An efficient convolutional neural network-extreme gradient boosting hybrid deep learning model for disease detection applications

Navaneeth Bhaskar, Aswathy Maruthompilli Ajithkumar, Priyanka Tupe-Waghmare


In this paper, we present an efficient deep-learning hybrid model comprising an extreme gradient boosting (XGBoost) supervised learning algorithm and convolutional neural networks (CNN) for the automated detection of diseases. The proposed model is implemented and tested to detect type-2 diabetes by measuring the acetone concentration in the exhaled breath. Acetone will be present in much higher concentrations in type-2 diabetic patients compared to non-diabetic people. A novel sensing module is designed and implemented in our study to measure the acetone concentration in exhaled breath. The proposed approach delivered good results, with a classification accuracy of 97.14%. The findings of this study show how effectively the proposed detection module functions in disease diagnosis applications. As the detection process is simple and non-invasive, people can undergo routine checks for diabetes with the proposed detection module.


Convolutional neural network; Deep learning; Extreme gradient boosting; Gas sensor; Signal processing; Type-2 diabetes

Full Text:



Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

International Journal of Electrical and Computer Engineering (IJECE)
p-ISSN 2088-8708, e-ISSN 2722-2578

This journal is published by the Institute of Advanced Engineering and Science (IAES) in collaboration with Intelektual Pustaka Media Utama (IPMU).