Automated Triage System for Intensive Care Admissions during the COVID-19 Pandemic Using Hybrid XGBoost-AHP Approach

The sudden increase in patients with severe COVID-19 has obliged doctors to make admissions to intensive care units (ICUs) in health care practices where capacity is exceeded by the demand. To help with difficult triage decisions, we proposed an integration system Xtreme Gradient Boosting (XGBoost) classifier and Analytic Hierarchy Process (AHP) to assist health authorities in identifying patients’ priorities to be admitted into ICUs according to the findings of the biological laboratory investigation for patients with COVID-19. The Xtreme Gradient Boosting (XGBoost) classifier was used to decide whether or not they should admit patients into ICUs, before applying them to an AHP for admissions’ priority ranking for ICUs. The 38 commonly used clinical variables were considered and their contributions were determined by the Shapley’s Additive explanations (SHAP) approach. In this research, five types of classifier algorithms were compared: Support Vector Machine (SVM), Decision Tree (DT), K-Nearest Neighborhood (KNN), Random Forest (RF), and Artificial Neural Network (ANN), to evaluate the XGBoost performance, while the AHP system compared its results with a committee formed from experienced clinicians. The proposed (XGBoost) classifier achieved a high prediction accuracy as it could discriminate between patients with COVID-19 who need ICU admission and those who do not with accuracy, sensitivity, and specificity rates of 97%, 96%, and 96% respectively, while the AHP system results were close to experienced clinicians’ decisions for determining the priority of patients that need to be admitted to the ICU. Eventually, medical sectors can use the suggested framework to classify patients with COVID-19 who require ICU admission and prioritize them based on integrated AHP methodologies.

Download Full-text

Early Prediction of Seven-Day Mortality in Intensive Care Unit Using a Machine Learning Model: Results from the SPIN-UTI Project

Journal of Clinical Medicine ◽

10.3390/jcm10050992 ◽

2021 ◽

Vol 10 (5) ◽

pp. 992

Author(s):

Martina Barchitta ◽

Andrea Maugeri ◽

Giuliana Favara ◽

Paolo Marco Riela ◽

Giovanni Gallo ◽

...

Keyword(s):

Machine Learning ◽

Intensive Care ◽

Intensive Care Units ◽

Learning Algorithm ◽

Area Under The Curve ◽

Support Vector ◽

Icu Admission ◽

Risk Of Death ◽

Saps Ii ◽

Svm Algorithm

Patients in intensive care units (ICUs) were at higher risk of worsen prognosis and mortality. Here, we aimed to evaluate the ability of the Simplified Acute Physiology Score (SAPS II) to predict the risk of 7-day mortality, and to test a machine learning algorithm which combines the SAPS II with additional patients’ characteristics at ICU admission. We used data from the “Italian Nosocomial Infections Surveillance in Intensive Care Units” network. Support Vector Machines (SVM) algorithm was used to classify 3782 patients according to sex, patient’s origin, type of ICU admission, non-surgical treatment for acute coronary disease, surgical intervention, SAPS II, presence of invasive devices, trauma, impaired immunity, antibiotic therapy and onset of HAI. The accuracy of SAPS II for predicting patients who died from those who did not was 69.3%, with an Area Under the Curve (AUC) of 0.678. Using the SVM algorithm, instead, we achieved an accuracy of 83.5% and AUC of 0.896. Notably, SAPS II was the variable that weighted more on the model and its removal resulted in an AUC of 0.653 and an accuracy of 68.4%. Overall, these findings suggest the present SVM model as a useful tool to early predict patients at higher risk of death at ICU admission.

Download Full-text

Prediction of Healing Performance of Autogenous Healing Concrete Using Machine Learning

Materials ◽

10.3390/ma14154068 ◽

2021 ◽

Vol 14 (15) ◽

pp. 4068

Author(s):

Xu Huang ◽

Mirna Wasouf ◽

Jessada Sresakoolchai ◽

Sakdirat Kaewunruen

Keyword(s):

Machine Learning ◽

Search Algorithm ◽

Weather Conditions ◽

Prediction Performance ◽

Machine Learning Algorithms ◽

Coefficient Of Determination ◽

Gradient Boosting ◽

Support Vector ◽

Self Healing ◽

Artificial Neural Network Ann

Cracks typically develop in concrete due to shrinkage, loading actions, and weather conditions; and may occur anytime in its life span. Autogenous healing concrete is a type of self-healing concrete that can automatically heal cracks based on physical or chemical reactions in concrete matrix. It is imperative to investigate the healing performance that autogenous healing concrete possesses, to assess the extent of the cracking and to predict the extent of healing. In the research of self-healing concrete, testing the healing performance of concrete in a laboratory is costly, and a mass of instances may be needed to explore reliable concrete design. This study is thus the world’s first to establish six types of machine learning algorithms, which are capable of predicting the healing performance (HP) of self-healing concrete. These algorithms involve an artificial neural network (ANN), a k-nearest neighbours (kNN), a gradient boosting regression (GBR), a decision tree regression (DTR), a support vector regression (SVR) and a random forest (RF). Parameters of these algorithms are tuned utilising grid search algorithm (GSA) and genetic algorithm (GA). The prediction performance indicated by coefficient of determination (R2) and root mean square error (RMSE) measures of these algorithms are evaluated on the basis of 1417 data sets from the open literature. The results show that GSA-GBR performs higher prediction performance (R2GSA-GBR = 0.958) and stronger robustness (RMSEGSA-GBR = 0.202) than the other five types of algorithms employed to predict the healing performance of autogenous healing concrete. Therefore, reliable prediction accuracy of the healing performance and efficient assistance on the design of autogenous healing concrete can be achieved.

Download Full-text

EXPRESS: Identifying mislabelled samples: machine learning models exceed human performance

Annals of Clinical Biochemistry International Journal of Laboratory Medicine ◽

10.1177/00045632211032991 ◽

2021 ◽

pp. 000456322110329

Author(s):

Christopher-John Lancaster Farrell

Keyword(s):

Machine Learning ◽

Human Performance ◽

Preliminary Investigation ◽

Gradient Boosting ◽

Support Vector ◽

Final Decision ◽

Laboratory Staff ◽

Clinical Laboratories ◽

Extreme Gradient Boosting ◽

Artificial Neural Network Ann

Background: It is difficult for clinical laboratories to identify samples that are labelled with the details of an incorrect patient. Many laboratories screen for these errors with delta checks, with final decision-making based on manual review of results by laboratory staff. Machine learning (ML) models have been shown to outperform delta checks for identifying these errors. However, a comparison of ML models to human-level performance has not yet been made. Methods: Deidentified data for current and previous (within seven days) electrolytes, urea and creatinine results was used in the computer simulation of mislabelled samples. Eight different ML models were developed on 127,256 sets of results using different algorithms: artificial neural network (ANN), extreme gradient boosting, support vector machine, random forest, logistic regression, k-nearest neighbours and two decision trees (one complex and one simple). A separate test dataset (n = 14,140) was used to evaluate the performance of these models as well as laboratory staff volunteers, who manually reviewed a random subset of this data (n = 500). Results: The best performing ML model was the ANN (92.1% accuracy), with the simple decision tree demonstrating the poorest accuracy (86.5%). The accuracy of laboratory staff for identifying mislabelled samples was 77.8%. Conclusions: The results of this preliminary investigation suggest that even relatively simple ML models can exceed human performance for identifying mislabelled samples. ML techniques should be considered for implementation in clinical laboratories to assist with error identification.

Download Full-text

Image and structured data analysis for prognostication of health outcomes in patients presenting to the Emergency Department during the COVID-19 pandemic

10.1101/2021.07.07.21260097 ◽

2021 ◽

Author(s):

Liam Butler ◽

Ibrahim Karabayir ◽

Mohammad Samie Tootooni ◽

Majid Afshar ◽

Ari Goldberg ◽

...

