Machine Learning Prediction Models for Mortality in Intensive Care Unit Patients with Lactic Acidosis

Background: Lactic acidosis is the most common cause of anion gap metabolic acidosis in the intensive care unit (ICU), associated with poor outcomes including mortality. We sought to compare machine learning (ML) approaches versus logistic regression analysis for prediction of mortality in lactic ac...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Pattharawin Pattharanitima, Charat Thongprayoon, Wisit Kaewput, Fawad Qureshi, Fahad Qureshi, Tananchai Petnak, Narat Srivali, Guido Gembillo, Oisin A. O’Corragain, Supavit Chesdachai, Saraschandra Vallabhajosyula, Pramod K. Guru, Michael A. Mao, Vesna D. Garovic, John J. Dillon, Wisit Cheungpasitporn
Formato: article
Lenguaje:EN
Publicado: MDPI AG 2021
Materias:
R
Acceso en línea:https://doaj.org/article/a3e7d1176aa740038116d80ea6a4475f
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:a3e7d1176aa740038116d80ea6a4475f
record_format dspace
spelling oai:doaj.org-article:a3e7d1176aa740038116d80ea6a4475f2021-11-11T17:38:57ZMachine Learning Prediction Models for Mortality in Intensive Care Unit Patients with Lactic Acidosis10.3390/jcm102150212077-0383https://doaj.org/article/a3e7d1176aa740038116d80ea6a4475f2021-10-01T00:00:00Zhttps://www.mdpi.com/2077-0383/10/21/5021https://doaj.org/toc/2077-0383Background: Lactic acidosis is the most common cause of anion gap metabolic acidosis in the intensive care unit (ICU), associated with poor outcomes including mortality. We sought to compare machine learning (ML) approaches versus logistic regression analysis for prediction of mortality in lactic acidosis patients admitted to the ICU. Methods: We used the Medical Information Mart for Intensive Care (MIMIC-III) database to identify ICU adult patients with lactic acidosis (serum lactate ≥4 mmol/L). The outcome of interest was hospital mortality. We developed prediction models using four ML approaches consisting of random forest (RF), decision tree (DT), extreme gradient boosting (XGBoost), artificial neural network (ANN), and statistical modeling with forward stepwise logistic regression using the testing dataset. We then assessed model performance using area under the receiver operating characteristic curve (AUROC), accuracy, precision, error rate, Matthews correlation coefficient (MCC), F1 score, and assessed model calibration using the Brier score, in the independent testing dataset. Results: Of 1919 lactic acidosis ICU patients, 1535 and 384 were included in the training and testing dataset, respectively. Hospital mortality was 30%. RF had the highest AUROC at 0.83, followed by logistic regression 0.81, XGBoost 0.81, ANN 0.79, and DT 0.71. In addition, RF also had the highest accuracy (0.79), MCC (0.45), F1 score (0.56), and lowest error rate (21.4%). The RF model was the most well-calibrated. The Brier score for RF, DT, XGBoost, ANN, and multivariable logistic regression was 0.15, 0.19, 0.18, 0.19, and 0.16, respectively. The RF model outperformed multivariable logistic regression model, SOFA score (AUROC 0.74), SAP II score (AUROC 0.77), and Charlson score (AUROC 0.69). Conclusion: The ML prediction model using RF algorithm provided the highest predictive performance for hospital mortality among ICU patient with lactic acidosis.Pattharawin PattharanitimaCharat ThongprayoonWisit KaewputFawad QureshiFahad QureshiTananchai PetnakNarat SrivaliGuido GembilloOisin A. O’CorragainSupavit ChesdachaiSaraschandra VallabhajosyulaPramod K. GuruMichael A. MaoVesna D. GarovicJohn J. DillonWisit CheungpasitpornMDPI AGarticlelactic acidlactic acidosislactatemortalityintensive care unitmachine learningMedicineRENJournal of Clinical Medicine, Vol 10, Iss 5021, p 5021 (2021)
institution DOAJ
collection DOAJ
language EN
topic lactic acid
lactic acidosis
lactate
mortality
intensive care unit
machine learning
Medicine
R
spellingShingle lactic acid
lactic acidosis
lactate
mortality
intensive care unit
machine learning
Medicine
R
Pattharawin Pattharanitima
Charat Thongprayoon
Wisit Kaewput
Fawad Qureshi
Fahad Qureshi
Tananchai Petnak
Narat Srivali
Guido Gembillo
Oisin A. O’Corragain
Supavit Chesdachai
Saraschandra Vallabhajosyula
Pramod K. Guru
Michael A. Mao
Vesna D. Garovic
John J. Dillon
Wisit Cheungpasitporn
Machine Learning Prediction Models for Mortality in Intensive Care Unit Patients with Lactic Acidosis
description Background: Lactic acidosis is the most common cause of anion gap metabolic acidosis in the intensive care unit (ICU), associated with poor outcomes including mortality. We sought to compare machine learning (ML) approaches versus logistic regression analysis for prediction of mortality in lactic acidosis patients admitted to the ICU. Methods: We used the Medical Information Mart for Intensive Care (MIMIC-III) database to identify ICU adult patients with lactic acidosis (serum lactate ≥4 mmol/L). The outcome of interest was hospital mortality. We developed prediction models using four ML approaches consisting of random forest (RF), decision tree (DT), extreme gradient boosting (XGBoost), artificial neural network (ANN), and statistical modeling with forward stepwise logistic regression using the testing dataset. We then assessed model performance using area under the receiver operating characteristic curve (AUROC), accuracy, precision, error rate, Matthews correlation coefficient (MCC), F1 score, and assessed model calibration using the Brier score, in the independent testing dataset. Results: Of 1919 lactic acidosis ICU patients, 1535 and 384 were included in the training and testing dataset, respectively. Hospital mortality was 30%. RF had the highest AUROC at 0.83, followed by logistic regression 0.81, XGBoost 0.81, ANN 0.79, and DT 0.71. In addition, RF also had the highest accuracy (0.79), MCC (0.45), F1 score (0.56), and lowest error rate (21.4%). The RF model was the most well-calibrated. The Brier score for RF, DT, XGBoost, ANN, and multivariable logistic regression was 0.15, 0.19, 0.18, 0.19, and 0.16, respectively. The RF model outperformed multivariable logistic regression model, SOFA score (AUROC 0.74), SAP II score (AUROC 0.77), and Charlson score (AUROC 0.69). Conclusion: The ML prediction model using RF algorithm provided the highest predictive performance for hospital mortality among ICU patient with lactic acidosis.
