DiabetesNet: A Deep Learning Approach to Diabetes Diagnosis

TL;DR

This paper introduces DiabetesNet, a deep learning model using BPNN with batch normalization and data re-sampling to improve non-invasive diabetes diagnosis accuracy across multiple datasets.

Contribution

It presents a novel application of BPNN with class balancing techniques for non-invasive diabetes diagnosis, outperforming traditional machine learning models.

Findings

Achieved up to 95.28% accuracy on Mesra Diabetes dataset

Significant improvements in sensitivity and specificity over traditional methods

Demonstrated robustness across three different datasets

Abstract

Diabetes, resulting from inadequate insulin production or utilization, causes extensive harm to the body. Existing diagnostic methods are often invasive and come with drawbacks, such as cost constraints. Although there are machine learning models like Classwise k Nearest Neighbor (CkNN) and General Regression Neural Network (GRNN), they struggle with imbalanced data and result in under-performance. Leveraging advancements in sensor technology and machine learning, we propose a non-invasive diabetes diagnosis using a Back Propagation Neural Network (BPNN) with batch normalization, incorporating data re-sampling and normalization for class balancing. Our method addresses existing challenges such as limited performance associated with traditional machine learning. Experimental results on three datasets show significant improvements in overall accuracy, sensitivity, and specificity compared to traditional methods. Notably, we achieve accuracies of 89.81% in Pima diabetes dataset, 75.49% in CDC BRFSS2015 dataset, and 95.28% in Mesra Diabetes dataset. This underscores the potential of deep learning models for robust diabetes diagnosis. See project website https://steve-zeyu-zhang.github.io/DiabetesDiagnosis/