Differential privacy preserved federated transfer learning for multi-institutional 68Ga-PET image artefact detection and disentanglement
European Journal of Nuclear Medicine and Molecular Imaging, vol. 51, pp. 40–53
Abstract
Abstract Purpose Image artefacts continue to pose challenges in clinical molecular imaging, resulting in misdiagnoses, additional radiation doses to patients and financial costs. Mismatch and halo artefacts occur frequently in gallium-68 ( 68 Ga)-labelled compounds whole-body PET/CT imaging. Correcting for these artefacts is not straightforward and requires algorithmic developments, given that conventional techniques have failed to address them adequately. In the current study, we employed differential privacy-preserving federated transfer learning (FTL) to manage clinical data sharing and tackle privacy issues for building centre-specific models that detect and correct artefacts present in PET images. Methods Altogether, 1413 patients with 68 Ga prostate-specific membrane antigen (PSMA)/DOTA-TATE (TOC) PET/CT scans from 3 countries, including 8 different centres, were enrolled in this study. CT-based attenuation and scatter correction (CT-ASC) was used in all centres for quantitative PET reconstruction. Prior to model training, an experienced nuclear medicine physician reviewed all images to ensure the use of high-quality, artefact-free PET images (421 patients’ images). A deep neural network (modified U2Net) was trained on 80% of the artefact-free PET images to utilize centre-based (CeBa), centralized (CeZe) and the proposed differential privacy FTL frameworks. Quantitative analysis was performed in 20% of the clean data (with no artefacts) in each centre. A panel of two nuclear medicine physicians conducted qualitative assessment of image quality, diagnostic confidence and image artefacts in 128 patients with artefacts (256 images for CT-ASC and FTL-ASC). Results The three approaches investigated in this study for 68 Ga-PET imaging (CeBa, CeZe and FTL) resulted in a mean absolute error (MAE) of 0.42 ± 0.21 ( CI 95%: 0.38 to 0.47), 0.32 ± 0.23 ( CI 95%: 0.27 to 0.37) and 0.28 ± 0.15 ( CI 95%: 0.25 to 0.31), respectively. Statistical analysis using the Wilcoxon test revealed significant differences between the three approaches, with FTL outperforming CeBa and CeZe ( p -value < 0.05) in the clean test set. The qualitative assessment demonstrated that FTL-ASC significantly improved image quality and diagnostic confidence and decreased image artefacts, compared to CT-ASC in 68 Ga-PET imaging. In addition, mismatch and halo artefacts were successfully detected and disentangled in the chest, abdomen and pelvic regions in 68 Ga-PET imaging. Conclusion The proposed approach benefits from using large datasets from multiple centres while preserving patient privacy. Qualitative assessment by nuclear medicine physicians showed that the proposed model correctly addressed two main challenging artefacts in 68 Ga-PET imaging. This technique could be integrated in the clinic for 68 Ga-PET imaging artefact detection and disentanglement using multicentric heterogeneous datasets.
Authors 24
-
University of Bern · University Hospital of Bern · University Hospital of Geneva
Affiliation as printed
Department of Cardiology, Inselspital, University of Bern, Bern, Switzerland
Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, CH-1211, Geneva, Switzerland
-
Affiliation as printed
Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, CH-1211, Geneva, Switzerland
-
Iran University of Medical Sciences · Shaheed Rajaei Cardiovascular Medical and Research Center
Affiliation as printed
Rajaie Cardiovascular Medical and Research Center, Iran University of Medical Sciences, Tehran, Iran
-
Shariati Hospital · Tehran University of Medical Sciences
Affiliation as printed
Research Center for Nuclear Medicine, Shariati Hospital, Tehran University of Medical Sciences, Tehran, Iran
-
Affiliation as printed
BC Cancer Research Institute, Vancouver, BC, Canada
-
Affiliation as printed
Department of Computer Science, University of Geneva, Geneva, Switzerland
-
Affiliation as printed
Department of Medical Epidemiology and Biostatistics, Karolinska Institute, Stockholm, Sweden
Division of Clinical Geriatrics, Department of Neurobiology, Care Sciences and Society, Karolinska Institutet, Stockholm, Sweden
-
Affiliation as printed
Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, CH-1211, Geneva, Switzerland
-
Affiliation as printed
Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, CH-1211, Geneva, Switzerland
-
Bushehr University of Medical Sciences
Affiliation as printed
The Persian Gulf Nuclear Medicine Research Center, Department of Nuclear Medicine, Molecular Imaging, and Theranostics, Bushehr Medical University Hospital, School of Medicine, Bushehr University of Medical Sciences, Bushehr, Iran
-
Affiliation as printed
Department of Medical Radiation Engineering, Shahid Beheshti University, Tehran, Iran
-
Islamic Azad University, Tehran · Islamic Azad University, Science and Research Branch
Affiliation as printed
Department of Medical Radiation Engineering, Science and Research Branch, Islamic Azad University, Tehran, Iran
-
Imam Khomeini Hospital · Tehran University of Medical Sciences
Affiliation as printed
Department of Nuclear Medicine, Imam Khomeini Hospital Complex, Tehran University of Medical Sciences, Tehran, Iran
-
Shariati Hospital · Tehran University of Medical Sciences
Affiliation as printed
Research Center for Nuclear Medicine, Shariati Hospital, Tehran University of Medical Sciences, Tehran, Iran
-
Imam Reza International University
Affiliation as printed
Cancer Research Center, Razavi Hospital, Imam Reza International University, Mashhad, Iran
-
Iran University of Medical Sciences · Shaheed Rajaei Cardiovascular Medical and Research Center
Affiliation as printed
Echocardiography Research Center, Rajaie Cardiovascular Medical and Research Center, Iran University of Medical Sciences, Tehran, Iran
Rajaie Cardiovascular Medical and Research Center, Iran University of Medical Sciences, Tehran, Iran
-
Bushehr University of Medical Sciences
Affiliation as printed
The Persian Gulf Nuclear Medicine Research Center, Department of Nuclear Medicine, Molecular Imaging, and Theranostics, Bushehr Medical University Hospital, School of Medicine, Bushehr University of Medical Sciences, Bushehr, Iran
-
BC Cancer Agency · University of British Columbia
Affiliation as printed
BC Cancer Research Institute, Vancouver, BC, Canada
Department of Radiology, University of British Columbia, Vancouver, BC, Canada
-
Pennsylvania State University · Universitätsklinikum Aachen · RWTH Aachen University
Affiliation as printed
Department of Public Health Sciences, College of Medicine, The Pennsylvania State University, Hershey, PA, 17033, USA
Institute of Pathology, RWTH Aachen University Hospital, Aachen, Germany
-
Affiliation as printed
Department of Computer Science, University of Geneva, Geneva, Switzerland
-
Affiliation as printed
Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, CH-1211, Geneva, Switzerland
-
BC Cancer Agency · University of British Columbia
Affiliation as printed
Department of Integrative Oncology, BC Cancer Research Institute, Vancouver, BC, Canada
Department of Radiology, University of British Columbia, Vancouver, BC, Canada
Molecular Imaging and Therapy, BC Cancer, Vancouver, BC, Canada
-
BC Cancer Agency · University of British Columbia
Affiliation as printed
Department of Integrative Oncology, BC Cancer Research Institute, Vancouver, BC, Canada
Department of Physics and Astronomy, University of British Columbia, Vancouver, Canada
Department of Radiology, University of British Columbia, Vancouver, BC, Canada
-
Habib Zaidi corresponding
University of Geneva · University Medical Center Groningen · University of Groningen · University of Southern Denmark · University Hospital of Geneva
Affiliation as printed
Department of Nuclear Medicine and Molecular Imaging, University of Groningen, University Medical Center Groningen, Groningen, Netherlands. habib.zaidi@hcuge.ch
Department of Nuclear Medicine, University of Southern Denmark, Odense, Denmark. habib.zaidi@hcuge.ch
Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, CH-1211, Geneva, Switzerland. habib.zaidi@hcuge.ch
Geneva University Neuro Center, Geneva University, Geneva, Switzerland. habib.zaidi@hcuge.ch
Department of Nuclear Medicine and Molecular Imaging, University of Groningen, University Medical Center Groningen, Groningen, Netherlands
Department of Nuclear Medicine, University of Southern Denmark, Odense, Denmark
Geneva University Neuro Center, Geneva University, Geneva, Switzerland
Cited by 33 stored of 33
No patents citing this paper on Lens.org (checked 2026-10-06).