Clinical Decision Support Systems: Decision-Making System for Clinical Data

Background Study:

In recent few decades, machine learning techniques are used in wide range to predict the disease based on the historical data. Many proposed algorithms are developed and studied from the researchers. In this section a few important works that are closely to the proposed work are discussed.

The prediction of readmission using big data tools based the drugs that are taken to the patient (Satish Boregowda, Rod Handy, 2016) .The care must be taken to the patient even after the discharge. The values of the patients are determined (Mingle Damian, 2017) by HbA1C results it is an important feature to control the glucose in in blood. Entropy is one of the important feature to select the (Yun Zheng, Chee Keong Kwoh, 2011) high dimensional features. High entropy alloy integrating with machine learning is doing better (Ziqing Zhou &Yeju Zhou, 2019). In this work the maximum entropy is used to optimize the objective solution of the proposed work (Rui Zhao & Xudong Sun, 2019). It also proposed three works based on the entropy. First, work is based on the weighted entropy second work is lower bound for optimization (Jayanthi, N & B. Vijaya Babu, 2017) and finally, prioritize framework on entropy. Differential entropy selecting the subset of features (Schulman & Chen, 2017) provides solution to the classification problem. The relevant features are selected it gives better accuracy. This entropy deals (Satish Boregowda & Rod Handy, 2016) nominal and real-valued data in the dataset. Neighborhood (Yanpeng Qu, & Rong, 2019) Entropy to select the relevant features (Andrea Marcello & Roberto Battiti,2018) using the neural network.