Using Big Data To Predict Your Potential Heart Problems

The doctor only measures your heart rate when you’re sick, but a new website aims to create a database of every normal beating heart in the world, so we can find out more about how the heart works.

It’s safe to say that Dr. Leslie Saxon, the founder of the USC Center for Body Computing, is more tech savvy than most doctors. Not only has she created a space at USC for academics, entrepreneurs, and venture capitalists to work on the future of wireless health, she also recently completed a study with AliveCor (maker of the iPhone ECG device) showing that wireless, ubiquitous heart rate monitoring—say, with the AliveCor iPhone case—can detect all sorts of heart conditions (note: I was part of this study).

Now Saxon is taking on another ambitious challenge: everyheartbeat, a website that will allow people to log their heart rate data using any available sensor—the iPhone light, the AliveCor iPhone case, or anything else that people have access to. The site, which intended to be a place for people to continuously monitor their health, will record and analyze all heartbeat data that comes in to find global patterns and even warn people of potential heart issues. Saxon’s ultimate goal is to record every heartbeat in the world.

"I’ve been thinking for awhile about how to realize the potential of continuous wireless communication as it relates to health," explains Saxon. "I came up with this concept because we have some traction here—a relatively complete understanding of wearable sensors and implantable devices."

Doctors have lots of experience measuring heart rates and rhythms during and after serious events—during a heart attack, for example, or in patients who have longstanding heart disease. But no one has ever been able to observe what heart rhythms are like on a continuous basis in the general population.

There are all sorts of abnormalities that everyheartbeat could pick up with relatively simple algorithms. Atrial fibrillation, a non-life threatening heart rhythm, often produces no symptoms. But up to 20% of all strokes are caused by the condition. A platform like everyheartbeat could alert patients of the need to see a doctor now, before a dangerous event takes place.

"I think we’re going to be able to make unbelievably predictive analytics across populations," says Saxon.

It’s not just personal health that will be impacted by everyheartbeat; doctors will be able to look at anonymized heart rhythm data on an unprecedented scale. "There will be some things that I’ll see and I’ll say, 'Wow, I wonder if that’s normal because nobody has ever recorded it before,'" she says. People looking at the everyheartbeat data will undoubtedly come across similar scenarios.

Saxon hopes that everyheartbeat will start collecting data by 2013.

Add New Comment


  • John Joyce

    Don't look now but Dr. Michael Shen has amassed a database of advanced ekgs correlated with  coronary angiographic studies of the same patients permitting 90% accuracy in cardiac disease burden description for new patient ekgs submitted for analysis. The device and system involved is the MCG Multifunction Cardiogram. john.joyce127@gmail.com

  • Bob Cable

    The funding for implementation of such technology shoud be from the savings to the medical care insurance funds. This savings can be quantified. Pick unedited zip code health care members of a target group. provide or install devices on all adult members. fund this through revenue bond type instruments. Use the claim history of the group , 5 year average, to determine the savings component. Use 40% of the savings to amortize the bonds.  Self funding for all medical funding companies
    Bob Cable, tvcable@aol.com