Toward Closing the Loop in Image-to-Image Conversion in Radiotherapy: A Quality Control Tool to Predict Synthetic Computed Tomography Hounsfield Unit Accuracy

In recent years, synthetic Computed Tomography (CT) images generated from Magnetic Resonance (MR) or Cone Beam Computed Tomography (CBCT) acquisitions have been shown to be comparable to real CT images in terms of dose computation for radiotherapy simulation. However, until now, there has been no in...

Full description

Saved in:
Bibliographic Details
Main Authors: Paolo Zaffino, Ciro Benito Raggio, Adrian Thummerer, Gabriel Guterres Marmitt, Johannes Albertus Langendijk, Anna Procopio, Carlo Cosentino, Joao Seco, Antje Christin Knopf, Stefan Both, Maria Francesca Spadea
Format: Article
Language:English
Published: MDPI AG 2024-12-01
Series:Journal of Imaging
Subjects:
Online Access:https://www.mdpi.com/2313-433X/10/12/316
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850050277210587136
author Paolo Zaffino
Ciro Benito Raggio
Adrian Thummerer
Gabriel Guterres Marmitt
Johannes Albertus Langendijk
Anna Procopio
Carlo Cosentino
Joao Seco
Antje Christin Knopf
Stefan Both
Maria Francesca Spadea
author_facet Paolo Zaffino
Ciro Benito Raggio
Adrian Thummerer
Gabriel Guterres Marmitt
Johannes Albertus Langendijk
Anna Procopio
Carlo Cosentino
Joao Seco
Antje Christin Knopf
Stefan Both
Maria Francesca Spadea
author_sort Paolo Zaffino
collection DOAJ
description In recent years, synthetic Computed Tomography (CT) images generated from Magnetic Resonance (MR) or Cone Beam Computed Tomography (CBCT) acquisitions have been shown to be comparable to real CT images in terms of dose computation for radiotherapy simulation. However, until now, there has been no independent strategy to assess the quality of each synthetic image in the absence of ground truth. In this work, we propose a Deep Learning (DL)-based framework to predict the accuracy of synthetic CT in terms of Mean Absolute Error (MAE) without the need for a ground truth (GT). The proposed algorithm generates a volumetric map as an output, informing clinicians of the predicted MAE slice-by-slice. A cascading multi-model architecture was used to deal with the complexity of the MAE prediction task. The workflow was trained and tested on two cohorts of head and neck cancer patients with different imaging modalities: 27 MR scans and 33 CBCT. The algorithm evaluation revealed an accurate HU prediction (a median absolute prediction deviation equal to 4 HU for CBCT-based synthetic CTs and 6 HU for MR-based synthetic CTs), with discrepancies that do not affect the clinical decisions made on the basis of the proposed estimation. The workflow exhibited no systematic error in MAE prediction. This work represents a proof of concept about the feasibility of synthetic CT evaluation in daily clinical practice, and it paves the way for future patient-specific quality assessment strategies.
format Article
id doaj-art-abd04604530d4137bbb54fdb67123655
institution DOAJ
issn 2313-433X
language English
publishDate 2024-12-01
publisher MDPI AG
record_format Article
series Journal of Imaging
spelling doaj-art-abd04604530d4137bbb54fdb671236552025-08-20T02:53:30ZengMDPI AGJournal of Imaging2313-433X2024-12-01101231610.3390/jimaging10120316Toward Closing the Loop in Image-to-Image Conversion in Radiotherapy: A Quality Control Tool to Predict Synthetic Computed Tomography Hounsfield Unit AccuracyPaolo Zaffino0Ciro Benito Raggio1Adrian Thummerer2Gabriel Guterres Marmitt3Johannes Albertus Langendijk4Anna Procopio5Carlo Cosentino6Joao Seco7Antje Christin Knopf8Stefan Both9Maria Francesca Spadea10Department of Experimental and Clinical Medicine, Magna Graecia University, viale Europa, 88100 Catanzaro, ItalyInstitute of Biomedical Engineering, Karlsruhe Institute of Technology (KIT), 76131 Karlsruhe, GermanyDepartment of Radiation Oncology, University Medical Center Groningen, University of Groningen, 9712 CP Groningen, The NetherlandsDepartment of Radiation Oncology, University Medical Center Groningen, University of Groningen, 9712 CP Groningen, The NetherlandsDepartment of Radiation Oncology, University Medical Center Groningen, University of Groningen, 9712 CP Groningen, The NetherlandsDepartment of Experimental and Clinical Medicine, Magna Graecia University, viale Europa, 88100 Catanzaro, ItalyDepartment of Experimental and Clinical Medicine, Magna Graecia University, viale Europa, 88100 Catanzaro, ItalyDepartment of Biomedical Physics in Radiation Oncology, Deutsches Krebsfoschungszentrum (DKFZ), 69120 Heidelberg, GermanyInstitute for Medical Engineering and Medical Informatics, School of Life Science FHNW, 4132 Muttenz, SwitzerlandDepartment of Radiation Oncology, University Medical Center Groningen, University of Groningen, 9712 CP Groningen, The NetherlandsInstitute of Biomedical Engineering, Karlsruhe Institute of Technology (KIT), 76131 Karlsruhe, GermanyIn recent