Özet:

Abstract Cardiovascular disease is a significant global health concern. A greater number of fatalities are seen at the first occurrence of a heart attack compared to other instances in the human population. However, its impact extends beyond heart attacks to include conditions such as breast cancer, lung cancer, some ventricular issues, and other related ailments. Heart failure occurs when the cardiac muscle is unable to adequately pump a sufficient volume of blood to meet the physiological needs of the body. It is possible to quantify symptoms, physical traits, and test results using the patient's computerized medical information that are readily available. And perform biometric analyzes designed to highlight patterns and correlations that physicians cannot detect. However, the existing system has some limitations. Healthcare plans should prioritize disease control efforts to reduce hospitalizations and mortality in individuals with heart failure. To forecast risk, we created a risk model. A patient's risk of death or hospitalization from heart failure using convolutional neural network algorithms (CNN) in a large health maintenance organization. The current potential of using deep learning algorithms in the early detection of heart disease. The primary objective of this research is to assess the accuracy of diagnosing a cardiac condition in individuals. In the recursive process of partitioning, the reordering of partitions is carried out in a greedy manner rather than seeking the optimal partition order. The suggested system utilizes a CNN to process input datasets for illness prediction. The system incorporates preprocessing, feature extraction, and classification techniques to analyze the data and provide relevant findings. Dimensionality reduction allows us to make more precise forecasts using the same data set. The majority of the time, many existing algorithms utilized in AI classifiers do not discover greater accuracy than Lasso or Ridge regression, which both produce superior results.

Anahtar Kelimeler: