Virtual Monoenergetic Images of Dual-Energy CT—Impact on Repeatability, Reproducibility, and Classification in Radiomics
The purpose of this study was to (i) evaluate the test–retest repeatability and reproducibility of radiomic features in virtual monoenergetic images (VMI) from dual-energy CT (DECT) depending on VMI energy (40, 50, 75, 120, 190 keV), radiation dose (5 and 15 mGy), and DECT approach (dual-source and...
Main Authors: | , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-09-01
|
Series: | Cancers |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-6694/13/18/4710 |
id |
doaj-7974033699bc482bba4cb36d2291b7b9 |
---|---|
record_format |
Article |
spelling |
doaj-7974033699bc482bba4cb36d2291b7b92021-09-25T23:50:34ZengMDPI AGCancers2072-66942021-09-01134710471010.3390/cancers13184710Virtual Monoenergetic Images of Dual-Energy CT—Impact on Repeatability, Reproducibility, and Classification in RadiomicsAndré Euler0Fabian Christopher Laqua1Davide Cester2Niklas Lohaus3Thomas Sartoretti4Daniel Pinto dos Santos5Hatem Alkadhi6Bettina Baessler7Institute of Diagnostic and Interventional Radiology, University Hospital Zurich, University of Zurich, Raemistrasse 100, 8091 Zurich, SwitzerlandInstitute of Diagnostic and Interventional Radiology, University Hospital Zurich, University of Zurich, Raemistrasse 100, 8091 Zurich, SwitzerlandInstitute of Diagnostic and Interventional Radiology, University Hospital Zurich, University of Zurich, Raemistrasse 100, 8091 Zurich, SwitzerlandInstitute of Diagnostic and Interventional Radiology, University Hospital Zurich, University of Zurich, Raemistrasse 100, 8091 Zurich, SwitzerlandInstitute of Diagnostic and Interventional Radiology, University Hospital Zurich, University of Zurich, Raemistrasse 100, 8091 Zurich, SwitzerlandMedical Faculty, University Hospital Cologne, Institute of Diagnostic and Interventional Radiology, University of Cologne, Kerpener Str. 62, 50937 Cologne, GermanyInstitute of Diagnostic and Interventional Radiology, University Hospital Zurich, University of Zurich, Raemistrasse 100, 8091 Zurich, SwitzerlandInstitute of Diagnostic and Interventional Radiology, University Hospital Zurich, University of Zurich, Raemistrasse 100, 8091 Zurich, SwitzerlandThe purpose of this study was to (i) evaluate the test–retest repeatability and reproducibility of radiomic features in virtual monoenergetic images (VMI) from dual-energy CT (DECT) depending on VMI energy (40, 50, 75, 120, 190 keV), radiation dose (5 and 15 mGy), and DECT approach (dual-source and split-filter DECT) in a phantom (ex vivo), and (ii) to assess the impact of VMI energy and feature repeatability on machine-learning-based classification in vivo in 72 patients with 72 hypodense liver lesions. Feature repeatability and reproducibility were determined by concordance–correlation–coefficient (CCC) and dynamic range (DR) ≥0.9. Test–retest repeatability was high within the same VMI energies and scan conditions (percentage of repeatable features ranging from 74% for SFDE mode at 40 keV and 15 mGy to 86% for DSDE at 190 keV and 15 mGy), while reproducibility varied substantially across different VMI energies and DECTs (percentage of reproducible features ranging from 32.8% for SFDE at 5 mGy comparing 40 with 190 keV to 99.2% for DSDE at 15 mGy comparing 40 with 50 keV). No major differences were observed between the two radiation doses (<10%) in all pair-wise comparisons. In vivo, machine learning classification using penalized regression and random forests resulted in the best discrimination of hemangiomas and metastases at low-energy VMI (40 keV), and for cysts at high-energy VMI (120 keV). Feature selection based on feature repeatability did not improve classification performance. Our results demonstrate the high repeatability of radiomics features when keeping scan and reconstruction conditions constant. Reproducibility diminished when using different VMI energies or DECT approaches. The choice of optimal VMI energy improved lesion classification in vivo and should hence be adapted to the specific task.https://www.mdpi.com/2072-6694/13/18/4710radiomicsspectral CTdual-energy CToncologymachine learningcomputed tomography |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
André Euler Fabian Christopher Laqua Davide Cester Niklas Lohaus Thomas Sartoretti Daniel Pinto dos Santos Hatem Alkadhi Bettina Baessler |
spellingShingle |
