An Innovative Deep Learning Framework for Healthcare Cost Prediction

$A$
crucial project with many applications, such as firm profiling, responsible care management, and capitated scientific price adjustment, is accurately calculating affected individual healthcare cost. Even while considering a common man, he/she doesn't know how much money to save for health. The proposed model enables the user to know his/her expenses in health care. Existing strategies rely heavily on manually designed points and models that rely on linear regression, that require a vast amount of clinical data and have limited predictive ability. Based on historical claims data, this study provides a deep learning approach for forecasting the future healthcare expenses at the person level. Our technique can accurately model the heterogeneous data, which includes patient demographics. These findings contribute to improved preventative and corrective care in the healthcare area. Here, a prediction model has been developed based on Neural Network - LSTM - Multi-View, Random Forest Regressor, Voting Regressor, Decision Tree Regressor. In which, this study prove that deep learning algorithm LSTM implemented using keras wins the race with less error recordings when compared to others. A web interface is used to collect user data to forecast health expenditure.