Mortality Prediction of COVID-19 Patients Using Radiomic and Neural Network Features Extracted from a Wide Chest X-ray Sample Size: A Robust Approach for Different Medical Imbalanced Scenarios
Aim: The aim of this study was to develop robust prognostic models for mortality prediction of COVID-19 patients, applicable to different sets of real scenarios, using radiomic and neural network features extracted from chest X-rays (CXRs) with a certified and commercially available software. Method...
| Main Authors: | , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-04-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/12/8/3903 |
| _version_ | 1797437048475353088 |
|---|---|
| author | Mauro Iori Carlo Di Castelnuovo Laura Verzellesi Greta Meglioli Davide Giosuè Lippolis Andrea Nitrosi Filippo Monelli Giulia Besutti Valeria Trojani Marco Bertolini Andrea Botti Gastone Castellani Daniel Remondini Roberto Sghedoni Stefania Croci Carlo Salvarani |
| author_facet | Mauro Iori Carlo Di Castelnuovo Laura Verzellesi Greta Meglioli Davide Giosuè Lippolis Andrea Nitrosi Filippo Monelli Giulia Besutti Valeria Trojani Marco Bertolini Andrea Botti Gastone Castellani Daniel Remondini Roberto Sghedoni Stefania Croci Carlo Salvarani |
| author_sort | Mauro Iori |
| collection | DOAJ |
| description | Aim: The aim of this study was to develop robust prognostic models for mortality prediction of COVID-19 patients, applicable to different sets of real scenarios, using radiomic and neural network features extracted from chest X-rays (CXRs) with a certified and commercially available software. Methods: 1816 patients from 5 different hospitals in the Province of Reggio Emilia were included in the study. Overall, 201 radiomic features and 16 neural network features were extracted from each COVID-19 patient’s radiography. The initial dataset was balanced to train the classifiers with the same number of dead and survived patients, randomly selected. The pipeline had three main parts: balancing procedure; three-step feature selection; and mortality prediction with radiomic features through three machine learning (ML) classification models: AdaBoost (ADA), Quadratic Discriminant Analysis (QDA) and Random Forest (RF). Five evaluation metrics were computed on the test samples. The performance for death prediction was validated on both a balanced dataset (Case 1) and an imbalanced dataset (Case 2). Results: accuracy (ACC), area under the ROC-curve (AUC) and sensitivity (SENS) for the best classifier were, respectively, 0.72 ± 0.01, 0.82 ± 0.02 and 0.84 ± 0.04 for Case 1 and 0.70 ± 0.04, 0.79 ± 0.03 and 0.76 ± 0.06 for Case 2. These results show that the prediction of COVID-19 mortality is robust in a different set of scenarios. Conclusions: Our large and varied dataset made it possible to train ML algorithms to predict COVID-19 mortality using radiomic and neural network features of CXRs. |
| first_indexed | 2024-03-09T11:12:23Z |
| format | Article |
| id | doaj.art-cd38cd393b094fd4ae3868e19ab6c866 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-09T11:12:23Z |
| publishDate | 2022-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-cd38cd393b094fd4ae3868e19ab6c8662023-12-01T00:41:13ZengMDPI AGApplied Sciences2076-34172022-04-01128390310.3390/app12083903Mortality Prediction of COVID-19 Patients Using Radiomic and Neural Network Features Extracted from a Wide Chest X-ray Sample Size: A Robust Approach for Different Medical Imbalanced ScenariosMauro Iori0Carlo Di Castelnuovo1Laura Verzellesi2Greta Meglioli3Davide Giosuè Lippolis4Andrea Nitrosi5Filippo Monelli6Giulia Besutti7Valeria Trojani8Marco Bertolini9Andrea Botti10Gastone Castellani11Daniel Remondini12Roberto Sghedoni13Stefania Croci14Carlo Salvarani15Medical Physics Unit, Azienda USL-IRCCS di Reggio Emilia, 42123 Reggio Emilia, ItalyMedical Physics Unit, Azienda USL-IRCCS di Reggio Emilia, 42123 Reggio