Efficient chronic disease diagnosis prediction and recommendation system

Abstract

Healthcare recommender systems for Chronic Disease Diagnosis (CDD) can play a major role in controlling the disease, through providing accurate and trustworthy disease risk diagnosis prediction and medical advices recommendation. That assists healthcare providers to have 24/7 remote patient monitoring system and assist patients to have 24 hour access to the medical care. Such systems are considered as extra tools to assist physicians and patients in controlling and managing the disease. Providing an accurate real-time recommendation for medical data is a challenge according to the complexity of the medical data represented by unbalance, large, multi-dimension, noisy and/or missing data. The CDD system expectation is to give a high accuracy of disease risk prediction and medical advices recommendation. In this paper, a CDD recommender system approach is proposed based on a hybrid method using multiple classifications and unified Collaborative Filtering. Multiple classifications based on decision tree algorithms are applied to buildan accurate predictive model that predicts the disease risk diagnosis for the monitored cases. The Unified CF method based on learning classification model using both historical binary ratings and external features will be utilized for achieving higher recommendation accuracy [1] when recommending medical advices for patients.Historical patients' medical data from the Middle East is used to train the model. Determining the relevant features through Attribute Selection method is used toreducedata generation and improve the predictive model performance. Merging patients' lab and home test readings is considered to leverage the diagnosis fidelity. Diabetes diagnosis case study is designed as experiments to show the feasibility of our model. To the written of this paper, multiple decision tree classifiers have been applied through the first stage of CDD system to achieve high accuracypredictive model. Our contribution is to provide more accurate and efficient recommendations to assist patients controlling their chronic disease and assist healthcare providers to have 24/7 remote patient monitoring system.