DeepMist: Toward Deep Learning Assisted Mist Computing Framework for Managing Healthcare Big Data

The prevalence of heart disease has remained a major cause mortalities across the world and been challenging for healthcare providers to detect early symptoms cardiac patients. To this end, several conventional machine learning models have gained popularity in providing precise prediction diseases by taking into account underlying conditions drawbacks associated with these methods are lack generalization convergence rate being much slower. As data systems scale up leading big issues, Cloud-Fog computing-based paradigm is necessary facilitate low-latency energy-efficient computation data. In paper, DeepMist framework suggested which exploits Deep Learning operating over Mist Computing infrastructure leverage fast predictive convergence, low-latency, energy efficiency smart systems. We exploit Q Network (DQN) algorithm building model identifying computing layer. Different performance evaluation metrics, like precision, recall, f-measure, accuracy, consumption, delay, used assess proposed framework. It provided an overall accuracy 97.6714 % loss value 0.3841, along consumption delay 32.1002 mJ 2.8002 ms respectively. validate efficacy DeepMist, we compare its outcomes dataset other benchmark Q-Reinforcement (QRL) Reinforcement (DRL) algorithms observe that scheme outperforms all others.