Real-time multi level chronic disease prediction and recommendation model using deep learning

The key highlights of the paper entitled “Real-time Multi Level Chronic Disease Prediction and Recommendation Model Using Deep Learning” are as follows:. • To develop a framework named as an efficient Real-time Multi Level Chronic Disease Prediction and Recommendation Model Using Deep Learning model with neural network. • The method uses to identify the stages of the chronic disease to start with the feature depth clustering algorithm it uses Depth Class Similarity (DCS) and average feature similarity value to find the stages of the disease. • The method computes the Depth Class Similarity (DCS) value The DCS value has been measured according to the number of features the tuple has and number of features the samples of the class have and also it computes the average feature similarity value AFSV The value of AFSV is measured according to the similarity of the feature values between the given sample and the samples of the class. • The paper may gain high interest from the readers and experts of. ü Chronic Disease Prediction. ü Deep Learning. ü Neural Network. Towards the problem of chronic disease prediction and recommendation, an efficient real-time multi-level chronic disease prediction and recommendation model (RMCDPRM) is presented. The model preprocesses the medical data set and cluster the data set under multiple levels according to the stages of disease. Once the data set is grouped, then the features of the trace are extracted and trained with neural network. At the test phase, the neurons of intermediate layer estimates Class Level Disease Support Value (CLDSV) for various features. This will be measured for all the class of diseases and received at the output layer. Finally, the method computes the value of Disease weight for all the classes. Finally, a class with higher DW is selected as result. Further, the method computes Disease Handling Support (DHS) and Patient Curing Weight (PCW) for various hospitals and medical practitioners. Based on the values of DHS and PCW, the method computes Medical Support MS and ranks them to generate recommendations for the user. The RMCDPRM model hikes the disease prediction accuracy up to 99 %.