Artificial Intelligence-Driven Single-Shot PET Image Artifact Detection and Disentanglement: Toward Routine Clinical Image Quality Assurance

Article Article scientifique Article Artificial Intelligence-Driven Single-Shot PET Image Artifact Detection and Disentanglement: Toward Routine Clinical Image Quality Assurance eng 2023-10-26 2023-12 author Shiri Lord Isaac 0000-0002-5735-0736 930888 author Salimi Yazdan 0000-0002-1233-9576 From the Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, Geneva. 943941 Département de radiologie et informatique médicale (RADIO) author Hervier Elsa From the Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, Geneva. 771859 author Pezzoni Agathe From the Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, Geneva. 873092 author Sanaat Amirhossein 0000-0001-8437-2060 From the Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, Geneva. 919308 Département de radiologie et informatique médicale (RADIO) author Mostafaei Shayan author Rahmim Arman author Mainta Ismini Charis From the Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, Geneva. 869258 author Zaidi Habib 0000-0001-7559-5297 98447 Département de radiologie et informatique médicale (RADIO) 1035-1046 10.1097/RLU.0000000000004912 37883015 0363-9762 https://journals.lww.com/nuclearmed/fulltext/2023/12000/artificial_intelligence_driven_single_shot_pet.4.aspx Purpose: Medical imaging artifacts compromise image quality and quantitative analysis and might confound interpretation and misguide clinical decision-making. The present work envisions and demonstrates a new paradigm PET image Quality Assurance NETwork (PET-QA-NET) in which various image artifacts are detected and disentangled from images without prior knowledge of a standard of reference or ground truth for routine PET image quality assurance. Methods: The network was trained and evaluated using training/validation/testing data sets consisting of 669/100/100 artifact-free oncological 18F-FDG PET/CT images and subsequently fine-tuned and evaluated on 384 (20% for fine-tuning) scans from 8 different PET centers. The developed DL model was quantitatively assessed using various image quality metrics calculated for 22 volumes of interest defined on each scan. In addition, 200 additional 18F-FDG PET/CT scans (this time with artifacts), generated using both CT-based attenuation and scatter correction (routine PET) and PET-QA-NET, were blindly evaluated by 2 nuclear medicine physicians for the presence of artifacts, diagnostic confidence, image quality, and the number of lesions detected in different body regions. Results: Across the volumes of interest of 100 patients, SUV MAE values of 0.13 ± 0.04, 0.24 ± 0.1, and 0.21 ± 0.06 were reached for SUVmean, SUVmax, and SUVpeak, respectively (no statistically significant difference). Qualitative assessment showed a general trend of improved image quality and diagnostic confidence and reduced image artifacts for PET-QA-NET compared with routine CT-based attenuation and scatter correction. Conclusion: We developed a highly effective and reliable quality assurance tool that can be embedded routinely to detect and correct for 18F-FDG PET image artifacts in clinical setting with notably improved PET image quality and quantitative capabilities. Artifacts Artificial Intelligence Fluorodeoxyglucose F18 Humans Image Processing, Computer-Assisted / methods Positron Emission Tomography Computed Tomography / methods Positron-Emission Tomography / methods Dewey 616.8 Dewey 616.0757 Swiss National Science Foundation Towards patient-specific hybrid whole-body PET parametric imaging 176052 CH Projects Private Foundation of Geneva University Hospitals RC-06-01 62 Département de radiologie et informatique médicale 227 16 Centre interfacultaire de neurosciences 0D9 false 287 Imagerie Médicale (TEP et TEMP) 542 Clinical nuclear medicine 48 12 151d32f3-5381-4e50-ac2e-9444f165ac32 Article (Published version) Public CNM20231.pdf 5925394 application/pdf CC BY-NC-ND-4.0 8eb6e67abcea55aea9b4fc750ad6baea9ec579c6995ddea748054e8bbea55ae7 e6e064fa-7f66-4b4d-8f16-73b2bb4cc42c Appendix Public cnm_2023_10_23_zaidi_cnm-d-23-00759.pdf 650245 application/pdf 5a66cc84337965adda8a8670f6f2e5512559ed94369c2b1b5776d9b91e51c8f2 openAccess Faculté de médecine / Section de médecine clinique / Département de radiologie et informatique médicale Centres et instituts / Centre interfacultaire de neurosciences

Title	Published in	Access level	OA Policy	Year	Views	Downloads

Artificial Intelligence-Driven Single-Shot PET Image Artifact Detection and Disentanglement: Toward Routine Clinical Image Quality Assurance	Clinical nuclear medicine			2023	110	62

Archive ouverte UNIGE

Pezzoni, Agathe