How your smartphone camera could detect diabetes

Despite the extensive functionality of modern smartphones, manufacturers for some reason have given all the health tracking tools to smart watches and fitness bracelets. Even those who do not produce anything other than smartphones on their own, implicitly recognize that wearable electronics have the right to a monopoly on tracking the health of users. The only company that decided to change the established practice was Samsung. She began to equip her smartphones with a heart rate monitor, but then she safely forgot about it. Meanwhile, it was possible to develop the topic of health care further.

Smartphone and smart watch

You can even diagnose diabetes using your smartphone

The camera of modern smartphones can be used to determine diabetes. To this conclusion, scientists from the University of California at San Francisco came in the course of multiple tests. Their experiment showed that an ordinary photo-module and a flash, not equipped with an additional body kit, provide the diagnosis of this disease with an accuracy of 80%. This is a good indicator, given that all calculations and diagnostic procedures are carried out not in the laboratory, but at home.

How to define diabetes

Blood oxygen measurement

Diabetes can be diagnosed in the same way that blood oxygen is measured.

The method proposed by scientists for the detection of diabetes is based on photoplethysmography, or, more simply, on the transmission of tissues using a light emitter to determine the characteristics of the blood flow. A flash is used as an emitter, and a camera matrix is ​​used as a scanner. Thus, for example, the level of oxygen in the blood is determined. But if, in the case of blood oxygen saturation, the hemoglobin content is estimated (oxygenated blood is darker than unsaturated blood), then when diagnosing diabetes, sensors will assess the degree of vascular damage caused by diabetes.

In order for smartphones to learn to recognize vascular damage, scientists had to test more than 50,000 experimental subjects suffering from diabetes. In total, they took 2.6 million measurements, and then provided the obtained data to the neural network, which on their basis began to self-learn. Based on the results of self-learning, the algorithms learned to detect the disease efficiently enough so that the technique could be used in everyday life. Indeed, despite the fact that it is inferior to invasive methods in accuracy, it surpasses them many times in painlessness and convenience.

Determine diabetes without puncture

Blood glucose meter

Diagnosing diabetes with a smartphone camera is completely non-invasive

Diabetes can be asymptomatic for a long time, making diagnosis much more difficult in the early stages of the disease. There is a clear shortage of non-invasive and widely used diabetes detection tools in the industry today, which prompted us to develop this method. We have learned from our own experience that the effectiveness of our diagnostic method is comparable to other widely used tests, such as mammography for breast cancer and cervical cytology for cervical cancer, ”explained Jeffrey Algin, co-author of the study.

Despite the fact that the method for diagnosing diabetes, developed by scientists at the University of California at San Francisco, looks quite simple and effective, it is unlikely to be used in the near future. First, it must undergo clinical trials, and then receive approval from regulatory authorities. After all, we are talking about a diagnostic procedure that aims to save lives. After all, if the technique is not effective enough, it can mislead the patient with a false result and only make him worse.

Rate article
Everything for Android and not only | Tips, instructions, root, news and app reviews.
Add a comment