An enhanced deep learning-based disease detection model in wireless body area network with energy efficient routing protocol

The group of connected small “Bio-sensor nodes (BSNs)” is employed in various parts of the human body that is called “Wireless body area networks (WBAN)”. It helps to recognize health-related data and to monitor the readings of blood pressure, “Electro-Cardiogram (ECG)”, heartbeat rate, “Electro-Myography (EMG)”, and glucose levels in the blood of the human body to know the real-time health. Many applications and research areas use the WBAN, like sports, social welfare, medical field, and entertainment. For WBAN, the major backbone is the BSNs, generally known as “Sensor nodes (SNs)”. Based on the small size of the SNs, they have basic resources. High energy is consumed when there is heavy data transmission. When all the energy is drained, that leads to the death of some SN. Routing is the data transfer method from the main source to the sink nodes. The minimum number of SNs is the efficient routing in the data transmission process, resulting in maximum energy consumption. Hence, an energy-efficient routing scheme is implemented with heuristic approaches to conserve more energy in the WBAN. To perform routing effectively, the Cluster Head (CH) needs to be selected initially. In this work, the optimal selection of the CH is carried out using a hybrid Red piranha and egret swarm algorithm (RPESA). Once the CH is optimally selected, the optimal routing is implemented using the RPESA algorithm. The data transmitted using this optimal routing scheme is then utilized for disease diagnosis using an Adaptive dilated cascaded recurrent neural network (ADC-RNN). The parameters in the ADC-RNN technique are optimally selected using the same RPESA algorithm. The classified disease outcome was obtained from ADC-RNN. The suggested heuristic-based energy-efficient routing approach for WBAN and the deep learning-based disease detection model was implemented, and its function was validated by differentiating it with other existing schemes.