Wearable gadgets can recognize infections, type 2 diabetes and other diseases.

Michael Snyder, a professor at the Medical School at Stanford University, the lead author of the scientific work on fixing abnormalities in human physiological parameters using wearable electronics and correlations with diseases, demonstrates wearable electronics

With the advent of fitness trackers and other wearable electronics, the idea immediately emerged that these gadgets can be used to monitor health. After all, it is known that many physiological parameters provide the most important information about human health. For example, heart rate, blood pressure, body temperature.

Wearable gadgets constantly monitor vital signs in a passive mode, sending the results to the user's computer or the attending physician, without requiring special care or prior training before use.
')
Specialists from the Medical School at Stanford University conducted a large-scale study to determine how useful the measurement results are in practice. In all cases, used devices that are in the open market.

Previously, it was possible to measure the physiological parameters only in the doctor’s office, which was a problem - changes in the indicators were often detected too late, because a “healthy” person rarely visits the doctor for a routine check. In addition, physiological parameters vary greatly depending on age, gender, fitness level and other factors. The doctor in the study of indicators of a particular patient usually compares them with the average indicators in the population, and this is not always an adequate assessment due to the large variation of individual indicators.

As a result, too long periods between measurements make it very difficult to determine changes in the health status of a particular person. To receive constant information about these indicators would be a great help for the attending physician.

In the process of research at Stanford University, about 2 billion measurements from 60 people were collected. This is a continuous stream of data from wearable biosensors and periodic analyzes in the medical laboratory. The laboratory took a blood test, checked the expression of genes and did other more advanced tests to identify the symptoms of the disease.



An important task of the study was to determine the "baseline" of normal physiological parameters for each individual person - and determine how deviations from this baseline level are determined using wearable electronics and how these deviations correlate with real diseases of people.

Each of the participants in the experiment constantly wore from 1 to 8 trackers. Doctors recorded changes in patient weight, heart rate, oxygen level in the blood, skin temperature and activity of people, including sleep, walking, running and cycling, wasted calories and even gamma and X-ray radiation in normal life.


The level of exposure of one of the patients in the 25-day period. Violet color (lower columns) indicates the time spent in flight. Stars - scanning carry-on baggage at check-in at airports

In general, a very large amount of data was collected.


Basic indicators of one of the patients

The study showed that after calculating the base level of physiological parameters, it is indeed possible to detect deviations from the base level. It is possible to establish the correlation of these deviations with external factors, including natural factors, diseases and other factors that affect health. The developed algorithm for identifying such template deviations and their explanations in the future will probably be used in medicine.

Interestingly, among the participants in the experiment was one of the authors of the scientific work - Professor Michael Snyder himself. Like other experimental subjects, there were seven biosensors on it during a flight from Norway, when the professor suddenly noticed unusual changes in his indicators . From previous flights, he knew that his pulse usually rises at the beginning of the flight, and the level of oxygen in the blood is lowered during the entire flight, but this time the indicators did not return to normal after landing. The professor saw that the indicators were substandard — and he was not completely surprised when he soon developed fever and other symptoms of an infectious disease.

Two weeks before, the professor had helped his brother build a fence in Massachusetts. Before the trip, he was very worried that in the countryside he could be bitten by a tick and hit Lyme disease, so he went to the doctor in advance and got a prescription for doxycycline, an antibiotic against Lyme disease. Now, after the onset of signs of infection, the doctor did tests and made sure that he was not mistaken - Lyme disease causative agents actually got into the body.

It is impressive how simple commercial biosensors have indicated the possibility of the onset of the disease even before the onset of the first symptoms.

The promising results of the correlation of early symptoms of the disease and deviations from normal indicators of physiological parameters, taken from biosensors, appeared in other patients during the study. For example, in several of them, elevated skin temperature and increased pulse correlated with an increased content of C-reactive protein in blood tests. C-reactive protein is a marker of the immune system as an indicator of inflammatory processes, as a result of which an infection, an autoimmune disease, cardiovascular disease or even a cancer can develop.


Changes in physiological parameters during infection

Wearable gadgets also revealed patients with insulin resistance - an indication of the risk of type 2 diabetes. Of the 20 patients who passed the test for glucose tolerance, 12 showed insulin resistance. The authors of the scientific work have compiled an algorithm that takes into account the number of steps taken per day, the daily heart rate and the difference between day and night heart rate, in order to predict with high probability which patient will not pass the glucose tolerance test.


The dependence of changes in body mass and pulse at rest on the number of steps taken per day, statistics from 43 patients

Such monitoring of one’s own health can be carried out on an ongoing basis by each person, and in the future they may well constitute a standard medical practice. “There are more sensors in our cars than on the human body,” says Professor Snyder. He considers such a situation abnormal and believes that in the future there will be the opposite: people will have more sensors than cars. Each person should know their basic indicators and instantly receive a signal if they deviate from the norm.

The scientific article was published on January 12, 2017 in the journal PLoS Biology .