format article
author Pattharawin Pattharanitima
Charat Thongprayoon
Wisit Kaewput
Fawad Qureshi
Fahad Qureshi
Tananchai Petnak
Narat Srivali
Guido Gembillo
Oisin A. O’Corragain
Supavit Chesdachai
Saraschandra Vallabhajosyula
Pramod K. Guru
Michael A. Mao
Vesna D. Garovic
John J. Dillon
Wisit Cheungpasitporn
author_facet Pattharawin Pattharanitima
Charat Thongprayoon
Wisit Kaewput
Fawad Qureshi
Fahad Qureshi
Tananchai Petnak
Narat Srivali
Guido Gembillo
Oisin A. O’Corragain
Supavit Chesdachai
Saraschandra Vallabhajosyula
Pramod K. Guru
Michael A. Mao
Vesna D. Garovic
John J. Dillon
Wisit Cheungpasitporn
author_sort Pattharawin Pattharanitima
title Machine Learning Prediction Models for Mortality in Intensive Care Unit Patients with Lactic Acidosis
title_short Machine Learning Prediction Models for Mortality in Intensive Care Unit Patients with Lactic Acidosis
title_full Machine Learning Prediction Models for Mortality in Intensive Care Unit Patients with Lactic Acidosis
title_fullStr Machine Learning Prediction Models for Mortality in Intensive Care Unit Patients with Lactic Acidosis
title_full_unstemmed Machine Learning Prediction Models for Mortality in Intensive Care Unit Patients with Lactic Acidosis
title_sort machine learning prediction models for mortality in intensive care unit patients with lactic acidosis
publisher MDPI AG
publishDate 2021
url https://doaj.org/article/a3e7d1176aa740038116d80ea6a4475f
work_keys_str_mv AT pattharawinpattharanitima machinelearningpredictionmodelsformortalityinintensivecareunitpatientswithlacticacidosis
AT charatthongprayoon machinelearningpredictionmodelsformortalityinintensivecareunitpatientswithlacticacidosis
AT wisitkaewput machinelearningpredictionmodelsformortalityinintensivecareunitpatientswithlacticacidosis
AT fawadqureshi machinelearningpredictionmodelsformortalityinintensivecareunitpatientswithlacticacidosis
AT fahadqureshi machinelearningpredictionmodelsformortalityinintensivecareunitpatientswithlacticacidosis
AT tananchaipetnak machinelearningpredictionmodelsformortalityinintensivecareunitpatientswithlacticacidosis
AT naratsrivali machinelearningpredictionmodelsformortalityinintensivecareunitpatientswithlacticacidosis
AT guidogembillo machinelearningpredictionmodelsformortalityinintensivecareunitpatientswithlacticacidosis
AT oisinaocorragain machinelearningpredictionmodelsformortalityinintensivecareunitpatientswithlacticacidosis
AT supavitchesdachai machinelearningpredictionmodelsformortalityinintensivecareunitpatientswithlacticacidosis
AT saraschandravallabhajosyula machinelearningpredictionmodelsformortalityinintensivecareunitpatientswithlacticacidosis
AT pramodkguru machinelearningpredictionmodelsformortalityinintensivecareunitpatientswithlacticacidosis
AT michaelamao machinelearningpredictionmodelsformortalityinintensivecareunitpatientswithlacticacidosis
AT vesnadgarovic machinelearningpredictionmodelsformortalityinintensivecareunitpatientswithlacticacidosis
AT johnjdillon machinelearningpredictionmodelsformortalityinintensivecareunitpatientswithlacticacidosis
AT wisitcheungpasitporn machinelearningpredictionmodelsformortalityinintensivecareunitpatientswithlacticacidosis
_version_ 1718432081321656320