Framework Two-Tier Feature Selection on the Intelligence System Model for Detecting Coronary Heart Disease

Abstract

Coronary heart disease is a non-communicable disease with high mortality. A good action to anticipate this is to do prevention, namely by carrying out a healthy lifestyle and routine early examinations. Early detection of coronary heart disease requires a number of examinations, such as demographics, ECG, laboratory, symptoms, and even angiography. The number of inspection parameters in the context of early detection will have an impact on the time and costs that must be incurred. Selection of the right and important inspection parameters will save time and costs. This study proposes an intelligence system model for the detection of coronary heart disease by using a minimal examination attribute, with performance in the good category. This research method is divided into a number of stages, namely data normalization, feature selection, classification, and performance analysis. Feature selection uses a Two-tier feature selection framework consisting of correlation-based filters and wrappers. The system model is tested using a number of datasets, and classification algorithms. The test results show that the proposed two-tier feature selection framework is able to reduce the highest attribute of 73.51% in the z-Alizadeh Sani dataset. The performance of the system using the bagging-PART algorithm is able to provide the best performance with parameters area under the curve (AUC) 95.4%, sensitivity 95.9% while accuracy is 94.1%. Referring to the AUC value, the proposed system model is included in the good category.

Use of Artificial Bee Swarm Optimization (ABSO) for Feature Selection in System Diagnosis for Coronary Heart Disease

“…Table 6 also shows the same, which requiring six features that include both investigations. These results are supported by several previous studies [16,17,23].…”

“…In this dataset, the feature selection process appears to eliminate the high-cost inspection variables immediately. In the research of Wiharto et al [15], performance was not significantly different from that of Wiharto et al [16], who used a stepwise greedy combination with Best First Search (BFS). This model can provide an AUC of 95.4% with a few features but at a much higher cost.…”

“…Many studies have been conducted on the use of feature selection in the diagnosis system for coronary heart disease. The feature selection methods used were genetic algorithms, particle swarm optimization, fast correlation-based filter (FCBF) [29], and greedy algorithms [16]. The proposed feature selection model can provide a relatively better performance than those in a number of previous studies.…”