Risk Assessment for Patient and Fetal Based on Threshold Limit Using Feature Extraction and Neural Networks

Recently, cloud-based technology has been used to meet data needs, socialize, collaborate, and handle formal and informal processes. The knowledge-based machine learning version comfort people in everyday life. This includes blood glucose monitors, pulse oximeters, activity trackers, blood pressure, scales for body examination and sleep. The user-mined data is stored in a cloud-based programme, and then the data is used with smart algorithms to train the machine learning techniques. In this paper, the proposed technique evaluates the risk prediction attributes based on the specific patient and the cross-validation model while selecting and extracting the features using RNN and CNN. These features have been classified using Faster Deep Neural Networks with Genetic Decision algorithms (FDNN-GD). The classified data has been determined based on the threshold limit. Then the data has been stored in cloud architecture with encryption and decryption keys for security. Finally, when they have a higher threshold value, the foetus and patient will be tracked, which gives the doctor warning. Otherwise, it provides monitored data.