Radiomics side experiments and DAFIT approach in identifying pulmonary hypertension using Cardiac MRI derived radiomics based machine learning models
Abstract Side experiments are performed on radiomics models to improve their reproducibility. We measure the impact of myocardial masks, radiomic side experiments and data augmentation for information transfer (DAFIT) approach to differentiate patients with and without pulmonary hypertension (PH) us...
Main Authors: | , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Nature Publishing Group
2021-06-01
|
Series: | Scientific Reports |
Online Access: | https://doi.org/10.1038/s41598-021-92155-6 |
id |
doaj-e3a8773830724016ab05eb04865c8652 |
---|---|
record_format |
Article |
spelling |
doaj-e3a8773830724016ab05eb04865c86522021-06-20T11:36:13ZengNature Publishing GroupScientific Reports2045-23222021-06-0111111310.1038/s41598-021-92155-6Radiomics side experiments and DAFIT approach in identifying pulmonary hypertension using Cardiac MRI derived radiomics based machine learning modelsSarv Priya0Tanya Aggarwal1Caitlin Ward2Girish Bathla3Mathews Jacob4Alicia Gerke5Eric A. Hoffman6Prashant Nagpal7Department of Radiology, University of Iowa Carver College of MedicineDepartment of Family Medicine, University of Iowa Carver College of MedicineDepartment of Biostatistics, University of Iowa College of Public HealthDepartment of Radiology, University of Iowa Carver College of MedicineDepartment of Electrical Engineering, University of Iowa College of EngineeringDepartment of Pulmonary Medicine, University of Iowa Carver College of MedicineDepartment of Radiology, University of Iowa Carver College of MedicineDepartment of Radiology, University of Iowa Carver College of MedicineAbstract Side experiments are performed on radiomics models to improve their reproducibility. We measure the impact of myocardial masks, radiomic side experiments and data augmentation for information transfer (DAFIT) approach to differentiate patients with and without pulmonary hypertension (PH) using cardiac MRI (CMRI) derived radiomics. Feature extraction was performed from the left ventricle (LV) and right ventricle (RV) myocardial masks using CMRI in 82 patients (42 PH and 40 controls). Various side study experiments were evaluated: Original data without and with intraclass correlation (ICC) feature-filtering and DAFIT approach (without and with ICC feature-filtering). Multiple machine learning and feature selection strategies were evaluated. Primary analysis included all PH patients with subgroup analysis including PH patients with preserved LVEF (≥ 50%). For both primary and subgroup analysis, DAFIT approach without feature-filtering was the highest performer (AUC 0.957–0.958). ICC approaches showed poor performance compared to DAFIT approach. The performance of combined LV and RV masks was superior to individual masks alone. There was variation in top performing models across all approaches (AUC 0.862–0.958). DAFIT approach with features from combined LV and RV masks provide superior performance with poor performance of feature filtering approaches. Model performance varies based upon the feature selection and model combination.https://doi.org/10.1038/s41598-021-92155-6 |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Sarv Priya Tanya Aggarwal Caitlin Ward Girish Bathla Mathews Jacob Alicia Gerke Eric A. Hoffman Prashant Nagpal |
spellingShingle |
Sarv Priya Tanya Aggarwal Caitlin Ward Girish Bathla Mathews Jacob Alicia Gerke Eric A. Hoffman Prashant Nagpal Radiomics side experiments and DAFIT approach in identifying pulmonary hypertension using Cardiac MRI derived radiomics based machine learning models Scientific Reports |
author_facet |
Sarv Priya Tanya Aggarwal Caitlin Ward Girish Bathla Mathews Jacob Alicia Gerke Eric A. Hoffman Prashant Nagpal |
author_sort |
Sarv Priya |
title |
Radiomics side experiments and DAFIT approach in identifying pulmonary hypertension using Cardiac MRI derived radiomics based machine learning models |
title_short |
Radiomics side experiments and DAFIT approach in identifying pulmonary hypertension using Cardiac MRI derived radiomics based machine learning models |
title_full |
Radiomics side experiments and DAFIT approach in identifying pulmonary hypertension using Cardiac MRI derived radiomics based machine learning models |
title_fullStr |
Radiomics side experiments and DAFIT approach in identifying pulmonary hypertension using Cardiac MRI derived radiomics based machine learning models |
title_full_unstemmed |
Radiomics side experiments and DAFIT approach in identifying pulmonary hypertension using Cardiac MRI derived radiomics based machine learning models |
title_sort |
radiomics side experiments and dafit approach in identifying pulmonary hypertension using cardiac mri derived radiomics based machine learning models |
publisher |
Nature Publishing Group |
series |
Scientific Reports |
issn |
2045-2322 |
publishDate |
2021-06-01 |
description |
Abstract Side experiments are performed on radiomics models to improve their reproducibility. We measure the impact of myocardial masks, radiomic side experiments and data augmentation for information transfer (DAFIT) approach to differentiate patients with and without pulmonary hypertension (PH) using cardiac MRI (CMRI) derived radiomics. Feature extraction was performed from the left ventricle (LV) and right ventricle (RV) myocardial masks using CMRI in 82 patients (42 PH and 40 controls). Various side study experiments were evaluated: Original data without and with intraclass correlation (ICC) feature-filtering and DAFIT approach (without and with ICC feature-filtering). Multiple machine learning and feature selection strategies were evaluated. Primary analysis included all PH patients with subgroup analysis including PH patients with preserved LVEF (≥ 50%). For both primary and subgroup analysis, DAFIT approach without feature-filtering was the highest performer (AUC 0.957–0.958). ICC approaches showed poor performance compared to DAFIT approach. The performance of combined LV and RV masks was superior to individual masks alone. There was variation in top performing models across all approaches (AUC 0.862–0.958). DAFIT approach with features from combined LV and RV masks provide superior performance with poor performance of feature filtering approaches. Model performance varies based upon the feature selection and model combination. |
url |
https://doi.org/10.1038/s41598-021-92155-6 |
work_keys_str_mv |
AT sarvpriya radiomicssideexperimentsanddafitapproachinidentifyingpulmonaryhypertensionusingcardiacmriderivedradiomicsbasedmachinelearningmodels AT tanyaaggarwal radiomicssideexperimentsanddafitapproachinidentifyingpulmonaryhypertensionusingcardiacmriderivedradiomicsbasedmachinelearningmodels AT caitlinward radiomicssideexperimentsanddafitapproachinidentifyingpulmonaryhypertensionusingcardiacmriderivedradiomicsbasedmachinelearningmodels AT girishbathla radiomicssideexperimentsanddafitapproachinidentifyingpulmonaryhypertensionusingcardiacmriderivedradiomicsbasedmachinelearningmodels AT mathewsjacob radiomicssideexperimentsanddafitapproachinidentifyingpulmonaryhypertensionusingcardiacmriderivedradiomicsbasedmachinelearningmodels AT aliciagerke radiomicssideexperimentsanddafitapproachinidentifyingpulmonaryhypertensionusingcardiacmriderivedradiomicsbasedmachinelearningmodels AT ericahoffman radiomicssideexperimentsanddafitapproachinidentifyingpulmonaryhypertensionusingcardiacmriderivedradiomicsbasedmachinelearningmodels AT prashantnagpal radiomicssideexperimentsanddafitapproachinidentifyingpulmonaryhypertensionusingcardiacmriderivedradiomicsbasedmachinelearningmodels |
_version_ |
1721369924994072576 |