Emilia, ItalyDepartment of Physics and Astronomy-DIFA, University of Bologna, 40126 Bologna, ItalyMedical Physics Unit, Azienda USL-IRCCS di Reggio Emilia, 42123 Reggio Emilia, ItalyMedical Physics Unit, Azienda USL-IRCCS di Reggio Emilia, 42123 Reggio Emilia, ItalyMedical Physics Unit, Azienda USL-IRCCS di Reggio Emilia, 42123 Reggio Emilia, ItalyRadiology Unit, Azienda USL-IRCCS di Reggio Emilia, 42123 Reggio Emilia, ItalyRadiology Unit, Azienda USL-IRCCS di Reggio Emilia, 42123 Reggio Emilia, ItalyMedical Physics Unit, Azienda USL-IRCCS di Reggio Emilia, 42123 Reggio Emilia, ItalyMedical Physics Unit, Azienda USL-IRCCS di Reggio Emilia, 42123 Reggio Emilia, ItalyMedical Physics Unit, Azienda USL-IRCCS di Reggio Emilia, 42123 Reggio Emilia, ItalyDepartment of Experimental, Diagnostic and Specialty Medicine–DIMES, 40126 Bologna, ItalyDepartment of Physics and Astronomy-DIFA, University of Bologna, 40126 Bologna, ItalyMedical Physics Unit, Azienda USL-IRCCS di Reggio Emilia, 42123 Reggio Emilia, ItalyClinical Immunology, Allergy and Advanced Biotechnologies Unit, Azienda USL-IRCCS di Reggio Emilia, 42123 Reggio Emilia, ItalyRheumatology Unit, Azienda USL-IRCCS di Reggio Emilia, 42123 Reggio Emilia, ItalyAim: The aim of this study was to develop robust prognostic models for mortality prediction of COVID-19 patients, applicable to different sets of real scenarios, using radiomic and neural network features extracted from chest X-rays (CXRs) with a certified and commercially available software. Methods: 1816 patients from 5 different hospitals in the Province of Reggio Emilia were included in the study. Overall, 201 radiomic features and 16 neural network features were extracted from each COVID-19 patient’s radiography. The initial dataset was balanced to train the classifiers with the same number of dead and survived patients, randomly selected. The pipeline had three main parts: balancing procedure; three-step feature selection; and mortality prediction with radiomic features through three machine learning (ML) classification models: AdaBoost (ADA), Quadratic Discriminant Analysis (QDA) and Random Forest (RF). Five evaluation metrics were computed on the test samples. The performance for death prediction was validated on both a balanced dataset (Case 1) and an imbalanced dataset (Case 2). Results: accuracy (ACC), area under the ROC-curve (AUC) and sensitivity (SENS) for the best classifier were, respectively, 0.72 ± 0.01, 0.82 ± 0.02 and 0.84 ± 0.04 for Case 1 and 0.70 ± 0.04, 0.79 ± 0.03 and 0.76 ± 0.06 for Case 2. These results show that the prediction of COVID-19 mortality is robust in a different set of scenarios. Conclusions: Our large and varied dataset made it possible to train ML algorithms to predict COVID-19 mortality using radiomic and neural network features of CXRs.https://www.mdpi.com/2076-3417/12/8/3903machine learningradiomicsCOVID-19X-ray radiographyunder-sampling |
| spellingShingle | Mauro Iori Carlo Di Castelnuovo Laura Verzellesi Greta Meglioli Davide Giosuè Lippolis Andrea Nitrosi Filippo Monelli Giulia Besutti Valeria Trojani Marco Bertolini Andrea Botti Gastone Castellani Daniel Remondini Roberto Sghedoni Stefania Croci Carlo Salvarani Mortality Prediction of COVID-19 Patients Using Radiomic and Neural Network Features Extracted from a Wide Chest X-ray Sample Size: A Robust Approach for Different Medical Imbalanced Scenarios Applied Sciences machine learning radiomics COVID-19 X-ray radiography under-sampling |
| title | Mortality Prediction of COVID-19 Patients Using Radiomic and Neural Network Features Extracted from a Wide Chest X-ray Sample Size: A Robust Approach for Different Medical Imbalanced Scenarios |
| title_full | Mortality Prediction of COVID-19 Patients Using Radiomic and Neural Network Features Extracted from a Wide Chest X-ray Sample Size: A Robust Approach for Different Medical Imbalanced Scenarios |
