One Drop Unveils New AI Predictions for CGM Users

One Drop presented long-term A1C, blood pressure, and weight outcomes forecasts for individuals, along with overnight hypoglycemia predictions for CGM users at the 80th Annual American Diabetes Association Virtual Scientific Sessions. New predictions lay the foundation for One Drop’s personalized and adaptive data-driven health products.
 
To date, One Drop has collected more than 11 billion health data points from nearly 3 million One Drop app users in 195 countries. The app collects health data through manual input, automatic upload from connected devices, HealthKit, GoogleFit, and direct integrations with Dexcom, Fitbit, Apple Watch, Companion Medical, and other strategic partners.
 
This pooled population data powers the machine learning models behind One Drop’s Predictive Insights. Currently, users can receive eight-hour glucose forecasts, along with real-time advice for exercise, diet, and lifestyle adjustments.
 
“The real power of our predictive models is how individualized they are,” said One Drop VP of data science, Dan Goldner. “We don’t bother looking for broad patterns of factors that might help in distinct groups. Instead, we jump straight to predicting the factors most likely to help you. We update that prediction every day and use that insight to steer your personal One Drop experience.”
 
Treatment goals for diabetes and related conditions, including hypertension and prediabetes, are defined as reductions in average blood glucose concentration, blood pressure, and weight. Accurately predicting these metrics can enable the delivery of timely treatment and lifestyle recommendations, leading to improved outcomes.
 
In a recent study, One Drop trained a suite of patent-pending supervised learning models to accurately forecast changes in glucose, blood pressure, and weight up to six months in advance. Sources for the study included health and self­care data from a sample of approximately 55,000 users.¹ 
 
Simultaneously, One Drop expanded its predictive capabilities for people using continuous blood glucose monitors (CGM) with new overnight hypoglycemia predictions. The patent-pending overnight hypoglycemia risk algorithms, trained on over 500,000 nights of data from app users, are up to 87% accurate.²
 
“A reactive, symptom-focused approach to health is unsustainable,” said One Drop CEO and founder, Jeff Dachis. “We cannot expect people living with chronic conditions to spend more and more on healthcare, year after year, but get less and less in return. We must enable everyone to become active participants in their health, not just recipients of services and advice. One Drop is leading the paradigm shift in health management from reactive to prospective, with our new short- and long-term predictions for diabetes-related biomarkers laying the foundation for affordable and highly personalized self-care products.”
 
One Drop expects to release new predictions to subscribers within the year.
 
References:
¹ Wexler Y,  Goldner D, Merchant G. Hirsch A, Huddleston B, Dachis. One- to Six-Month Outcomes Forecasts for Diabetes and Related Conditions. Poster presented at: The 80th Annual American Diabetes Association Scientific Sessions; June 12, 2020; Virtual.
² Wexler Y,  Goldner D, Merchant G. Hirsch A, Huddleston B, Dachis. One- to Six-Month Outcomes Forecasts for Diabetes and Related Conditions. Overnight Hypoglycemia Prediction for CGM Users. Poster presented at: The 80th Annual American Diabetes Association Scientific Sessions; June 12, 2020; Virtual.