Multicenter PET image harmonization using generative adversarial networks
Authors:
Open Access
- 2 May 2024
- journal article
- research article
- Published by Springer Nature in European Journal of Nuclear Medicine and Molecular Imaging
- Vol. 51 (9) , 2532-2546
Abstract
Purpose: To improve reproducibility and predictive performance of PET radiomic features in multicentric studies by cycle-consistent generative adversarial network (GAN) harmonization approaches. Methods: GAN-harmonization was developed to harmonize whole-body PET scans to perform image style and texture translation between different centers and scanners. GAN-harmonization was evaluated by application to two retrospectively collected open datasets and different tasks. First, GAN-harmonization was performed on a dual-center lung cancer cohort (127 female, 138 male) where the reproducibility of radiomic features in healthy liver tissue was evaluated. Second, GAN-harmonization was applied to a head and neck cancer cohort (43 female, 154 male) acquired from three centers. Here, the clinical impact of GAN-harmonization was analyzed by predicting the development of distant metastases using a logistic regression model incorporating first-order statistics and texture features from baseline 18F-FDG PET before and after harmonization. Results: Image quality remained high (structural similarity: left kidney $$\ge$$ 0.800, right kidney $$\ge$$ 0.806, liver $$\ge$$ 0.780, lung $$\ge$$ 0.838, spleen $$\ge$$ 0.793, whole-body $$\ge$$ 0.832) after image harmonization across all utilized datasets. Using GAN-harmonization, inter-site reproducibility of radiomic features in healthy liver tissue increased at least by $$\ge$$ 5 ± 14% (first-order), $$\ge$$ 16 ± 7% (GLCM), $$\ge$$ 19 ± 5% (GLRLM), $$\ge$$ 16 ± 8% (GLSZM), $$\ge$$ 17 ± 6% (GLDM), and $$\ge$$ 23 ± 14% (NGTDM). In the head and neck cancer cohort, the outcome prediction improved from AUC 0.68 (95% CI 0.66–0.71) to AUC 0.73 (0.71–0.75) by application of GAN-harmonization. Conclusions: GANs are capable of performing image harmonization and increase reproducibility and predictive performance of radiomic features derived from different centers and scanners.
Keywords
Funding Information
- ERACoSysMed (4724-B HOLY 2020)
- Medical University of Vienna
This publication has 44 references indexed in Scilit:
- Contrastive image adaptation for acquisition shift reduction in medical imagingArtificial Intelligence in Medicine, 2023
- A review of harmonization strategies for quantitative PETAnnals of Nuclear Medicine, 2023
- Radiogenomic markers enable risk stratification and inference of mutational pathway states in head and neck cancerEuropean Journal of Nuclear Medicine and Molecular Imaging, 2022
- Fully automatic segmentation of diffuse large B cell lymphoma lesions on 3D FDG-PET/CT for total metabolic tumour volume prediction using a convolutional neural network.European Journal of Nuclear Medicine and Molecular Imaging, 2020
- Novel prognostic clinical factors and biomarkers for outcome prediction in head and neck cancer: a systematic reviewThe Lancet Oncology, 2019
- Harmonization of cortical thickness measurements across scanners and sitesNeuroImage, 2018
- Harmonization of multi-site diffusion tensor imaging dataNeuroImage, 2017
- FDG PET/CT: EANM procedure guidelines for tumour imaging: version 2.0European Journal of Nuclear Medicine and Molecular Imaging, 2014
- The Cancer Imaging Archive (TCIA): Maintaining and Operating a Public Information RepositoryJournal of Digital Imaging, 2013
- PET-CT for response assessment and treatment adaptation in head and neck cancerThe Lancet Oncology, 2010