Mortality Prediction in ICU Patients Using Machine Learning Models

Effective utilization of limited intensive care unit (ICU) allocations is a challenging task for medical experts to save precious human lives. The prolonged ICU stay of relatively secure patients and patients with low chances of recovery can cause life-threatening effects on patients waiting for ICU accommodation. Machine learning-based techniques for early prediction of mortality can help in this regard. This paper presents two mortality prediction models using the support vector machine (SVM) and linear discriminant analysis. The proposed models use clinical data of the ICU patients for early prediction of mortality. Distribution filtering is performed using the chi-square distribution during pre-processing. A subset of the publicly available Medical Information Mart for Intensive Care (MIMIC-III v1.4) dataset is used for the evaluation of the proposed models. The problem of class imbalance is handled by synthetic minority oversampling technique. The comparison of obtained results is performed with existing SVM and multiple logistic regression models to show the effectiveness of the proposed models. The results of the study can be helpful for clinical experts for better decision making regarding the utilization of ICU allocations.

[1]  E. Draper,et al.  APACHE II: A severity of disease classification system , 1985, Critical care medicine.

[2]  T. Wilt,et al.  The Association of Registered Nurse Staffing Levels and Patient Outcomes: Systematic Review and Meta-Analysis , 2007, Medical care.

[3]  Hamid Mohamadlou,et al.  Using electronic health record collected clinical variables to predict medical intensive care unit mortality , 2016, Annals of medicine and surgery.

[4]  Jing Li,et al.  Admissions optimisation and premature discharge decisions in intensive care units , 2015 .

[5]  Ankur Teredesai,et al.  Interpretable Machine Learning in Healthcare , 2018, 2018 IEEE International Conference on Healthcare Informatics (ICHI).

[6]  I. Kohane,et al.  Big Data and Machine Learning in Health Care. , 2018, JAMA.

[7]  Yixin Chen,et al.  An integrated data mining approach to real-time clinical monitoring and deterioration warning , 2012, KDD.

[8]  Carri W. Chan,et al.  Critical Care Capacity Management: Understanding the Role of a Step Down Unit , 2018 .

[9]  G. Clermont,et al.  Growth of intensive care unit resource use and its estimated cost in Medicare* , 2008, Critical care medicine.

[10]  Wenqi Hu,et al.  An Examination of Early Transfers to the ICU Based on a Physiologic Risk Score , 2018, Manuf. Serv. Oper. Manag..

[11]  Yan Liu,et al.  Benchmark of Deep Learning Models on Large Healthcare MIMIC Datasets , 2017, ArXiv.

[12]  Maurice Bruynooghe,et al.  Mining data from intensive care patients , 2007, Adv. Eng. Informatics.

[13]  N. Halpern,et al.  Critical care medicine in the United States 2000–2005: An analysis of bed numbers, occupancy rates, payer mix, and costs* , 2010, Critical care medicine.

[14]  D. E. Lawrence,et al.  APACHE—acute physiology and chronic health evaluation: a physiologically based classification system , 1981, Critical care medicine.

[15]  Tanvi Banerjee,et al.  Early Hospital Mortality Prediction using Vital Signals , 2018, Smart health.

[16]  S. Lemeshow,et al.  A new Simplified Acute Physiology Score (SAPS II) based on a European/North American multicenter study. , 1993, JAMA.

[17]  W. Knaus,et al.  The APACHE III prognostic system. Risk prediction of hospital mortality for critically ill hospitalized adults. , 1991, Chest.

[18]  J. Tu,et al.  Use of a neural network as a predictive instrument for length of stay in the intensive care unit following cardiac surgery. , 1993, Computers and biomedical research, an international journal.

[19]  Romain Pirracchio,et al.  Mortality Prediction in the ICU Based on MIMIC-II Results from the Super ICU Learner Algorithm (SICULA) Project , 2016 .

[20]  Filip De Turck,et al.  Bmc Medical Informatics and Decision Making Support Vector Machine versus Logistic Regression Modeling for Prediction of Hospital Mortality in Critically Ill Patients with Haematological Malignancies , 2008 .

[21]  Alistair E. W. Johnson,et al.  Patient specific predictions in the intensive care unit using a Bayesian ensemble , 2012, 2012 Computing in Cardiology.

[22]  Cheng H. Lee,et al.  An imputation-enhanced algorithm for ICU mortality prediction , 2012, 2012 Computing in Cardiology.

[23]  Jorma Laurikkala,et al.  Improving Identification of Difficult Small Classes by Balancing Class Distribution , 2001, AIME.

[24]  Gustavo E. A. P. A. Batista,et al.  A study of the behavior of several methods for balancing machine learning training data , 2004, SKDD.

[25]  Peter Szolovits,et al.  MIMIC-III, a freely accessible critical care database , 2016, Scientific Data.

[26]  Tammy Rotem,et al.  Improved ICU mortality prediction based on SOFA scores and gastrointestinal parameters , 2019, PloS one.

[27]  Tamás D. Gedeon,et al.  Balancing Bias and Variance: Network Topology and Pattern Set Reduction Techniques , 1995, IWANN.

[28]  Cynthia Rudin,et al.  Optimized Scoring Systems: Toward Trust in Machine Learning for Healthcare and Criminal Justice , 2018, Interfaces.

[29]  N. D. de Keizer,et al.  Mortality prediction by SOFA score in ICU-patients after cardiac surgery; comparison with traditional prognostic–models , 2020, BMC Anesthesiology.

[30]  Vaibhav Rajan,et al.  ICU Mortality Prediction: A Classification Algorithm for Imbalanced Datasets , 2017, AAAI.

[31]  Roger G. Mark,et al.  Real-time Mortality Prediction in the Intensive Care Unit , 2018, AMIA.

[32]  Ming Chen,et al.  Big Data Analytics in Medicine and Healthcare , 2018, J. Integr. Bioinform..

[33]  Carri W. Chan,et al.  The Impact of Delays on Service Times in the Intensive Care Unit , 2017, Manag. Sci..