Artificial Neural Networks Based Predictive Model for Detecting the Early-Stage Diabetes

https://doi.org/10.55529/jaimlnn.241.8

Authors

  • Shokhjakhon Abdufattokhov Automatic Control and Computer Engineering Department, Turin Polytechnic University in Tashkent, 17 Little Ring Road, Tashkent, Uzbekistan
  • Nodira Normatova Automatic Control and Computer Engineering Department, Turin Polytechnic University in Tashkent, 17 Little Ring Road, Tashkent, Uzbekistan
  • Makhbuba Shermatova Automatic Mathematical Natural Sciences Department, Turin Polytechnic University in Tashkent, 17 Little Ring Road, Tashkent, Uzbekistan

Keywords:

Artificial Neural Networks, Blood Glucose, Diabetes, Relief-Based Filter.

Abstract

High blood glucose levels cause diabetes, and it is characterized as a chronic disease that will disrupt fat and protein metabolism. The blood glucose levels rise because it cannot be burned in the cells due to a shortage of insulin secretion by the pancreas, or the insulin produced by the cell is insufficient. If exact early detection is possible, the hazard and prevalence of diabetes can be decreased considerably. With this, the application of technology has been an essential part of providing accurate and acceptable results in the prevention and early detection of the illness. This research implements artificial neural networks to predict the early stage of diabetes by incorporating methods involving feature selection or dimension reduction using a Relief-Based Filter for testing and training data. The results show 99.3% prediction accuracy and can be essential in contributing to a new way that is highly accurate in determining diabetes among patients.

Published

2022-06-15

How to Cite

Shokhjakhon Abdufattokhov, Nodira Normatova, & Makhbuba Shermatova. (2022). Artificial Neural Networks Based Predictive Model for Detecting the Early-Stage Diabetes. Journal of Artificial Intelligence,Machine Learning and Neural Network , 2(04), 1–8. https://doi.org/10.55529/jaimlnn.241.8

Similar Articles

1 2 3 4 5 6 7 > >> 

You may also start an advanced similarity search for this article.