Predicting disease risks from highly imbalanced data using random forest

Abstract

BackgroundWe present a method utilizing Healthcare Cost and Utilization Project (HCUP) dataset for predicting disease risk of individuals based on their medical diagnosis history. The presented methodology may be incorporated in a variety of applications such as risk management, tailored health communication and decision support systems in healthcare.MethodsWe employed the National Inpatient Sample (NIS) data, which is publicly available through Healthcare Cost and Utilization Project (HCUP), to train random forest classifiers for disease prediction. Since the HCUP data is highly imbalanced, we employed an ensemble learning approach based on repeated random sub-sampling. This technique divides the training data into multiple sub-samples, while ensuring that each sub-sample is fully balanced. We compared the performance of support vector machine (SVM), bagging, boosting and RF to predict the risk of eight chronic diseases.ResultsWe predicted eight disease categories. Overall, the RF ensemble learning method outperformed SVM, bagging and boosting in terms of the area under the receiver operating characteristic (ROC) curve (AUC). In addition, RF has the advantage of computing the importance of each variable in the classification process.ConclusionsIn combining repeated random sub-sampling with RF, we were able to overcome the class imbalance problem and achieve promising results. Using the national HCUP data set, we predicted eight disease categories with an average AUC of 88.79%.

Development of 365-day Life Expectancy Models: Application of Machine Learning Methods to a Prospective Study of Critically Ill Cirrhotic Patients (Preprint)

“…The main strength of this study is the use of models based on machine learning validated by leave-one-out cross-validation using prospective data, which demonstrated the clinical utility of the SVM/Ensemble/RANSAC/Random Forest methods in predicting survival [36][37][38][39]. We believe that our results may have positive clinical implications, because the information provided is objective and can be used in family meetings to help set expectations for reasonable clinical care.…”

Discovery and characterisation of dietary patterns in two Nordic countries

“…54 Actually, RF is proven highly versatile and has demonstrated high classification accuracy in numerous cases. 56,57 Because of its popularity in many areas special and/or enhanced varieties of RF have appeared. [58][59][60] The second of the two families of predictive modelling, brought into play in the study outlined here, is known as Nearest Shrunken Centroid (NSC) classifier.…”

Data Mining Big Data Inpatient Database Using Cuckoo Search

“…Such knowledge would be beneficial to hospitals, and also to insurance compaines, which can make evidence based decisions, and can optimize, validate and refine the rules that govern their business [6]. This important hidden knowledge can be found with the help of data mining, with methods such as clustering, feature selection, association rule mining, and many more.…”