The comparison of selected machine learning techniques and correlation matrix in ICU mortality risk prediction
Introduction: Identifying and analyzing mortality risk factors will lead to more accurate planning and prevention in health platforms. This research provides models for predicting mortality in the intensive care unit with machine-learning techniques. Method: We extracted data from 1400 patients'...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-01-01
|
| Series: | Informatics in Medicine Unlocked |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2352914822001381 |
| _version_ | 1828461914705887232 |
|---|---|
| author | Parnian Asgari Mir Mohammad Miri Fahimeh Asgari |
| author_facet | Parnian Asgari Mir Mohammad Miri Fahimeh Asgari |
| author_sort | Parnian Asgari |
| collection | DOAJ |
| description | Introduction: Identifying and analyzing mortality risk factors will lead to more accurate planning and prevention in health platforms. This research provides models for predicting mortality in the intensive care unit with machine-learning techniques. Method: We extracted data from 1400 patients' medical records admitted to the intensive care unit of Imam Hossein Hospital in Tehran. Period of inclusion was January 1, 2018, to February 31, 2020. In this study, we used twelve predictor algorithms for data analysis, including: SVM, k-nearest neighbor, decision tree, logistic regression, random forest, MLP, RBF, Gradient Boosting, Fast Large Margin, Rule Model, Naive Bayesian, and Deep learning. We reported the performance of algorithms based on the precision, accuracy, sensitivity, specificity, and AUC. Results: The findings showed that algorithms according to the percentage of AUC were 0.80 in Gradient Boosting, 0.781 in deep learning, 0.746 in logistic regression, 0.740 in SVM and 0.735 in MLP. Algorithm with the highest level of AUC was Gradient Boosting. Gradient Boosting's AUC = (0.8) was better at predicting mortality in intensive care units. 1) bilirubin 2) INR 3) Age 4) WBC had the strongest relationship with mortality in Gradient Boosting. Conclusion: Data analysis in ICU patients can be a valuable tool for predicting mortality and morbidity associated with mortality, but the results vary according to the data quality. However, the processes and methods mentioned in this study suggested that the rules extracted from the Gradient Boosting can be used as a model to predict mortality in intensive care units. |
| first_indexed | 2024-12-11T02:27:16Z |
| format | Article |
| id | doaj.art-9aeedd8459c0421a997e51faad583e57 |
| institution | Directory Open Access Journal |
| issn | 2352-9148 |
| language | English |
| last_indexed | 2024-12-11T02:27:16Z |
| publishDate | 2022-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Informatics in Medicine Unlocked |
| spelling | doaj.art-9aeedd8459c0421a997e51faad583e572022-12-22T01:23:54ZengElsevierInformatics in Medicine Unlocked2352-91482022-01-0131100995The comparison of selected machine learning techniques and correlation matrix in ICU mortality risk predictionParnian Asgari0Mir Mohammad Miri1Fahimeh Asgari2Department of Health Information Technology, School of Paramedicine, Mashhad University of Medical Sciences, Mashhad, IranDepartment of Anesthesiology, Faculty of Medicine, Shahid Beheshti University of Medical Sciences, Tehran, IranDepartment of Health Economics, School of Medicine, Shahed University of Medical Sciences, Tehran, Iran; Corresponding author. Department of Health Economics, Faculty of Medicine, Shahed University of Medical Sciences, Tehran, Iran.Introduction: Identifying and analyzing mortality risk factors will lead to more accurate planning and prevention in health platforms. This research provides models for predicting mortality in the intensive care unit with machine-learning techniques. Method: We extracted data from 1400 patients' medical records admitted to the intensive care unit of Imam Hossein Hospital in Tehran. Period of inclusion was January 1, 2018, to February 31, 2020. In this study, we used twelve predictor algorithms for data analysis, including: SVM, k-nearest neighbor, decision tree, logistic regression, random forest, MLP, RBF, Gradient Boosting, Fast Large Margin, Rule Model, Naive Bayesian, and Deep learning. We reported the performance of algorithms based on the precision, accuracy, sensitivity, specificity, and AUC. Results: The findings showed that algorithms according to the percentage of AUC were 0.80 in Gradient Boosting, 0.781 in deep learning, 0.746 in logistic regression, 0.740 in SVM and 0.735 in MLP. Algorithm with the highest level of AUC was Gradient Boosting. Gradient Boosting's AUC = (0.8) was better at predicting mortality in intensive care units. 1) bilirubin 2) INR 3) Age 4) WBC had the strongest relationship with mortality in Gradient Boosting. Conclusion: Data analysis in ICU patients can be a valuable tool for predicting mortality and morbidity associated with mortality, but the results vary according to the data quality. However, the processes and methods mentioned in this study suggested that the rules extracted from the Gradient Boosting can be used as a model to predict mortality in intensive care units.http://www.sciencedirect.com/science/article/pii/S2352914822001381MortalityData miningMachine learningIntensive care unitPredictive model |
| spellingShingle | Parnian Asgari Mir Mohammad Miri Fahimeh Asgari The comparison of selected machine learning techniques and correlation matrix in ICU mortality risk prediction Informatics in Medicine Unlocked Mortality Data mining Machine learning Intensive care unit Predictive model |
| title | The comparison of selected machine learning techniques and correlation matrix in ICU mortality risk prediction |
| title_full | The comparison of selected machine learning techniques and correlation matrix in ICU mortality risk prediction |
| title_fullStr | The comparison of selected machine learning techniques and correlation matrix in ICU mortality risk prediction |
| title_full_unstemmed | The comparison of selected machine learning techniques and correlation matrix in ICU mortality risk prediction |
| title_short | The comparison of selected machine learning techniques and correlation matrix in ICU mortality risk prediction |
| title_sort | comparison of selected machine learning techniques and correlation matrix in icu mortality risk prediction |
| topic | Mortality Data mining Machine learning Intensive care unit Predictive model |
| url | http://www.sciencedirect.com/science/article/pii/S2352914822001381 |
| work_keys_str_mv | AT parnianasgari thecomparisonofselectedmachinelearningtechniquesandcorrelationmatrixinicumortalityriskprediction AT mirmohammadmiri thecomparisonofselectedmachinelearningtechniquesandcorrelationmatrixinicumortalityriskprediction AT fahimehasgari thecomparisonofselectedmachinelearningtechniquesandcorrelationmatrixinicumortalityriskprediction AT parnianasgari comparisonofselectedmachinelearningtechniquesandcorrelationmatrixinicumortalityriskprediction AT mirmohammadmiri comparisonofselectedmachinelearningtechniquesandcorrelationmatrixinicumortalityriskprediction AT fahimehasgari comparisonofselectedmachinelearningtechniquesandcorrelationmatrixinicumortalityriskprediction |