Deep Optimized Smart Healthcare System for the Detection of Diabetes and Cardiovascular Ailments

The Internet of Things (loT) is a valuable tool in the field of smart healthcare for diagnosing a wide range of illnesses in Healthcare Systems. It is noticed that the traditional diagnostic techniques consume much more time and may result in inaccuracies making the ailment identification faulty. This necessitates a smart using the concept of Deep Learning (DL). As a motivation, the concept of blending ailment prediction with DL for precise prediction is attempted in this study. In this work, an innovative structure for detecting the occurrence of diabetes and cardiac ailment from the medical database is developed and it is referred to as the Deep Optimized Smart Healthcare system. In this study, an advanced optimization technique, the Boosted Binary Harris Hawks (BB-HH) is used to reduce the dimensionality of the dataset to improve the performance of the classifier implemented. Subsequently, the traits of the BB-HH optimization technique are used for the development of the Gated Recurrent Deep Convolutional Network (GDRCNet) technique for performing the classification of the ailment. Furthermore, the loss function of the Gated Recurrent Deep Convolutional Network is estimated with the help of the Krill Herd Optimization technique to enhance the accuracy of the developed classifier. The performance evaluation and validation of the proposed Deep Optimized Smart Healthcare System is carried out using diabetes and cardiovascular ailment datasets from PIMA.