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.

Early Detection of Alzheimer's Disease with the C4.5 Algorithm Based on BPSO (Binary Particle Swarm Optimization)

“…Binary Particle Swarm Optimization (BPSO) merupakan hasil modifikasi dari algoritma sebelumnya yaitu Particle Swarm Optimization (PSO) dimana algoritma ini menggunakan metode pendekatan meta-heuristik yang terinspirasi dari sekumpulan burung dan ikan [17]. Algoritma PSO dikembangkan menjadi algoritma Binary PSO (BPSO) bertujuan untuk memudahkan dalam melakukan seleksi fitur.…”

A hybrid approach to medical decision-making: diagnosis of heart disease with machine-learning model

“…We have employed iterative feature elimination for removing irrelevant features of heart disease dataset that does not have effect on the predictive outcome of the developed model to obtain good outcome on heart disease prediction. Iterative feature elimination removes irrelevant features that mislead the model's predictive capability and ultimately reduce the performance of classification model [20], [23]. Moreover, with reduced feature, the computational time required for model training and storage space requirement is optimized [22], [24].…”

“…As demonstrated in Table 5, the proposed model outperforms compared to the existing work. [5] 2019 NB, RF 86.81% Wan Hajarul [6] 2018 DT and RF 82.99% with RF Amin Ul Haq [8] 2018 SVM, DT, RF, NB, DT 86% with SVM Kathleen H. Miaoa [11] 2018 Deep neural network 83.67% Wiharto Wiharto [12] 2019 Ensemble classifier 88.33% Noor Basha [18] 2019 KNN, NB, SVM, DT 85%, with KNN Edsel Ing [19] 2019 SVM and LR 82.71% with LR Márcio Dias [20] 2020 SVM 87.71% Khaled Mohamad [21] 2020 SVM, NB 84.19% with SVM Pooja Rani [22] 2021 NB, LR, NB, SVM, RF 84.79% with SVM Suja Panicker [23] 2020 SVM 90% G. Magesh [24] 2020 RF 89.30% Ashir Javeed [25] 2020 Deep neural network 91.83% G. Saranya [ A hybrid approach to medical decision-making: diagnosis of heart disease … (Tamilarasi Suresh) 1837 5. CONCLUSION Automated intelligent approaches are crucial for timely prediction of heart disease.…”