| title_fullStr | Mortality Prediction of COVID-19 Patients Using Radiomic and Neural Network Features Extracted from a Wide Chest X-ray Sample Size: A Robust Approach for Different Medical Imbalanced Scenarios |
| title_full_unstemmed | Mortality Prediction of COVID-19 Patients Using Radiomic and Neural Network Features Extracted from a Wide Chest X-ray Sample Size: A Robust Approach for Different Medical Imbalanced Scenarios |
| title_short | Mortality Prediction of COVID-19 Patients Using Radiomic and Neural Network Features Extracted from a Wide Chest X-ray Sample Size: A Robust Approach for Different Medical Imbalanced Scenarios |
| title_sort | mortality prediction of covid 19 patients using radiomic and neural network features extracted from a wide chest x ray sample size a robust approach for different medical imbalanced scenarios |
| topic | machine learning radiomics COVID-19 X-ray radiography under-sampling |
| url | https://www.mdpi.com/2076-3417/12/8/3903 |
| work_keys_str_mv | AT mauroiori mortalitypredictionofcovid19patientsusingradiomicandneuralnetworkfeaturesextractedfromawidechestxraysamplesizearobustapproachfordifferentmedicalimbalancedscenarios AT carlodicastelnuovo mortalitypredictionofcovid19patientsusingradiomicandneuralnetworkfeaturesextractedfromawidechestxraysamplesizearobustapproachfordifferentmedicalimbalancedscenarios AT lauraverzellesi mortalitypredictionofcovid19patientsusingradiomicandneuralnetworkfeaturesextractedfromawidechestxraysamplesizearobustapproachfordifferentmedicalimbalancedscenarios AT gretameglioli mortalitypredictionofcovid19patientsusingradiomicandneuralnetworkfeaturesextractedfromawidechestxraysamplesizearobustapproachfordifferentmedicalimbalancedscenarios AT davidegiosuelippolis mortalitypredictionofcovid19patientsusingradiomicandneuralnetworkfeaturesextractedfromawidechestxraysamplesizearobustapproachfordifferentmedicalimbalancedscenarios AT andreanitrosi mortalitypredictionofcovid19patientsusingradiomicandneuralnetworkfeaturesextractedfromawidechestxraysamplesizearobustapproachfordifferentmedicalimbalancedscenarios AT filippomonelli mortalitypredictionofcovid19patientsusingradiomicandneuralnetworkfeaturesextractedfromawidechestxraysamplesizearobustapproachfordifferentmedicalimbalancedscenarios AT giuliabesutti mortalitypredictionofcovid19patientsusingradiomicandneuralnetworkfeaturesextractedfromawidechestxraysamplesizearobustapproachfordifferentmedicalimbalancedscenarios AT valeriatrojani mortalitypredictionofcovid19patientsusingradiomicandneuralnetworkfeaturesextractedfromawidechestxraysamplesizearobustapproachfordifferentmedicalimbalancedscenarios AT marcobertolini mortalitypredictionofcovid19patientsusingradiomicandneuralnetworkfeaturesextractedfromawidechestxraysamplesizearobustapproachfordifferentmedicalimbalancedscenarios AT andreabotti mortalitypredictionofcovid19patientsusingradiomicandneuralnetworkfeaturesextractedfromawidechestxraysamplesizearobustapproachfordifferentmedicalimbalancedscenarios AT gastonecastellani mortalitypredictionofcovid19patientsusingradiomicandneuralnetworkfeaturesextractedfromawidechestxraysamplesizearobustapproachfordifferentmedicalimbalancedscenarios AT danielremondini mortalitypredictionofcovid19patientsusingradiomicandneuralnetworkfeaturesextractedfromawidechestxraysamplesizearobustapproachfordifferentmedicalimbalancedscenarios AT robertosghedoni mortalitypredictionofcovid19patientsusingradiomicandneuralnetworkfeaturesextractedfromawidechestxraysamplesizearobustapproachfordifferentmedicalimbalancedscenarios AT stefaniacroci mortalitypredictionofcovid19patientsusingradiomicandneuralnetworkfeaturesextractedfromawidechestxraysamplesizearobustapproachfordifferentmedicalimbalancedscenarios AT carlosalvarani mortalitypredictionofcovid19patientsusingradiomicandneuralnetworkfeaturesextractedfromawidechestxraysamplesizearobustapproachfordifferentmedicalimbalancedscenarios |