Medical adherence is an enormous challenge for the healthcare industry, with a failure to take the correct medicine at the correct time believed to cost in the region of $300bn a year, and approximately 120,000 deaths.

Throw into the mix the damaging impact non-adherence can have on the clinical trial process and you have a situation that needs urgent remedy.

Low-tech vs high-tech

I’ve written previously about a relatively low technology approach to improving things, with a study that was published in the Annual Review of Public Health exploring the use of text messaging to improve adherence.

It found that a simple text message reminder can be a great way to prod us into taking the right medication at the right time.  Suffice to say, this does still require an element of trust as doctors still have little evidence that the patient has done what they claimed.

This is where AI based startup AI Cure, who presented at the recent Re:Work Deep Learning in Healthcare Summit, aims to improve matters.  They combine machine learning with smartphone technology to both remind people to take their medicine, and also try and prove that they did so.

The technology sends the patient a reminder, and then requests that they use the camera built into their phone to video themselves taking the medicine.

The machine learning then kicks into gear, attempting to recognize that the person in the video is the patient, and then to identify the pill in the mouth of the patient to prove that they have taken their medicine.

This successful (or unsuccessful) identification is then communicated with the patients doctor, with an analytics dashboard available to both parties charting the adherence to their medical plan.

Suffice to say, it does require the patient to have a smartphone in order to take footage, but it’s a potentially nice answer to a significant challenge for both the healthcare and medical research industries.

Check out the video below to see AI Cure in action.