Open AccessPrediction of obesity, type 2 diabetes mellitus, metabolic syndrome and coronary heart disease using Backpropagation algorithm
Author(s) -
Widiyawati,
A Nurrahman,
Budi Darmawan,
Agus Indra Jaya,
Rina Ratianingsih
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1218/1/012039
Subject(s) - obesity , medicine , disease , diabetes mellitus , metabolic syndrome , type 2 diabetes mellitus , type 2 diabetes , risk factor , stroke (engine) , backpropagation , coronary heart disease , cardiology , algorithm , artificial neural network , endocrinology , machine learning , computer science , mechanical engineering , engineering
Along with the development of increasingly modern life styles, the risk of community healthiness to gain weighted is rising due to the positive and negative impact of social interactions. Many negative impacts of social interaction tend the people of such community member being obese, in other way some positive impacts make the possibility of the obese people being normal weighted. This study discusses the prognosis obesity, metabolic syndrome MS), type 2 diabetes mellitus (DM) and coronary heart disease (CHD) people in point view of some risk factor of it. Obesity is the main stimulator for MS, Type 2 DM and hypertension. These diseases are the main factors for the cardiovascular disease. MS is a metabolic disorder that mostly suffered by obese and raises the risk of heart disease or stroke three times higher. Moreover, it raises the risk of type 2 DM five times higher. The type 2 DM also becomes as the one of the major health problems that causes vascular diseases such as coronary heart disease. The disease is the world’s leading cause of death. A prognostic detection of obese, DM, SM and CHD architecture is constructed in this paper and Backpropagation is proposed to derive the prediction model. The datasets taken from Palu Anutapura Hospital’s patients is trained to get the model. After that the model is tested until having a high accurate prediction. The accuracy of the detection to predict DM or MS is 93.33%, while the accuracy to predict CHD is 90.67%. The accuracy is obtained by identify the variation of the learning rate and hidden layer and got the best value of them are α = 0.5 and 5 hidden layers for not greater than 0.001 and maximum epoch 1000 of error target.