No abstract

The use of data mining approaches in medicine and medical science has become necessary especially with the evolution of these approaches and their contributions medical decision support. Coronary artery disease (CAD) touches millions of people all over the world including a major portion in Algeria. However, much advancement has been done in medical science, but the early detection of CAD is still a challenge for prevention. Although, the early detection of CAD is a prevention challenge for clinicians. The subject of this paper is to propose new clinical decision support system (CDSS) for evaluating risk of CAD called CADSS. In this paper, the authors describe the characteristics of clinical decision support systems CDSSs for the diagnosis of CAD. The aim of this study is to explain the clinical contribution of CDSSs for medical decision-making and compare data mining techniques used for their implementation. Then, they describe their new fuzzy logic-based approach for detecting CAD at an early stage. Rules were extracted using a data mining technique and validated by experts, and the fuzzy expert system was used to handle the uncertainty present in the medical field. This work presents the main risk factors responsible for CAD and presents the designed CASS. The developed CADSS leads to 94.05% of accuracy, and its effectiveness was compared with different CDSS.

Feature Selection (FS)

Industry 4.0 is characterized by the availability of sensors to operate the so-called intelligent factory. Predictive maintenance, in particular, failure prediction, is an important issue to cut the costs associated with production breaks. We studied more than 40 publications on predictive maintenance. We point out that they focus on various machine learning algorithms rather than on the selection of suitable datasets. In fact, most publications consider a single, usually non-public, benchmark. More benchmarks are needed to design and test the generality of the proposed approaches. This paper is the first to define the requirements on these benchmarks. It highlights that there are only two benchmarks that can be used for supervised learning among the six publicly available ones we found in the literature. We also illustrate how such a benchmark can be used with deep learning to successfully train and evaluate a failure prediction model. We raise several perspectives for research.

Feature Selection (FS) has become the motivation of much research on decision support systems areas for which datasets with large number of features are analyzed. This paper presents a new method for the diagnosis of Coronary Artery Diseases (CAD) founded on Genetic Algorithm (GA) wrapper Bayes Naïve (BN). Initially, thirteen attributes were involved in predicting CAD. In GA–BN algorithm, GA produces in each iteration a subset of attributes that will be evaluated using the BN in the second step of the selection procedure. The final result set of attribute holds the most pertinent feature model that increases the accuracy. The accuracy results showed that the algorithm produces 85.50% classification accuracy in the diagnosis of CAD. Therefore, the strength of the Algorithm is then compared with other machine learning algorithms such as Support Vector Machine (SVM), Multi-Layer Perceptron (MLP) and C4.5 decision tree Algorithm. The result of classification accuracy for those algorithms are respectively 83.5%, 83.16% and 80.85%. Then, the GA wrapper BN Algorithm is similarly compared with other FS algorithms. The Obtained results have shown very favorable outcomes for the diagnosis of CAD.