Fully Automated Region-Specific Human-Perceptive-Equivalent Image Quality Assessment: Application to 18F-FDG PET Scans
Mehdi Amini,Yazdan Salimi,Ghasem Hajianfar,Ismini Mainta,Elsa Hervier,Amirhossein Sanaat,Arman Rahmim,Isaac Shiri,Habib Zaidi
DOI: https://doi.org/10.1097/rlu.0000000000005526
IF: 10.6
2024-11-02
Clinical Nuclear Medicine
Abstract:Quantitative metabolic imaging using 18 F-FDG PET is a cornerstone of cancer care and has become an important tool in disease management, particularly for oncology applications involving diagnosis, staging, prognosis, and monitoring of treatment response. 1,2 A number of factors in the pipeline of the PET imaging process can influence the quality of the produced images. 3,4 These factors can emerge from the scanner hardware (eg, the miscalibration of the equipment), the radiopharmaceutical injection process (eg, IV injection, 18 F-FDG injected dose), patient characteristics (eg, patients' weight, size, and motion during imaging), acquisition procedure (eg, patient and bed positioning, scan duration per bed position), as well as image reconstruction and attenuation and scatter correction (ASC) algorithms. 3–5 In addition, the appearance of different artifacts that emerge from the PET image or propagated from the concomitant anatomical CT image used for ASC can degrade the image quality and reduce the diagnostic confidence of the interpreters. 6 Due to the quantitative nature of PET images, small variations in the images can bias the SUV, thus leading to misinterpretation by physicians, potentially altering the diagnosis. 7 Hence, prior to clinical use of PET images, high image quality is essential for a precise qualitative interpretation and accurate quantitative analysis.
radiology, nuclear medicine & medical imaging