Remote healthcare systems for diabetes diagnosis : A bio-inspired deep learning approach using NCA-CNN

The development of healthcare systems has been greatly enhanced by recent advances in the IoT and AI, especially machine learning technology, which will now be used to prevent, diagnose and treat patients they have been monitored remotely and in the comfort of your home so This study focuses more on in-depth learning about building remote health care systems for diagnosing diabetes. Both the speed of data processing and the accuracy of detection are enhanced in the proposed system. The two steps in the hybrid deep learning approach (NCA-CNN) method present are as follows: First, significant features are selected from all data, then selected features are classified in NCA algorithm is a statistical method that selects important features and scores features based on data analysis findings. Subsequently, the variable depth of tissue classifies the most dramatic features, leading to more accurate diagnosis. The evaluation results show that the accuracy of the proposed method is 96.88%.