Rapid introduction of new diseases and the severity improvement of existing dead diseases due to the bad food habits and lacking of awareness over the health conscious food items those are available in the market. The Internet of Things (IoT) gets more attention for reducing the disease severity by knowing the current status of their disease according to the dynamic inputs of human body through IoT devices today. Moreover, the combination of IoT and cloud computing technologies are playing major roles in e-health services. In this scenario, security is a major issue in the process of data storage and communication. For this purpose, we propose a new e-healthcare system for monitoring the dead disease level by using the technologies such as IoT and Cloud with the help of deep learning approach and fuzzy rules with temporal features. In this system, the medical data is retrieved from various located patients who are utilizing the e-healthcare assisting devices. First, the retrieved and encrypted data is stored in cloud by applying a newly proposed secured cloud storage algorithm. Second, the stored data can be retrieved the data as original data by applying the decryption process. Third, a new cloud framework is introduced for predicting the status of heart beat rates and diabetes levels by using the medical data that is created by applying the UCI Repository dataset. In addition, a new deep learning approach which applies the Convolutional Neural Network for predicting the disease severity. The experimental results are obtained by conducting various experiments for the proposed model by using the dataset and the hospital patient records. The proposed model results outperforms the available disease prediction systems in terms of prediction accuracy.
Health Risk Prediction Using Big Medical Data - a Collaborative Filtering-Enhanced Deep Learning Approach