Keyword(s):

Machine Learning ◽

Emergency Department ◽

Intensive Care ◽

Deep Learning ◽

Clinical Data ◽

Chest Radiographs ◽

Gradient Boosting ◽

Learning Models ◽

Icu Admission ◽

Light Gradient

Background: Patients admitted to the emergency department (ED) with COVID-19 symptoms are routinely required to have chest radiographs and computed tomography (CT) scans. COVID-19 infection has been directly related to development of acute respiratory distress syndrome (ARDS) and severe infections lead to admission to intensive care and can also lead to death. The use of clinical data in machine learning models available at time of admission to ED can be used to assess possible risk of ARDS, need for intensive care unit (ICU) admission as well as risk of mortality. In addition, chest radiographs can be inputted into a deep learning model to further assess these risks. Purpose: This research aimed to develop machine and deep learning models using both structured clinical data and image data from the electronic health record (EHR) to adverse outcomes following ED admission. Materials and Methods: Light Gradient Boosting Machines (LightGBM) was used as the main machine learning algorithm using all clinical data including 42 variables. Compact models were also developed using 15 the most important variables to increase applicability of the models in clinical settings. To predict risk of the aforementioned health outcome events, transfer learning from the CheXNet model was implemented on our data as well. This research utilized clinical data and chest radiographs of 3571 patients 18 years and older admitted to the emergency department between 9th March 2020 and 29th October 2020 at Loyola University Medical Center. Main Findings: Our research results show that we can detect COVID-19 infection (AUC = 0.790 (0.746-0.835)) and predict the risk of developing ARDS (AUC = 0.781 (0.690-0.872), ICU admission (AUC = 0.675 (0.620-0.713)), and mortality (AUC = 0.759 (0.678-0.840)) at moderate accuracy from both chest X-ray images and clinical data. Principal Conclusions: The results can help in clinical decision making, especially when addressing ARDS and mortality, during the assessment of patients admitted to the ED with or without COVID-19 symptoms.

Download Full-text

Severity and Outcome Detection for the Cancer Patients with COVID-19 Using Machine Learning Models

10.21203/rs.3.rs-989924/v1 ◽

2021 ◽

Author(s):

Akbar Davoodi ◽

Shaghayegh Haghjooy Javanmard ◽

Golnaz Vaseghi ◽

Amirreza Manteghinejad

Keyword(s):

Machine Learning ◽

Intensive Care Unit ◽

Intensive Care ◽

Cancer Patients ◽

Cancer Center ◽

Support Vector ◽

Neutrophil Lymphocyte Ratio ◽

Icu Admission ◽

Reactive Protein ◽

Risk Of Death

Abstract Background:The COVID-19 pandemic challenges the healthcare system to provide enough resources to battle the pandemic without jeopardizing routine treatments. As a result, this is important that we can predict the outcomes of patients at the time of admission. This study aims to apply different machine learning (ML) models for predicting Intensive Care Unit (ICU) admission and mortality of Cancer Patients infected with COVID-19.Methods:This study's data were collected from a referral cancer center in Iran. The study included all patients with cancer and a confirmed diagnosis of COVID-19.Different ML prediction algorithms like Logistic Regression (LR), Naïve Bayes (NB), k-Nearest Neighbours (kNN), Random Forest (RF), and Support Vector Machine (SVM) were used. Also, we applied the SelectKBest method to find the most important features for predicting ICU admission and mortality.Results:Three hundred thirty-nine patients enrolled in the study. One hundred fifteen were admitted to the Intensive Care Unit (ICU), and 118 patients died during the hospital admission. The Area Under Curve (AUC) for predicting mortality is 0.61 for LR, 0.74 for NB, 0.61 for kNN, 0.6 for SVM, and 0.79 for RF. The AUC for predicting ICU admission is 0.61 for LR, 0.74 for NB, 0.56 for kNN, 0.55 for SVM, and 0.7 for RF.C-reactive protein (CRP), Aspartate transaminase (AST), and Neutrophil-Lymphocyte Ratio (NLR) also are the most common features in predicting ICU admission and mortality.Conclusion:Our findings show the promise of different AI methods for predicting the risk of death or ICU in cancer patients infected with COVID-19, highlighting the importance of first laboratory results and patients' symptoms.

