Comparison of 7 artificial intelligence models in predicting venous thromboembolism in COVID-19 patients

Background: An artificial intelligence (AI) approach can be used to predict venous thromboembolism (VTE). Objectives: To compare different AI models in predicting VTE using data from patients with COVID-19. Methods: We used feature ranking through recursive feature elimination with AI algorithms (lo...

Full description

Saved in:
Bibliographic Details
Main Authors: Indika Rajakaruna, Mohammad Hossein Amirhosseini, Mike Makris, Mike Laffan, Yang Li, Deepa J. Arachchillage
Format: Article
Language:English
Published: Elsevier 2025-02-01
Series:Research and Practice in Thrombosis and Haemostasis
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2475037925000354
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850156347423719424
author Indika Rajakaruna
Mohammad Hossein Amirhosseini
Mike Makris
Mike Laffan
Yang Li
Deepa J. Arachchillage
author_facet Indika Rajakaruna
Mohammad Hossein Amirhosseini
Mike Makris
Mike Laffan
Yang Li
Deepa J. Arachchillage
author_sort Indika Rajakaruna
collection DOAJ
description Background: An artificial intelligence (AI) approach can be used to predict venous thromboembolism (VTE). Objectives: To compare different AI models in predicting VTE using data from patients with COVID-19. Methods: We used feature ranking through recursive feature elimination with AI algorithms (logistic regression and random forest classifier) and standard statistical methods to identify the significant factors that contribute to developing VTE in COVID-19 patients using a large dataset from “Coagulopathy associated with COVID-19,” a multicenter observational study. We developed 7 AI models (Multilayer perceptron classifier, Artificial neural network with backpropagation, eXtreme gradient boosting, Support vector classifier, Stochastic gradient descent classifier, Random forest classifier and Logistic regression classifier) using the selected significant features to predict the development of VTE during hospitalization and used K-fold cross-validation and hyperparameter tuning to validate and optimize the models. The models’ predictive power was tested on 2649 (33% of 8027 overall patients), which were previously separated and not used during model training and validation stages. Results: Age, female sex, white ethnicity, comorbidities (diabetes, liver disease, autoimmune disease), and laboratory features (increased hemoglobin, white cell count, D-dimer, lactate dehydrogenase, ferritin), and presence of multiorgan failure were major factors associated with the development of thrombosis. Support vector classifier (SVC) model outperformed all other models, achieving an accuracy of 97%. The SVC model also led in precision (0.98), recall (0.97), and F1 score (0.97), and recorded the lowest log-loss score (0.112 on the test dataset), reflecting better model convergence and an improved fit to the data. Additionally, it achieved the highest area under the curve score (0.983). Conclusion: The SVC model delivered the best overall performance outperforming similar studies that developed deep learning and machine-learning models for COVID-19.
format Article
id doaj-art-e03c752d18b94217a12acdde5363f20a
institution OA Journals
issn 2475-0379
language English
publishDate 2025-02-01
publisher Elsevier
record_format Article
series Research and Practice in Thrombosis and Haemostasis
spelling doaj-art-e03c752d18b94217a12acdde5363f20a2025-08-20T02:24:34ZengElsevierResearch and Practice in Thrombosis and Haemostasis2475-03792025-02-019210271110.1016/j.rpth.2025.102711Comparison of 7 artificial intelligence models in predicting venous thromboembolism in COVID-19 patientsIndika Rajakaruna0Mohammad Hossein Amirhosseini1Mike Makris2Mike Laffan3Yang Li4Deepa J. Arachchillage5Department of Computer Science and Digital Technologies, University of East London, London, United KingdomDepartment of Computer Science and Digital Technologies, University of East London, London, United KingdomDepartment of Cardiovascular Sciences University of Sheffield, Sheffield, United KingdomDepartment of Immunology and Inflammation, Centre for Haematology, Imperial College London, London, United Kingdom; Department of Haematology, Imperial College Healthcare NHS Trust, London, United KingdomDepartment of Computer Science and Digital Technologies, University of East London, London, United KingdomDepartment of Immunology and