Related Experiment Videos
[Computer program for diagnosis, monitoring, and prophylaxis of cardiovascular diseases]
Insights
This study developed software to automatically identify cardiovascular disease risk factors and predict patient outcomes. The tool aids lipid centers in early detection and prevention strategies for coronary heart disease.
Area of Science:
- Cardiology
- Medical Informatics
- Artificial Intelligence
Background:
- Cardiovascular diseases are a leading cause of adult mortality.
- Effective primary and secondary prevention are crucial for reducing incidence.
- Automated tools can enhance risk assessment in clinical settings.
Purpose of the Study:
- To develop software for automated recognition of cardiovascular risk factors.
- To enable automated prognosis and evaluation of dyslipoproteinemia phenotypes.
- To support lipid centers in managing cardiovascular health.
Main Methods:
- Utilized a neuronal network for analysis.
- Input vectors included lipid and lipoprotein concentrations.
- Output neurons provided diagnostic signs for dyslipoproteinemia and prognosis.
Main Results:
- The software estimates prognostic indexes and lipidogram values.
- It detects risk factors like overweight, poor diet, and smoking.
- It predicts the overall risk of coronary heart disease.
Conclusions:
- The developed software can automate cardiovascular risk factor identification and prognosis.
- It aids in evaluating dyslipoproteinemia phenotypes.
- This technology can improve the management of cardiovascular diseases in specialized centers.
Abstract:
Cardiovascular diseases remain the main cause of death among adult population; their incidence can be reduced only using an effective system of primary and secondary prevention. The aim of our study was to develop software for automated recognition of cardiovascular risk factors, prognosis, and evaluation of dyslipoproteinemia phenotypes in lipid centers. Prognostic indexes, lipidogram values are estimated, and the program detects the risk factors (overweight, poor physical training, irrational nutrition, tobacco smoking, etc.) and predicts the summary risk of coronary disease. Neuronal network was used for evaluating the types of dyslipoproteinemias and for cardiovascular prognosis. The input vectors were lipid and lipoprotein concentrations. Output neurons were diagnostic signs of dislipoproteinemia phenotypes or cardiovascular prognosis.