Download Full-text

A Multimodal Approach for the Risk Prediction of Intensive Care and Mortality in Patients with COVID-19

Diagnostics ◽

10.3390/diagnostics12010056 ◽

2021 ◽

Vol 12 (1) ◽

pp. 56

Author(s):

Vasileios C. Pezoulas ◽

Konstantina D. Kourou ◽

Costas Papaloukas ◽

Vassiliki Triantafyllia ◽

Vicky Lampropoulou ◽

...

Keyword(s):

Risk Factors ◽

Intensive Care ◽

Biological Markers ◽

Lymphocyte Count ◽

Dynamic Bayesian Networks ◽

Gradient Boosting ◽

Multimodal Approach ◽

Icu Admission ◽

Explainable Ai ◽

Scoring Index

Background: Although several studies have been launched towards the prediction of risk factors for mortality and admission in the intensive care unit (ICU) in COVID-19, none of them focuses on the development of explainable AI models to define an ICU scoring index using dynamically associated biological markers. Methods: We propose a multimodal approach which combines explainable AI models with dynamic modeling methods to shed light into the clinical features of COVID-19. Dynamic Bayesian networks were used to seek associations among cytokines across four time intervals after hospitalization. Explainable gradient boosting trees were trained to predict the risk for ICU admission and mortality towards the development of an ICU scoring index. Results: Our results highlight LDH, IL-6, IL-8, Cr, number of monocytes, lymphocyte count, TNF as risk predictors for ICU admission and survival along with LDH, age, CRP, Cr, WBC, lymphocyte count for mortality in the ICU, with prediction accuracy 0.79 and 0.81, respectively. These risk factors were combined with dynamically associated biological markers to develop an ICU scoring index with accuracy 0.9. Conclusions: to our knowledge, this is the first multimodal and explainable AI model which quantifies the risk of intensive care with accuracy up to 0.9 across multiple timepoints.

Download Full-text

Clinical Predictive Models for COVID-19: Systematic Study (Preprint)

10.2196/preprints.21439 ◽

2020 ◽

Author(s):

Patrick Schwab ◽

August DuMont Schütte ◽

Benedikt Dietz ◽

Stefan Bauer

Keyword(s):

Machine Learning ◽

Health Care ◽

Intensive Care ◽

Clinical Data ◽

Predictive Models ◽

Health Care Systems ◽

Gradient Boosting ◽

Support Vector ◽

Care Systems ◽

To Receive

BACKGROUND COVID-19 is a rapidly emerging respiratory disease caused by SARS-CoV-2. Due to the rapid human-to-human transmission of SARS-CoV-2, many health care systems are at risk of exceeding their health care capacities, in particular in terms of SARS-CoV-2 tests, hospital and intensive care unit (ICU) beds, and mechanical ventilators. Predictive algorithms could potentially ease the strain on health care systems by identifying those who are most likely to receive a positive SARS-CoV-2 test, be hospitalized, or admitted to the ICU. OBJECTIVE The aim of this study is to develop, study, and evaluate clinical predictive models that estimate, using machine learning and based on routinely collected clinical data, which patients are likely to receive a positive SARS-CoV-2 test or require hospitalization or intensive care. METHODS Using a systematic approach to model development and optimization, we trained and compared various types of machine learning models, including logistic regression, neural networks, support vector machines, random forests, and gradient boosting. To evaluate the developed models, we performed a retrospective evaluation on demographic, clinical, and blood analysis data from a cohort of 5644 patients. In addition, we determined which clinical features were predictive to what degree for each of the aforementioned clinical tasks using causal explanations. RESULTS Our experimental results indicate that our predictive models identified patients that test positive for SARS-CoV-2 a priori at a sensitivity of 75% (95% CI 67%-81%) and a specificity of 49% (95% CI 46%-51%), patients who are SARS-CoV-2 positive that require hospitalization with 0.92 area under the receiver operator characteristic curve (AUC; 95% CI 0.81-0.98), and patients who are SARS-CoV-2 positive that require critical care with 0.98 AUC (95% CI 0.95-1.00). CONCLUSIONS Our results indicate that predictive models trained on routinely collected clinical data could be used to predict clinical pathways for COVID-19 and, therefore, help inform care and prioritize resources.

