first 3,000 Fitbit study (finished in November of last year) are:
- Motor activity contains signatures of particular chronic deceases (metabolic, psychiatric and neurological)
- Low-resolution trackers (e.g. Fitbit, Jawbone, etc.) can also be used with GERO’s mathematical model with sufficient tracking time
- We are already passed the proof of concept phase to detect particular health conditions with accuracy
We keep working on increasing the accuracy of our algorithms. Along with disease risks and trends, we have learned to detect biological age and gender. At the moment we are focusing on diabetes and soon will publish some of our very interesting findings.
How does the app / data interface help users?
As we are still in the research stage we don’t claim that our app helps users at the moment. It collects activity data and helps to develop our technology. Individual health reports that we will release to our participants of course might potentially help by giving awareness of health conditions and showing their trends.