André Euler Fabian Christopher Laqua Davide Cester Niklas Lohaus Thomas Sartoretti Daniel Pinto dos Santos Hatem Alkadhi Bettina Baessler Virtual Monoenergetic Images of Dual-Energy CT—Impact on Repeatability, Reproducibility, and Classification in Radiomics Cancers radiomics spectral CT dual-energy CT oncology machine learning computed tomography |
author_facet |
André Euler Fabian Christopher Laqua Davide Cester Niklas Lohaus Thomas Sartoretti Daniel Pinto dos Santos Hatem Alkadhi Bettina Baessler |
author_sort |
André Euler |
title |
Virtual Monoenergetic Images of Dual-Energy CT—Impact on Repeatability, Reproducibility, and Classification in Radiomics |
title_short |
Virtual Monoenergetic Images of Dual-Energy CT—Impact on Repeatability, Reproducibility, and Classification in Radiomics |
title_full |
Virtual Monoenergetic Images of Dual-Energy CT—Impact on Repeatability, Reproducibility, and Classification in Radiomics |
title_fullStr |
Virtual Monoenergetic Images of Dual-Energy CT—Impact on Repeatability, Reproducibility, and Classification in Radiomics |
title_full_unstemmed |
Virtual Monoenergetic Images of Dual-Energy CT—Impact on Repeatability, Reproducibility, and Classification in Radiomics |
title_sort |
virtual monoenergetic images of dual-energy ct—impact on repeatability, reproducibility, and classification in radiomics |
publisher |
MDPI AG |
series |
Cancers |
issn |
2072-6694 |
publishDate |
2021-09-01 |
description |
The purpose of this study was to (i) evaluate the test–retest repeatability and reproducibility of radiomic features in virtual monoenergetic images (VMI) from dual-energy CT (DECT) depending on VMI energy (40, 50, 75, 120, 190 keV), radiation dose (5 and 15 mGy), and DECT approach (dual-source and split-filter DECT) in a phantom (ex vivo), and (ii) to assess the impact of VMI energy and feature repeatability on machine-learning-based classification in vivo in 72 patients with 72 hypodense liver lesions. Feature repeatability and reproducibility were determined by concordance–correlation–coefficient (CCC) and dynamic range (DR) ≥0.9. Test–retest repeatability was high within the same VMI energies and scan conditions (percentage of repeatable features ranging from 74% for SFDE mode at 40 keV and 15 mGy to 86% for DSDE at 190 keV and 15 mGy), while reproducibility varied substantially across different VMI energies and DECTs (percentage of reproducible features ranging from 32.8% for SFDE at 5 mGy comparing 40 with 190 keV to 99.2% for DSDE at 15 mGy comparing 40 with 50 keV). No major differences were observed between the two radiation doses (<10%) in all pair-wise comparisons. In vivo, machine learning classification using penalized regression and random forests resulted in the best discrimination of hemangiomas and metastases at low-energy VMI (40 keV), and for cysts at high-energy VMI (120 keV). Feature selection based on feature repeatability did not improve classification performance. Our results demonstrate the high repeatability of radiomics features when keeping scan and reconstruction conditions constant. Reproducibility diminished when using different VMI energies or DECT approaches. The choice of optimal VMI energy improved lesion classification in vivo and should hence be adapted to the specific task. |
topic |
radiomics spectral CT dual-energy CT oncology machine learning computed tomography |
url |
https://www.mdpi.com/2072-6694/13/18/4710 |
work_keys_str_mv |
AT andreeuler virtualmonoenergeticimagesofdualenergyctimpactonrepeatabilityreproducibilityandclassificationinradiomics AT fabianchristopherlaqua virtualmonoenergeticimagesofdualenergyctimpactonrepeatabilityreproducibilityandclassificationinradiomics AT davidecester virtualmonoenergeticimagesofdualenergyctimpactonrepeatabilityreproducibilityandclassificationinradiomics AT niklaslohaus virtualmonoenergeticimagesofdualenergyctimpactonrepeatabilityreproducibilityandclassificationinradiomics AT thomassartoretti virtualmonoenergeticimagesofdualenergyctimpactonrepeatabilityreproducibilityandclassificationinradiomics AT danielpintodossantos virtualmonoenergeticimagesofdualenergyctimpactonrepeatabilityreproducibilityandclassificationinradiomics AT hatemalkadhi virtualmonoenergeticimagesofdualenergyctimpactonrepeatabilityreproducibilityandclassificationinradiomics AT bettinabaessler virtualmonoenergeticimagesofdualenergyctimpactonrepeatabilityreproducibilityandclassificationinradiomics |
_version_ |
1717367763900563456 |