< br />
MOSCOW, 16 Aug. The mathematical model of SSMU scientists will allow to reduce by half the number of errors in diagnosing cardiovascular diseases. The results are published in The European Physical Journal Special Topics.
The heart of a healthy person beats irregularly, that is, the time of heart contractions can be different, the scientists explained. At the same time, in people with diseases of the heart and blood vessels, such as hypertension, the heart beats much more evenly than in healthy people.
According to experts, the degree of heart rhythm irregularity can be used for medical diagnosis. Most often, a spectral analysis of the electrical activity of the heart, obtained by the method of electrocardiography (ECG), is used for this. //ria.ru/20230731/nauka-1886911712.html» data->
However, building accurate spectra of biological systems is an extremely difficult task, so all diagnostic methods involve an error, the specific value of which is known only very approximately, the scientists noted.
Researchers of the Saratov State Medical University named after. IN AND. Razumovsky (SSMU) created a mathematical model for accurate calculation of the error in assessing the spectra of a biological signal. According to them, the use of the obtained data will make it possible to halve the number of incorrect cardiovascular diagnoses.
“The accepted diagnostic methods, for example, hypertension, as it turned out, give the correct diagnosis only in 70% of cases. The main reason for the high error in that both world and Russian standards recommend short electrocardiograms of 3–5 minutes for analysis. Our calculations have shown that ECG data for 15–20 minutes will help raise this value to 85%,” said Yury Ishbulatov.
The proposed model, according to the creators, reliably imitates the biological signal, but has a pre-known spectrum. The study made it possible not only to determine the error of the existing methods of spectral analysis in medicine, but also to improve their mathematical implementation, the scientists explained.
«Our recommendation to increase the length of recorded signals to 15-20 can be relatively easily implemented using wearable devices, such as fitness bracelets or smart watches that can write the necessary signals for days,» Ishbulatov emphasized.
The SSMU team has made the developed model publicly available so that any medical specialist can evaluate the accuracy of spectral analysis for diagnosing a pathology of interest.
< br />In the future, the research team intends to develop mathematical models for assessing errors in the analysis of long-term ECG signals using wearable electronics and use the data obtained to improve the efficiency of spectral analysis during continuous monitoring of patients' condition.