Analysis of Cardiac Disease based on AI Prediction Techniques using data analytics approach

Patients who have an increased likelihood of coronary heart disease can reduce their consequences by changing their lifestyle, with the support of early diagnosis. Healthcare expenses are rising above, both company budgets and the average cost of medical treatment nationally, due to asymptomatic conditions like cardiovascular disorders. Early detection and treatment of these illnesses are essential. Utilizing the data that is accessible in healthcare systems has been significantly impacted by recent technological advancements. As part of this, patients' Heart Disease (HD) may be predicted using the massive amount of data. However, due to concerns with prediction and diagnosis performance and accuracy, projecting HD has emerged as one of the main obstacles facing the healthcare sector. The main challenges were to apply homogeneous data mining techniques to define the most significant causes of heart disease and to accurately predict the overall risk. Using Data Analytics (DA) to speed up the prediction process is one way to make use of massive amounts of health care data. Selecting the optimal analytical methods, however, is the most crucial step as it will impact the final result of the diagnostic and forecasting. In order to assess the effectiveness of the methods for HD forecasting, this study intends to implement DA approach in collecting of databases. In this research work, Best First Search (BFS) has been used to identify the most important attributes for HD prediction. This study employs a range of ML approaches based on AI to predict HD. The Random Forest (RF) method, which uses specific characteristics, has the highest accuracy rate (98.63%) in this experiment. The BFS extracted 14 relevant features and then compared them using AI approaches. Based on the analysis, feature selection has enhanced the DA technique by improving the efficiency and precision of the classification model's prediction.