Influence of Statistical Fluctuation on Reproducibility and Accuracy of SUV_max and SUV_peak: A Phantom Study

Abstract

Standardized uptake values (SUVs) have been widely used in the diagnosis of malignant tumors and in clinical trials of tumor therapies as semiquantitative metrics of tumor 18 F-FDG uptake. However, SUVs for small lesions are liable to errors due to partialvolume effect and statistical noise. The purpose of this study was to evaluate the reproducibility and accuracy of maximum and peak SUV (SUV max and SUV peak , respectively) of small lesions in phantom experiments. Methods: We used a body phantom with 6 spheres in a quarter warm background. The PET data were acquired for 1,800 s in list-mode, from which data were extracted to generate 15 PET images for each of the 60-, 90-, 120-, 150-, and 180-s scanning times. The SUV max and SUV peak of the hot spheres in the 1,800-s scan were used as a reference (SUV ref,max and SUV ref,peak ). Coefficients of variation for both SUV max and SUV peak in hot spheres (CV max and CV peak ) were calculated to evaluate the variability of the SUVs. On the other hand, percentage differences between SUV max and SUV ref,max and between SUV peak and SUV ref,peak were calculated for evaluation of the accuracy of SUV. We additionally examined the coefficients of variation of background activity and the percentage background variability as parameters for the physical assessment of image quality. Results: Visibility of a 10-mm-diameter hot sphere was considerably different among scan frames. The CV max and CV peak increased as the sphere size became smaller and as the acquisition time became shorter. SUV max was generally overestimated as the scan time shortened and the sphere size increased. The SUV max and SUV peak of a 37-mm-diameter sphere for 60-s scans had average positive biases of 28.3% and 4.4%, compared with the reference. Conclusion: SUV max was variable and overestimated as the scan time decreased and the sphere size increased. In contrast, SUV peak was a more robust and accurate metric than SUV max . The measurements of SUV peak (or SUV peak normalized to lean body mass) in addition to SUV max are desirable for reproducible and accurate quantification in clinical situations.

A better prediction of progression‐free survival in diffuse large B‐cell lymphoma by a prognostic model consisting of baseline TLG and %ΔSUVmax

“…In particular, for low-uptake lesions, their uptake values were often overestimated as a result of background noise. 21 In contrast, SUV mean made up for the shortcomings of SUV max . Consequently, the volume parameter TLG, derived from SUV mean and TMTV, may perform better in predicting the metabolic activity of the total lesion.…”

“…However, the SUV max representing only one‐pixel point of the lesion could not reflect the condition of the whole lesion. In particular, for low‐uptake lesions, their uptake values were often overestimated as a result of background noise . In contrast, SUV mean made up for the shortcomings of SUV max .…”

Optimization of injected 68Ga-PSMA activity based on list-mode phantom data and clinical validation

“…It seemed unexpected that the RC max values of this scan are lower than the RC max of the 20 min/bed scan. Upward bias is expected with low count statistics, because of a higher chance that variation will lead to a higher maximum value [19][20][21]. This effect increases with larger regions.…”

Protocols for Harmonized Quantification and Noise Reduction in Low-Dose Oncologic 18F-FDG PET/CT Imaging

“…During the last few years, several facilities have adopted the European Association of Nuclear Medicine (EANM) guidelines through the EANM Research Ltd. (EARL) 18 F-FDG PET/CT accreditation program (12) as a standard for PET SUV quantification (12)(13)(14). Despite the wide literature on optimization of PET protocols for lesion detectability (6)(7)(8)(9)15) or the impact of noise on quantitation (16)(17)(18)(19), there is still limited work optimizing low-dose PET as a quantitative biomarker using guidelines such as the EANM criteria (20,21). Advances in PET image reconstruction algorithms that incorporate PSF modeling have reduced noise and increased resolution, potentially allowing for reduced injected activity and thus lower radiation exposure to patients (15,(22)(23)(24)(25), opening new scenarios for cancer screening with PET (15,26,27).…”