IOT-Based Smart Health Care Patient Monitoring System Using Dual Sampling Dilated Pre- Activation Simplicial Convolution Neural Network with Artificial Hummingbird Algorithm

Abstract

Heart disease prediction is the leading cause of death globally at the moment, and its prevalence is rising. There are difficulties in identifying heart disease in its earliest stages before a cardiac event occurs. A vast variety of information about heart illness is accessible in the medical field, including in clinics and hospitals. To find the hidden patterns, this data is not treated intelligently enough. Still, existing techniques have many shortcomings, such as low accuracy and a high error rate. The detection and classification of heart disease remains a major challenge in healthcare. In this concern, a Dual Sampling Dilated Pre- Activation Simplicial Convolution Neural Network (DuSDPreASCNN) is proposed that provides high accuracy with less error rates during heart disease prediction and classification. First, pre-processing is applied to the input ECG data that were gathered from the Cleveland dataset. During this stage, the irrelevant noise components are suppressed. Following that, the pre-processing is done by using Enhanced Gaussian Short-Time Fourier Transform and Spectrograms Continuous Wavelet Transform (EGSFT-SCWT) method then the preprocessed data undergoes prediction and classification stage. It is done by using Dual Sampling Dilated Pre- Activation Simplicial Convolution Neural Network (DuSDPreASCNN) and optimized using Artificial Hummingbird Algorithm (AHA) for monitoring the Smart-health care Patient monitoring system of IoT-based healthcare system with outstanding accuracy and effectiveness. The efficiency of the proposed DuSDPreASCNN -AHA is analyzed using Cleveland dataset and attains 99.5% accuracy, 96.06 % recall and attains better results in comparison with the existing techniques. The outcomes of the proposed technique showed that it could improve the patient outcomes, early stage diagnosis and reduced healthcare cost method.