Beyond the Average: Notes on the Future of Medicine

Four announcements crossed our desks recently. On the surface they are unrelated. Together, they hint at the future of medicine. Allow us to philosophize a little.

Sometimes it pays to climb out of your own rabbit hole and look around to see where you fit in the landscape. These four news items made us do exactly that.

Beyond BMI. Beyond AHI. Beyond the daily average.

Four announcements, one theme

• Google showed that a smartphone photo can estimate the body fat driving metabolic risk, including visceral fat, and predict insulin resistance almost as well as a DXA scan. Google calls it “seeing beyond BMI.”
• Cleveland Clinic and IBM trained AI on more than 10,000 sleep studies. AHI, the number used to grade sleep apnea for decades, failed to predict who actually dies. The raw data knew. The summary number didn’t. We cover the study in Why AHI Fails to Predict Sleep Apnea Risk.
• Sixth Lab CEO Aniket Lunagariya tested two wearables on the same body on the same night. Their “recovery” scores differed by more than 20 points, and both claimed to be right.
• Clarkson Law Firm filed a proposed class action against Oura in August 2026, alleging false advertising about the accuracy of its sleep tracking. Oura disputes the allegations.

Different companies, different diseases and different technology, with one theme running through all of them.

A century of summary numbers

For a century, medicine has run on crude summary numbers: BMI for metabolic health, AHI for sleep apnea, and many more. Those numbers did a job. They also threw away most of the information in the underlying data, because that was the best anyone could do at the time. Sadly, wearables have adopted the same habit without needing to.

Today we can extract the rich, moment-to-moment structure of human physiology cheaply, from devices people already own, and use it to spot disease years before it shows itself. Insulin resistance silently damages the body long before blood sugar looks abnormal. Sleep-disordered breathing fragments nights, strains the heart and harms the brain long before anyone orders a sleep study. The damage is written in the signals we used to discard.

Medicine 3.0

This is what some call Medicine 3.0: proactive, personalized and accessible detection of disease while it is still forming, in place of reactive treatment of end-stage disease. Universal screening that catches problems while there is still time to change course.

Where we fit

Standing back to look at the changing landscape gave us a clearer view of where our own team fits. Heart-rate wearables climbed from heart rate to heart-rate variability to AFib detection. We are doing the same for breathing, specifically the microstructure of breathing control itself: the fine variability that once-a-day numbers hide. A phone becomes a breathing screening tool, because one of the most consequential sleep and cardiometabolic risk factors should not require a referral or a trip to a lab.

Google is doing it for metabolic risk. Cleveland Clinic is doing it for sleep. We are doing it for the breathing control system underlying both. Beyond BMI. Beyond AHI. Beyond the daily or nightly average.

The era of crude summary numbers is ending. What comes next is the richest picture of human health we have ever been able to see, available to everyone. That is the future worth building.

Key takeaways

What do BMI and AHI have in common? Both reduce a complex body signal to a single summary number, which throws away most of the underlying information.

What is Medicine 3.0? A shift from treating disease at its end stage toward detecting it early, using rich data from everyday devices.

How does BreatheSimple fit in? We measure the moment-to-moment structure of breathing control, the detail that daily averages hide, using a phone you already own.

Where BreatheSimple fits

BreatheSimple’s free screening uses your phone’s microphone to give you a Breathing Control Index in a few minutes. It is a wellness and screening tool, not a medical device, and it does not diagnose or treat any condition. Join the waitlist to be among the first to try it.