Smartphone diagnoses diabetes 82% accurately

Diabetes mellitus is the seventh most common cause of death in the world. Even more, people die from diseases that increase the risk of diabetes: patients with type 2 diabetes are twice as likely to die from heart disease as people without diabetes. At the same time, there is a problem with the early diagnosis of diabetes. The disease develops slowly and invisibly, so people may not know about the problem for a long time until they face serious and often irreversible complications. Scientists at the University of California have suggested a very simple and effective way to diagnose diabetes using an usual smartphone. 

The method developed by experts is based on a technology called photoplethysmography (PPG) - measuring the optical density of the tissue. The volume of each organ consists of its own tissue and the volume of blood that fills its vessels as the heart shrinks. The essence of the method of photoplethysmography is to "illuminate" the area of the body under study with infrared light - it is necessary that the wavelength of the light allows it to be absorbed by red blood cells. The reflected light signal hits the photo converter: its characteristics allow estimating the level of blood supply to the organ. 

The method of plethysmography underlies the work of fitness bracelets: sending and receiving the reflected signal, the device receives information about the nature of the pulse wave, blood filling of the wrist, which allows assessing the cardiovascular system. Experts from the University of California decided to use this technology to detect vascular disorders at the fingertips in case of diabetes. The researchers used a flashlight and a camera to measure PPG, capturing changes in the color of the fingertips with each heartbeat. 

In order to understand exactly how blood vessels at the fingertips change during diabetes mellitus, the researchers analyzed about three million PPG records received from 53,870 patients with diagnosed diabetes. The data were processed using a computer algorithm: the developed program for the smartphone has learned to diagnose the disease with high precision. Testing showed that the algorithm was able to correctly diagnose 82% of patients with photoplethysmography via smartphone.

The accuracy of determining the absence of diabetes was even higher, ranging from 92 to 97%. When the results of PPG were supplemented with information about the patients' field, age, weight, and body mass index, the accuracy of diagnosis increased even more. According to the authors, the simplicity and accessibility of the proposed diagnostic method will allow us to detect more cases of diabetes at an early stage and to seek help in time. 

Aug. 19, 2020, 10:26 a.m.

Similar articles:

AI-based fitness coach on your smartphone

Exer has created a training app that runs on artificial ...

An app for early diagnosis of eye diseases in diabetics

The affection of the retina vessels is one of the ...

The brand new app detects diseases by face

Israeli specialists have developed a technology that can recognize symptoms ...

Somryst app for the treatment of insomnia approved for use

American company Pear Therapeutics*, which creates high-tech medical solutions, including ...

An app can diagnosis COVID-19 by voice

Scientists at the University of Cambridge (England) have developed an ...

Artificial intelligence will detect heart disease by four selfies

Scientists from Tsinghua University, in collaboration with clinicians from Fuwai ...

The new app will help doctors avoid professional burnout and teach mindfulness

As coronavirus cases continue to rise in the world, health ...

The light implant restored the rats' hearing

Scientists at the University of Göttingen Medical Center in Germany ...

COOKIE

Our site collects information using cookies to be more convenient and customized to your needs interests. The purposes of the use of cookies are defined in Policy the processing of personal data .If you agree to continue to receive cookies, please click the "Accept" button. If you don't agree or want to resolve this issue later, please change your browser cookie settings.