years, synthetic Computed Tomography (CT) images generated from Magnetic Resonance (MR) or Cone Beam Computed Tomography (CBCT) acquisitions have been shown to be comparable to real CT images in terms of dose computation for radiotherapy simulation. However, until now, there has been no independent strategy to assess the quality of each synthetic image in the absence of ground truth. In this work, we propose a Deep Learning (DL)-based framework to predict the accuracy of synthetic CT in terms of Mean Absolute Error (MAE) without the need for a ground truth (GT). The proposed algorithm generates a volumetric map as an output, informing clinicians of the predicted MAE slice-by-slice. A cascading multi-model architecture was used to deal with the complexity of the MAE prediction task. The workflow was trained and tested on two cohorts of head and neck cancer patients with different imaging modalities: 27 MR scans and 33 CBCT. The algorithm evaluation revealed an accurate HU prediction (a median absolute prediction deviation equal to 4 HU for CBCT-based synthetic CTs and 6 HU for MR-based synthetic CTs), with discrepancies that do not affect the clinical decisions made on the basis of the proposed estimation. The workflow exhibited no systematic error in MAE prediction. This work represents a proof of concept about the feasibility of synthetic CT evaluation in daily clinical practice, and it paves the way for future patient-specific quality assessment strategies.https://www.mdpi.com/2313-433X/10/12/316synthetic CTconversion predictionMR-only adaptive radiotherapydeep learning
spellingShingle Paolo Zaffino
Ciro Benito Raggio
Adrian Thummerer
Gabriel Guterres Marmitt
Johannes Albertus Langendijk
Anna Procopio
Carlo Cosentino
Joao Seco
Antje Christin Knopf
Stefan Both
Maria Francesca Spadea
Toward Closing the Loop in Image-to-Image Conversion in Radiotherapy: A Quality Control Tool to Predict Synthetic Computed Tomography Hounsfield Unit Accuracy
Journal of Imaging
synthetic CT
conversion prediction
MR-only adaptive radiotherapy
deep learning
title Toward Closing the Loop in Image-to-Image Conversion in Radiotherapy: A Quality Control Tool to Predict Synthetic Computed Tomography Hounsfield Unit Accuracy
title_full Toward Closing the Loop in Image-to-Image Conversion in Radiotherapy: A Quality Control Tool to Predict Synthetic Computed Tomography Hounsfield Unit Accuracy
title_fullStr Toward Closing the Loop in Image-to-Image Conversion in Radiotherapy: A Quality Control Tool to Predict Synthetic Computed Tomography Hounsfield Unit Accuracy
title_full_unstemmed Toward Closing the Loop in Image-to-Image Conversion in Radiotherapy: A Quality Control Tool to Predict Synthetic Computed Tomography Hounsfield Unit Accuracy
title_short Toward Closing the Loop in Image-to-Image Conversion in Radiotherapy: A Quality Control Tool to Predict Synthetic Computed Tomography Hounsfield Unit Accuracy
title_sort toward closing the loop in image to image conversion in radiotherapy a quality control tool to predict synthetic computed tomography hounsfield unit accuracy
topic synthetic CT
conversion prediction
MR-only adaptive radiotherapy
deep learning
url https://www.mdpi.com/2313-433X/10/12/316
work_keys_str_mv AT paolozaffino towardclosingtheloopinimagetoimageconversioninradiotherapyaqualitycontroltooltopredictsyntheticcomputedtomographyhounsfieldunitaccuracy
AT cirobenitoraggio towardclosingtheloopinimagetoimageconversioninradiotherapyaqualitycontroltooltopredictsyntheticcomputedtomographyhounsfieldunitaccuracy
AT adrianthummerer towardclosingtheloopinimagetoimageconversioninradiotherapyaqualitycontroltooltopredictsyntheticcomputedtomographyhounsfieldunitaccuracy
AT gabrielguterresmarmitt towardclosingtheloopinimagetoimageconversioninradiotherapyaqualitycontroltooltopredictsyntheticcomputedtomographyhounsfieldunitaccuracy
AT johannesalbertuslangendijk towardclosingtheloopinimagetoimageconversioninradiotherapyaqualitycontroltooltopredictsyntheticcomputedtomographyhounsfieldunitaccuracy
AT annaprocopio towardclosingtheloopinimagetoimageconversioninradiotherapyaqualitycontroltooltopredictsyntheticcomputedtomographyhounsfieldunitaccuracy
AT carlocosentino towardclosingtheloopinimagetoimageconversioninradiotherapyaqualitycontroltooltopredictsyntheticcomputedtomographyhounsfieldunitaccuracy
AT joaoseco towardclosingtheloopinimagetoimageconversioninradiotherapyaqualitycontroltooltopredictsyntheticcomputedtomographyhounsfieldunitaccuracy
AT antjechristinknopf towardclosingtheloopinimagetoimageconversioninradiotherapyaqualitycontroltooltopredictsyntheticcomputedtomographyhounsfieldunitaccuracy
AT stefanboth towardclosingtheloopinimagetoimageconversioninradiotherapyaqualitycontroltooltopredictsyntheticcomputedtomographyhounsfieldunitaccuracy
AT mariafrancescaspadea towardclosingtheloopinimagetoimageconversioninradiotherapyaqualitycontroltooltopredictsyntheticcomputedtomographyhounsfieldunitaccuracy