Download Full-text

Prediction of Hemorrhagic Transformation after Ischemic Stroke Using Machine Learning

Journal of Personalized Medicine ◽

10.3390/jpm11090863 ◽

2021 ◽

Vol 11 (9) ◽

pp. 863

Author(s):

Jeong-Myeong Choi ◽

Soo-Young Seo ◽

Pum-Jun Kim ◽

Yu-Seop Kim ◽

Sang-Hwa Lee ◽

...

Keyword(s):

Machine Learning ◽

Ischemic Stroke ◽

Binary Logistic Regression ◽

Hemorrhagic Transformation ◽

Structured Data ◽

Gradient Boosting ◽

Support Vector ◽

Test Dataset ◽

Extreme Gradient Boosting ◽

Artificial Neural Network Ann

Hemorrhagic transformation (HT) is one of the leading causes of a poor prognostic marker after acute ischemic stroke (AIS). We compared the performances of the several machine learning (ML) algorithms to predict HT after AIS using only structured data. A total of 2028 patients with AIS, who were admitted within seven days of symptoms onset, were included in this analysis. HT was defined based on the criteria of the European Co-operative Acute Stroke Study-II trial. The whole dataset was randomly divided into a training and a test dataset with a 7:3 ratio. Binary logistic regression, support vector machine, extreme gradient boosting, and artificial neural network (ANN) algorithms were used to assess the performance of predicting the HT occurrence after AIS. Five-fold cross validation and a grid search technique were used to optimize the hyperparameters of each ML model, which had its performance measured by the area under the receiver operating characteristic (AUROC) curve. Among the included AIS patients, the mean age and number of male subjects were 69.6 years and 1183 (58.3%), respectively. HT was observed in 318 subjects (15.7%). There were no significant differences in corresponding variables between the training and test dataset. Among all the ML algorithms, the ANN algorithm showed the best performance in terms of predicting the occurrence of HT in our dataset (0.844). Feature scaling including standardization and normalization, and the resampling strategy showed no additional improvement of the ANN’s performance. The ANN-based prediction of HT after AIS showed better performance than the conventional ML algorithms. Deep learning may be used to predict important outcomes for structured data-based prediction.

Download Full-text

Clinical Predictive Models for COVID-19: Systematic Study

Journal of Medical Internet Research ◽

10.2196/21439 ◽

2020 ◽

Vol 22 (10) ◽

pp. e21439 ◽

Cited By ~ 1

Author(s):

Patrick Schwab ◽

August DuMont Schütte ◽

Benedikt Dietz ◽

Stefan Bauer

Keyword(s):