Inflammation, Centre for Haematology, Imperial College London, London, United Kingdom; Department of Haematology, Imperial College Healthcare NHS Trust, London, United Kingdom; Correspondence Deepa R.J. Arachchillage, Centre for Haematology, Department of Immunology and Inflammation, Imperial College London, 5th Floor, Commonwealth Building, Du Cane Road, London, W12 0NN, UK.Background: An artificial intelligence (AI) approach can be used to predict venous thromboembolism (VTE). Objectives: To compare different AI models in predicting VTE using data from patients with COVID-19. Methods: We used feature ranking through recursive feature elimination with AI algorithms (logistic regression and random forest classifier) and standard statistical methods to identify the significant factors that contribute to developing VTE in COVID-19 patients using a large dataset from “Coagulopathy associated with COVID-19,” a multicenter observational study. We developed 7 AI models (Multilayer perceptron classifier, Artificial neural network with backpropagation, eXtreme gradient boosting, Support vector classifier, Stochastic gradient descent classifier, Random forest classifier and Logistic regression classifier) using the selected significant features to predict the development of VTE during hospitalization and used K-fold cross-validation and hyperparameter tuning to validate and optimize the models. The models’ predictive power was tested on 2649 (33% of 8027 overall patients), which were previously separated and not used during model training and validation stages. Results: Age, female sex, white ethnicity, comorbidities (diabetes, liver disease, autoimmune disease), and laboratory features (increased hemoglobin, white cell count, D-dimer, lactate dehydrogenase, ferritin), and presence of multiorgan failure were major factors associated with the development of thrombosis. Support vector classifier (SVC) model outperformed all other models, achieving an accuracy of 97%. The SVC model also led in precision (0.98), recall (0.97), and F1 score (0.97), and recorded the lowest log-loss score (0.112 on the test dataset), reflecting better model convergence and an improved fit to the data. Additionally, it achieved the highest area under the curve score (0.983). Conclusion: The SVC model delivered the best overall performance outperforming similar studies that developed deep learning and machine-learning models for COVID-19.http://www.sciencedirect.com/science/article/pii/S2475037925000354artificial intelligenceCOVID-19deep learningmachine learningthrombosis
spellingShingle Indika Rajakaruna
Mohammad Hossein Amirhosseini
Mike Makris
Mike Laffan
Yang Li
Deepa J. Arachchillage
Comparison of 7 artificial intelligence models in predicting venous thromboembolism in COVID-19 patients
Research and Practice in Thrombosis and Haemostasis
artificial intelligence
COVID-19
deep learning
machine learning
thrombosis
title Comparison of 7 artificial intelligence models in predicting venous thromboembolism in COVID-19 patients
title_full Comparison of 7 artificial intelligence models in predicting venous thromboembolism in COVID-19 patients
title_fullStr Comparison of 7 artificial intelligence models in predicting venous thromboembolism in COVID-19 patients
title_full_unstemmed Comparison of 7 artificial intelligence models in predicting venous thromboembolism in COVID-19 patients
title_short Comparison of 7 artificial intelligence models in predicting venous thromboembolism in COVID-19 patients
title_sort comparison of 7 artificial intelligence models in predicting venous thromboembolism in covid 19 patients
topic artificial intelligence
COVID-19
deep learning
machine learning
thrombosis
url http://www.sciencedirect.com/science/article/pii/S2475037925000354
work_keys_str_mv AT indikarajakaruna comparisonof7artificialintelligencemodelsinpredictingvenousthromboembolismincovid19patients
AT mohammadhosseinamirhosseini comparisonof7artificialintelligencemodelsinpredictingvenousthromboembolismincovid19patients
AT mikemakris comparisonof7artificialintelligencemodelsinpredictingvenousthromboembolismincovid19patients
AT mikelaffan comparisonof7artificialintelligencemodelsinpredictingvenousthromboembolismincovid19patients
AT yangli comparisonof7artificialintelligencemodelsinpredictingvenousthromboembolismincovid19patients
AT deepajarachchillage comparisonof7artificialintelligencemodelsinpredictingvenousthromboembolismincovid19patients