Comparison of Predictive Models for Transferring Stroke In-Patients to Intensive Care Unit
Journal Title: Transactions on Machine Learning and Artificial Intelligence - Year 2016, Vol 4, Issue 3
Abstract
The cost for Intensive Care Unit (ICU) resources is extremely high and it affects healthcare budget that provides quality healthcare service for patients. Thus, the need for a predictive model for the decision to transfer stroke in-patients to the ICU is very important to utilize the resources effectively. Also, it will help to lower morbidity and mortality rates through earlier detection and intervention. In this research, initially, a Decision Tree (DT) model, an Artificial Neural Network (ANN) model, a Support Vector Machine (SVM) model, and a Logistic Regression (LR) model are evaluated for predicting the need to transfer the stroke in-patients to the ICU or not. The study is conducted on a clinical dataset consisting of 1,415 observations using the vital signs with six variables. This original dataset was having data imbalance and hence the result was misleading. In order to overcome this situation the Synthetic Minority Oversampling Technique was applied on the dataset. A DT model, an ANN model, a SVM model, and a LR model are evaluated again on the balanced dataset for prediction. Tree-based ensemble approaches such as Generalized Boosted Model, Adaptive Boosting (AdaBoost.M1), Random Forest and Bagged AdaBoost (Adabag) are used to improve the accuracy and performance of models. These methods were trained and tested on the balanced stroke in-patients dataset. The boosting model, AdaBoost.M1, and bagging model, Random Forest, achieved better accuracy compared to the other models. Therefore, these two models could be used for helping healthcare professionals in decision-making.
Authors and Affiliations
Nawal N. Alotaibi, Sreela Sasi
Keywords
Related Articles
- EP ID EP277213
- DOI 10.14738/tmlai.43.2051
- Views 66
- Downloads 0
Hello Message Scheme Enhancement in MANET based on Neighbor Mobility
The key idea behind the routing protocols based in multi hope is to find the required path from the source node to the destination. Since those protocols does not consider the node mobility in their mechanism, we propose...
Robot-Server Architecture for Optimizing Solar Panel Power Output
Solar panel facilities for generating electricity have increased exponentially in the recent years. Dust and bird droppings on the solar panels inhibit the energy production. Having people to inspect them and, if needed,...
Accuracy of the Java Simulation for the Charge Motion in Electric and Magnetic Fields
The accuracy of the Java simulation by the Runge-Kutta method for the charge motion in electric and magnetic fields has been investigated in comparison with the analytical solution. The error of the simulation depends on...
Psychological and Physiological Profiles in Oncology Caregivers: a Multivariable Cross-Sectional Study
Scientific research has led to enhanced medical assistance, and hence higher disease control and a continuous lifespan dilation. However, research mainly focusses on the patient’s condition, and, in some cases, caregiver...
Application of Genetic Algorithms Coupled with Neural Networks to Optimization of Reinforced Concrete Footings
This paper first applies genetic algorithms to optimally design reinforced concrete isolated footings subjected to concentric loading. Based on the ACI Building Code, constraints are built by considering wide-beam and pu...