Machine Learning ◽

Health Care ◽

Intensive Care ◽

Clinical Data ◽

Predictive Models ◽

Health Care Systems ◽

Gradient Boosting ◽

Support Vector ◽

Care Systems ◽

To Receive

Background COVID-19 is a rapidly emerging respiratory disease caused by SARS-CoV-2. Due to the rapid human-to-human transmission of SARS-CoV-2, many health care systems are at risk of exceeding their health care capacities, in particular in terms of SARS-CoV-2 tests, hospital and intensive care unit (ICU) beds, and mechanical ventilators. Predictive algorithms could potentially ease the strain on health care systems by identifying those who are most likely to receive a positive SARS-CoV-2 test, be hospitalized, or admitted to the ICU. Objective The aim of this study is to develop, study, and evaluate clinical predictive models that estimate, using machine learning and based on routinely collected clinical data, which patients are likely to receive a positive SARS-CoV-2 test or require hospitalization or intensive care. Methods Using a systematic approach to model development and optimization, we trained and compared various types of machine learning models, including logistic regression, neural networks, support vector machines, random forests, and gradient boosting. To evaluate the developed models, we performed a retrospective evaluation on demographic, clinical, and blood analysis data from a cohort of 5644 patients. In addition, we determined which clinical features were predictive to what degree for each of the aforementioned clinical tasks using causal explanations. Results Our experimental results indicate that our predictive models identified patients that test positive for SARS-CoV-2 a priori at a sensitivity of 75% (95% CI 67%-81%) and a specificity of 49% (95% CI 46%-51%), patients who are SARS-CoV-2 positive that require hospitalization with 0.92 area under the receiver operator characteristic curve (AUC; 95% CI 0.81-0.98), and patients who are SARS-CoV-2 positive that require critical care with 0.98 AUC (95% CI 0.95-1.00). Conclusions Our results indicate that predictive models trained on routinely collected clinical data could be used to predict clinical pathways for COVID-19 and, therefore, help inform care and prioritize resources.

Download Full-text

AKI Prediction Models in ICU: A Comparative Study (Preprint)

10.2196/preprints.18257 ◽

2020 ◽

Author(s):

Qing Qian ◽

Haixia Sun ◽

Jinming Wu ◽

Jiayang Wang ◽

Lei Yang

Keyword(s):

Kidney Disease ◽

Comparative Study ◽

Critically Ill ◽

Critically Ill Patients ◽

Gradient Boosting ◽

Support Vector ◽

Early Prediction ◽

Icu Admission ◽

Extreme Gradient Boosting ◽

Mimic Iii

BACKGROUND Acute kidney injury (AKI) is highly prevalent in critically ill patients and associated with significant morbidity and mortality as well as high financial costs. Early prediction of AKI provides an opportunity to develop strategies for early diagnosis, effective prevention, and timely treatment. Machine learning models have been developed for early prediction of AKI on critically ill patients by different researchers under different scenario. OBJECTIVE This comparative study aims to assess the performances of existing models for early prediction of AKI in the Intensive Care Unit (ICU) setting. METHODS The data was collected from the MIMIC-III database for all patients above 18 years old who had valid creatinine measured for 72 hours following ICU admission. Those with existing condition of kidney disease on admission were excluded. 17 predictor variables including patient demographics and physiological indicators were selected on the basis of the Kidney Disease Improving Global Outcomes (KDIGO) and medical literatures. Six models from three different types of methods were tested including Logistic Regression (LR), Support Vector Machines (SVM), Random Forest (RF), eXtreme Gradient Boosting (XGBoost), Light Gradient Boosting Decision (LightGBM), and Convolutional Neural Network (CNN). The area under ROC curve (AUC), accuracy, precision, recall and F1 value were calculated for each model to evaluate the performance. RESULTS We extracted 17205patient ICU records from MIMIC-III dataset. LightGBM had the best performance, with all the evaluation indicators achieved the highest (with average AUC 0.905, F1 0.897, Recall 0.836, P<.001). XGBoost had the second best performance (P<.001) and LR, RF, SVM performed similarly (P=0.082, 0.158, 0.710) on AUC. CNN got the lowest score on accuracy, precision, F1 and AUC. SVM and LR had relatively low recall than others. Creatinine were found to have the most significant effect on the early prediction of AKI onset in LR, RF, SVM and LightGBM. CONCLUSIONS LightGBM demonstrated the best predictive capability in predicting AKI present at the first 72 hours of ICU admission. LightGBM and XGBoost showed great potential for clinical application owing to their high recall. This study can provide references for AI-powered clinical decision support system for early AKI prediction in ICU setting.

Download Full-text