Mortality Rate Prediction in ICU Using Logistic Regression Method
DOI:
https://doi.org/10.26438/ijcse/v6i5.668674Keywords:
Predictive learning, logistic regression, SMOTE sampling, stratified sampling, time series dataAbstract
High risk of illness is observed for the patients admitted in hospital’s cardiac intensive care units (ICU). Patient’s dead/alive categorical outcome prediction would benefits for patients as well as medical professionals in creating awareness and making clinical decisions respectively. In this work, a model is proposed for predicting life outcomes of cardiac patients admitted in ICU. The model is prepared on the basis of data collected from the regular medication treatments and clinical laboratory test results. A logistic regression model is prepared and compared with two standard algorithms in machine learning such as artificial neural network (ANN) and random forest algorithms, which are the classifiers of decision tree. The performance parameters were compared for both Synthetic Minority Oversampling Technique and stratified sampling for all predictive learning models. It is concluded that logistic regression with stratified sampling techniques would be preferable as a predictive model for the inconsistent time series data set.
References
R Awang, S Palaniappa, “Intelligent heart disease prediction system using data mining techniques”, IEEE/ACS International Conference on Computer Systems and Applications, ISSN: 2161-5322,2008.
J.Sun, S. Ebadollahi, D. Gotz, J. Hu, D. Sow and C.Neti , “Predicting Patient’s Trajectory of Physiological Data using Temporal Trends in Similar Patients: A System for Near-Term Prognostics”, AMIA Symposium Proceedings, pp-192-195, 2010.
S.Wang , F.Huang, and C.Chan , “Predicting Disease By Using Data Mining Based on Healthcare Information System”, IEEE International Conference on Granular Computing, 2012.
M. Rouzbahman, R. Kealey, E. Yu, M. Chignell ,R. Samavi and T. Sieminowski, “Development of Non-Confidential Patient Types for Use in Emergency Medicine Clinical Decision Support,” IEEE Securiy & Privacy, vol. 11, pp. 12-18, 2013.
L.Morissette and S.Chartier, “The k-means clustering technique: General considerations and implementation in Mathematica”, Tutorials in Quantitative Methods for Psychology, Vol. 9 (1), p. 15-24,2013.
M. Rouzbahman, R. Kealey, E. Yu, M. Chignell ,R. Samavi and T. Sieminowski, “Development of Non-Confidential Patient Types for Use in Emergency Medicine Clinical Decision Support,” IEEE Security & Privacy, vol. 11, pp. 12-18, 2013.
J.Lee, H.Lim, D.Kim, S.Shin, J.Kim,B.Yoo, and K.Cho, “The development and implementation of stroke risk prediction model in National Health Insurance Service's personal health record”, Computer Methods and Programs in Biomedicine, Vol 153, pp. 253-257, 2018.
M.Rouzbahman, A. Jovicic, and M.Chignell, “Can Cluster-Boosted Regression Improve Prediction of Death and Length of Stay in the ICU?”, IEEE Journal of Biomedical and Health Informatics, Vol 21, pp. 851 – 858, 2016.
Nitesh V. Chawla, Kevin W. Bowyer, Lawrence O. Hall, W. Philip Kegelmeyer , “SMOTE: Synthetic Minority Over-sampling Technique”, Journal of Artificial Intelligence Research, vol.16, pp 321-357, 2002.
Kevin Lang, Edo Liberty, Konstantin Shmakov, “Stratified Sampling Meets Machine Learning”, International Conference on machine Learning, vol.48, pp 2320-2329, 2016.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors contributing to this journal agree to publish their articles under the Creative Commons Attribution 4.0 International License, allowing third parties to share their work (copy, distribute, transmit) and to adapt it, under the condition that the authors are given credit and that in the event of reuse or distribution, the terms of this license are made clear.
