New Approach based on Machine Learning for Short-Term Mortality Prediction in Neonatal Intensive Care Unit
Journal Title: International Journal of Advanced Computer Science & Applications - Year 2019, Vol 10, Issue 7
Abstract
Mortality remains one of the most important outcomes to predict in Intensive Care Units (ICUs). In fact, the sooner mortality is predicted, the better critical decisions are made by doctors based on patient’s illness severity. In this paper, a new approach based on Machine Learning (ML) techniques for short-term mortality prediction in Neonatal Intensive Care Unit (NICU) is proposed. This approach relies on many steps. At first, relevant features are selected from available data upon neonates’ admission and from the time-series variables collected within the two first hours of stay in the NICU from the Medical Information Mart for Intensive Care III (MIMIC-III). After that, to predict mortality, many classifiers were tested which are Linear Discriminant Analysis (LDA), K-Nearest Neighbors (KNN), Classification and Regression Trees (CART), Logistic Regression (LR), Support Vector Machine (SVM), Naïve Bayes (NB) and Random Forest (RF). The experimental results showed that LDA was the best performing classifier with an accuracy equal to 0.947 and AUROC equal to 0.97 with 31 features. The third step of this approach is mortality time prediction using the Galaxy-Random Forest method achieving an f-score equal to 0.871. The proposed approach compared favorably in terms of time, accuracy and AUROC with existing scoring systems and ML techniques. It is the first work predicting neonates mortality based on ML techniques and time-series data after only two hours of admission to the NICU.
Authors and Affiliations
Zaineb Kefi, Kamel Aloui, Mohamed Saber Naceur
Keywords
Related Articles
- EP ID EP611501
- DOI 10.14569/IJACSA.2019.0100778
- Views 96
- Downloads 0
Towards Usability Guidelines for the Design of Effective Arabic Websites
The Arabic websites constitute 1% of the web content with more than 225 million viewers and 41% Internet penetration. However, there is a lack of design guidelines related to the selection and use of appropriate font typ...
Conceptual Modeling in Simulation: A Representation that Assimilates Events
Simulation is often based on some type of model of the evolved portion of the world being studied. The underlying model is a static description; the simulation itself is executed by generating events or dynamic aspects i...
Improving Routing Performances to Provide Internet Connectivity in VANETs over IEEE 802.11p
In the intelligent transportation systems, many applications and services could be offered in the road via Internet. Providing these applications over vehicular ad hoc network (VANETs) technology may require good perform...
A Robust Method for Diagnostic Energetic System with Bond Graph
Surveillance and supervision systems have a major role in insuring the safety and availability of industrial equipments and installations. Default detection and diagnosis is highly important to facilitate the planning an...
Impact of Heterogeneous Deployment on Source Initiated Reactive Approach
Selection of an optimal number of high energy level nodes and the most appropriate heterogeneity level is a prerequisite in the heterogeneous deployment of wireless sensor network, and it serves several purposes like enh...