Deep Learning based Spectrum predictions for Cognitive Healthcare system

The prevailing pandemic scenario and the constant explosion of wireless devices lead to demand for a large spectrum. There is a crucial need for regular supervision of patient's health in these dire situations of a pandemic. This demands a smart solution for healthcare based on the latest technologies. This research work presents a detailed review of the latest advancements in the context of smart healthcare and medical imaging. This research work also proposes a novel framework for Cognitive healthcare based on Deep Learning. A unique spectrum prediction model has been proposed that combines the Grey model and Long Short Term Memory functionalities and allocates spectrum to the patients according to their risk classification (high, medium, and low). This research work describes the distinct novel applications of Cognitive radio, Deep learning, and the Internet of things in the Smart healthcare and Medical Imaging context. The proposed LSTM-Grey model has also been compared with existing spectrum sensing methodologies based on LSTM and CNN. All simulation results based on detection probability, false alarm probability and Accuracy of prediction indicate that LSTM-Grey model outperforms the existing solutions for healthcare system. The unique solution proposed based on the Grey model and LSTM for spectrum prediction enables transmission of medical data on free spectrum slots for real-time supervision of patient's health. The Deep Learning-based Cognitive Healthcare system is a promising candidate for Smart Healthcare supporting swift diagnosis, tracking, and patient health supervision.