Machine learning: The technology exists that can cut cost of healthcare in U.S.

Posted: January 19, 2015 in Uncategorized
Tags: , , ,


So what’s being done with all the data we record today with our wearable technology? From here, it doesn’t seem like it’s enough just to be able to glance at a readout from time to time to check calorie-burn or blood pressure. Numbers are just numbers. To be valuable, the data must be transformed into actionable information.

One of the problems is that today’s wearables cannot transmit that data in compliance with the patient privacy required by government HIPAA regulations. And this is too bad because the technology exists that would allow the application of this data to drive down a least some of the onerous costs of chronic diseases. Think type-2 diabetes and some cancers. We’re familiar with the depressing numbers. According to the American Diabetes Association, for example,, this disease accounts today for one out of every five dollars spent on health care in America. Between 2007 and 2012, cost of treatment rose 41% to $245 billion. Diabetes is just one example.

So what do do? For starters, we could develop and embed those ubiquitous wearables with the firmware that would enable a secure, virtual network compliant with HIPAA rules. This network could transmit the readings through a secure wireless link to the wearer’s smartphone and then on to a back-end cloud where, utlizing machine learning, the data could be analyzed to determine the inflection points in the monitored data. Then this could signal the onset of trouble such as as a critical reading in a diabetic’s glucometer. Receiving an alert in real-time, the appropriate clinician could take steps immediately to mitigate any complications.

The per-patient cost savings of reduced diabetic complications amount from $67 to $105 per month in the U.S. for the non-Medicare population and $99 to $158 per month for Medicare patients. Beyond diabetes, cancer treatment presents another opportunity to drive down costs by enabling chemo pumps to collect and transmit important patient data between the patient’s home and hospital.

So, for a for a major HMO such as Kaiser with nine million subscribers and an estimated 20% of them suffering from diabetes, this kind of secure network and back-end cloud employing machine learning algorithms could represent savings of $180M per month in the treatment of complications.  That’s a sizable dent in the cost of treating and managing chronic disease for a single HMO.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s