Clinical decision support system: risk level prediction of heart disease using weighted fuzzy rules and decision tree rules

Abstract

The development of medical domain applications has been one of the most active research areas recently. One example of a medical domain application is a detection system for heart disease based on computer-aided diagnosis methods, where the data is obtained from some other sources and is evaluated by computer based applications. Up to now, computers have usually been used to build knowledge based clinical decision support systems which used the knowledge from medical experts, and transferring this knowledge into computer algorithms was done manually. This process is time consuming and really depends on the medical expert's opinion, which may be subjective. To handle this problem, machine learning techniques have been developed to gain knowledge automatically from examples or raw data. Here, a weighted fuzzy rule-based clinical decision support system (CDSS) is presented for the diagnosis of heart disease, automatically obtaining the knowledge from the patient's clinical data. The proposed clinical decision support system for risk prediction of heart patients consists of two phases, (1) automated approach for generation of weighted fuzzy rules and decision tree rules, and, (2) developing a fuzzy rule-based decision support system. In the first phase, we have used the mining technique, attribute selection and attribute weightage method to obtain the weighted fuzzy rules. Then, the fuzzy system is constructed in accordance with the weighted fuzzy rules and chosen attributes. Finally, the experimentation is carried out on the proposed system using the datasets obtained from the UCI repository and the performance of the system is compared with the neural network-based system utilizing accuracy, sensitivity and specificity.

A Study of Cardiovascular Disease Prediction Models Using Discriminant Analysis

“…In order to calculate these results we first calculated the True positive, True negative, False negative, and False positive values based on the definitions given in Table II(2) [9]. When only considering the accuracy of the model, both the FRS and our discriminant analysis had accuracy rates around 80%.…”

“…For the experiment, the 4th (2009) [9] and 5th (2010) Korean National Health and Nutrition Examination Surveys were used. As an experimental scenario, creation of the classification function coefficients to predict cardiovascular disease were prepared using the 5th KNHANES(2010) data set, and the accuracy of the classification in predictingcardiovascular disease was assessed through comparison of our model with the FRS Guidelines using the 4th (2009) and 5th (2010) data KNHANES data sets.…”

A robust smart device app assisting medical diagnosis

“…[7][8][9] Moreover, new methods are posed for comparing their performances by giving the necessary evaluation criteria. [8,10,11] Today advances in MDMS based on ANNs are radical and multifaceted. [4,5,7,8,[12][13][14][15] ANNs have many advantages such as improved speed factor, dynamic data storage, robustness, parallel searching, and generalization virtues so AI and medical experts utilize ANNs in MDMS implementations.…”

“…It is a requirement by the Clinical Differential Diagnosis Methodology (CDDM) to singly weight all MD, judge intermediate results and process all data from the more broad to the more explicit. This is why CDDM should be followed by any MDMS [7,11] and ANNs are the optimal implementation for CDDM. [5,17] In order to establish the problem domain boundaries, clinical experts in PDs, set a definite number of questions / inputs, the same way that they would be asked by MDs during patient examination.…”

Classification of coronary artery disease data sets by using a deep neural network

“…While conventional methods such as decision trees (1), naive Bayes (3), etc., have some speed benefits and easily applied to data sets, these methods cannot yield significant classification performance. Therefore, machine learning based classification methods, such as neural network classifiers and fuzzy classifiers (6,8,11), have been applied in recent years to classify the CAD data to improve the classification performance.…”