Algorithm To Detect Diabetes With Regular Smartphone Cameras

50% of the people worldwide who have the condition are completely unaware of it.
Deniz Yildiran
The photo credit line may appear like this1,2

Lately, we stopped using cellphones merely for calls and switched our attention to texting, taking photos, checking up on social media, and so on. They have become little portable computers after all. 

So it is only natural to try to put them to good use as such. This time for health-related issues.

Last year, it was one of those new discoveries that a smartphone could detect bioagents that cause anthrax and melioidosis, according to a study conducted by the researchers from the University of Arizona and University of Nevada, Reno. In a sense, involving smartphones in diagnosis processes is not that new.


And now, researchers from the University of California, San Francisco developed a technique to detect type 2 diabetes. The method reportedly needs no additional hardware but a smartphone camera. 

The research was published in the journal Nature Medicine.

Able to detect both sides

The camera was successful in diagnosing 81% of the subjects accurately. And that's quite a number.

Basically, researchers used a smartphone camera to detect vascular damage from diabetes by photoplethysmography (PPG). PPG is a technique that identifies volumetric changes in blood circulation. The camera of a smartphone along with its flashlight was used on the fingertip to detect color changes corresponding to heartbeats and measure PPG.

Before drawing on randomized human subjects, old PPG data of confirmed diabetics were firstly revised to see if the algorithm could really detect diabetes. Out of 3 million recordings from 53,870 patients in the Health eHeart Study who used the Azumio Instant Heart Rate app on the iPhone, the predictions were mostly correct.

Results of 92 to 97 percent of patients without diabetes were correct as well. In this case, the algorithm passed the crosscheck successfully.

A faster and shorter way

Conventional ways to test diabetes can take a long time if we consider the amount of time for fasting. In one of the methods, the blood sample is obtained after an overnight fast, and the results might not come out in a short time, depending on where you get tested.

“We demonstrated that the algorithm’s performance is comparable to other commonly used tests, such as mammography for breast cancer or cervical cytology for cervical cancer, and its painlessness makes it attractive for repeated testing,” added, author of the study Jeffrey Olgin.


Subscribe today

For full access to all features
and product updates.

%30 Save Quarterly




Subscribe Now
You can cancel anytime.
View Other Options

Already have an account? Log in

0 Comment
Already have an account? Log in