The methods of duo output neural network ensemble for prediction of coronary heart disease

Abstract

The occurrence of Coronary heart disease (CHD) is hard to predict yet, but the assessment of CHD risk for the next ten years is possible. The prediction of coronary heart disease can be modelled using multi-layer perceptron neural network (MLP-ANN). Prediction model with MLP-ANN has either positive or negative CHD output, which is a binary classification. A prediction model with binary classification requires determination of threshold value before the classification process which increases the uncertainty in the classification process. Another weakness of the MLP-ANN model is the presence of overfitting. This study proposes a prediction model for coronary heart disease using the duo output artificial neural network ensemble (DOANNE) method to overcome the problems of overfitting and uncertainty of classification in MLP-ANN. This research method was divided into several stages, namely data acquisition, pre-processing, modelling into DOANNE, neural network ensemble training with Levenberg-Marquard (LM) algorithm, system performance testing, and evaluation. The results of the study showed that the use of DOANNE-LM method was able to provide a significant improvement from the MLP-ANN method, indicated by the results of statistical tests with pvalue <0.05.

Sequential feature selection for heart disease detection using random forest

“…The goal of the FR is to rank input features used for training the heart disease diagnosis ISSN: 0067-2904 model, ultimately improving the precision and accuracy of the model for heart disease diagnosis. Unrelated and redundant features in a real-world heart disease dataset propose relationships between irrelevant features and the target class rising by chance, and a strong relationship between irrelevant features and the target class tends to decrease the classification accuracy of a decision support model [2]. Furthermore, a larger number of input features results in substantially higher computational time complexity without resultant random forest model performance improvement.…”

Determination of the obesity prevalence and its associated factors among community in Selangor, Malaysia: an ordinal logistic regression approach

“…Most attractive methods for uncovering the associated factors in medical fields are machine learning algorithm [22][23][24]. Although machine learning algorithm has been applied successfully in medical fields, its advantage requires massive data which lead to long processing time.…”