AP13068289 Application of machine learning methods for early diagnosis of pathologies of the cardiovascular system

The aim of the project is to develop a fully functional prototype of a software and hardware platform for diagnosing pathologies of the cardiovascular system based on the analysis of electrocardiograms using machine learning methods and algorithms.

Cardiovascular diseases (CVD) remain as one of the leading causes of death worldwide. In most of the developed European countries, this indicator reaches 40% of all deaths, and in Kazakhstan - more than 50% [1-3].

According to the latest data, in Kazakhstan, most people die from diseases of the circulatory system - 24.44% of all deaths, in second place - respiratory diseases - 12.89%, third place - from neoplasms - 12% [4-5]. The main part of this paper is the study of digital monitoring methods and the development of a hardware-software complex for early detection of CVD using machine learning algorithms, thereby making a significant contribution to reducing mortality from CVD.

The high sensitivity of electronic stethoscopes, compared to a conventional stethoscope, allows us to use them for screening obstructive coronary artery disease. Conventional stethoscopes lack auscultation ability to detect intracoronary murmur of turbulent blood flow arising from hemodynamically significant coronary artery disease. According to our research, the sensitivity of electronic stethoscopes has grown significantly. At this moment, an electronic stethoscope has a higher sensitivity than an acoustic stethoscope, both for a cardiologist and for a patient in the analysis.

The introduction of mathematical methods of data analysis has significantly expanded the capabilities of phonocardiography (PCG) and electrocardiography (ECG). Such advantages as non-invasiveness, safety, the absence of contraindications, relatively inexpensive equipment, create the prerequisites for the use of PCG and ECG in telemedicine. Machine learning will play an integral role in this, as a tool that allows us to find samples in the data generated by diagnostic tests for cardiovascular diseases.