Common Disease Forecaster

Our technology forecasts future risks of 28 common diseases for individuals who are between 45-75 years old. We integrate routine clinical measurements that you receive at your annual physical such as blood pressure, cholesterol, blood counts, etc., and, if you provide a whole genome sequencing file (VCF), we also integrate your genetic information. By giving you an early indication of elevated disease risk before symptoms are clinically apparent, we hope to help you take action to improve your health.

Our technology is prognostic, not diagnostic. We algorithmically predict future disease onset within the next 10 years of your life from the information gathered at your doctor visit. We developed this technology from the UK Biobank dataset, which is the largest and most comprehensive biobank in the world.

In the images below, we externally validate our predictive performance on an independent test set of 20,000 families that were completely withheld from all technology development. We can comprehensively test the accuracy of our 10-year risk predictions on these individuals as their biomarkers were all measured more than 10 years ago and we have their complete diagnosis records.

We are confident that our technology is capturing direct biological effects that influence disease risk. The gold-standard test is whether the technology can predict disease between members of the same family. Suppose two siblings in their 60s, e.g., brothers Sam and Frank, go to their annual physical and receive their basic physical measurements and standard bloodwork. One of them goes on to develop coronary artery disease within the next 10 years, the other does not. From the information gathered at their annual physical, can we consistently predict which sibling is more likely to develop coronary artery disease, better than random guessing? We show across 28 diseases that our technology is able to do this.