Search

Background: [18F]FDG PET denoising by SubtlePET™ using deep learning artificial intelligence (AI) was previously found to induce slight modifications in lesion and reference organs' quantification and in lesion detection. As a next step, we aimed to evaluate its clinical impact on [18F]FDG PET solid tumour treatment response assessments, while comparing "standard PET" to "AI denoised half-duration PET" ("AI PET") during follow-up. Results: 110 patients referred for baseline and follow-up standard digital [18F]FDG PET/CT were prospectively included. "Standard" EORTC and, if applicable, PERCIST response classifications by 2 readers between baseline standard PET1 and follow-up standard PET2 as a "gold standard" were compared to "mixed" classifications between standard PET1 and AI PET2 (group 1; n = 64), or between AI PET1 and standard PET2 (group 2; n = 46). Separate classifications were established using either standardized uptake values from ultra-high definition PET with or without AI denoising (simplified to "UHD") or EANM research limited v2 (EARL2)-compliant values (by Gaussian filtering in standard PET and using the same filter in AI PET). Overall, pooling both study groups, in 11/110 (10%) patients at least one EORTCUHD or EARL2 or PERCISTUHD or EARL2 mixed vs. standard classification was discordant, with 369/397 (93%) concordant classifications, unweighted Cohen's kappa = 0.86 (95% CI: 0.78–0.94). These modified mixed vs. standard classifications could have impacted management in 2% of patients. Conclusions: Although comparing similar PET images is preferable for therapy response assessment, the comparison between a standard [18F]FDG PET and an AI denoised half-duration PET is feasible and seems clinically satisfactory. [ABSTRACT FROM AUTHOR]

Given the constant pressure to increase patient throughput while respecting radiation protection, global body PET image quality (IQ) is not satisfactory in all patients. We first studied the association between IQ and other variables, in particular body habitus, on a digital PET/CT. Second, to improve and homogenize IQ, we evaluated a deep learning PET denoising solution (Subtle PETTM) using convolutional neural networks. We analysed retrospectively in 113 patients visual IQ (by a 5-point Likert score in two readers) and semi-quantitative IQ (by the coefficient of variation in the liver, CVliv) as well as lesion detection and quantification in native and denoised PET. In native PET, visual and semi-quantitative IQ were lower in patients with larger body habitus (p < 0.0001 for both) and in men vs. women (p ≤ 0.03 for CVliv). After PET denoising, visual IQ scores increased and became more homogeneous between patients (4.8 ± 0.3 in denoised vs. 3.6 ± 0.6 in native PET; p < 0.0001). CVliv were lower in denoised PET than in native PET, 6.9 ± 0.9% vs. 12.2 ± 1.6%; p < 0.0001. The slope calculated by linear regression of CVliv according to weight was significantly lower in denoised than in native PET (p = 0.0002), demonstrating more uniform CVliv. Lesion concordance rate between both PET series was 369/371 (99.5%), with two lesions exclusively detected in native PET. SUVmax and SUVpeak of up to the five most intense native PET lesions per patient were lower in denoised PET (p < 0.001), with an average relative bias of −7.7% and −2.8%, respectively. DL-based PET denoising by Subtle PETTM allowed [18F]FDG PET global image quality to be improved and homogenized, while maintaining satisfactory lesion detection and quantification. DL-based denoising may render body habitus adaptive PET protocols unnecessary, and pave the way for the improvement and homogenization of PET modalities. [ABSTRACT FROM AUTHOR]

This study aimed to investigate if combining clinical characteristics with pre-therapeutic 18 F-fluorodeoxyglucose (18 F-FDG) positron emission tomography (PET) radiomics could predict the presence of molecular alteration(s) in key molecular targets in lung adenocarcinoma. This non-interventional monocentric study included patients with newly diagnosed lung adenocarcinoma referred for baseline PET who had tumour molecular analyses. The data were randomly split into training and test datasets. LASSO regression with 100-fold cross-validation was performed, including sex, age, smoking history, AJCC cancer stage and 31 PET variables. In total, 109 patients were analysed, and it was found that 63 (57.8%) patients had at least one molecular alteration. Using the training dataset (n = 87), the model included 10 variables, namely age, sex, smoking history, AJCC stage, excessKustosis_HISTO, sphericity_SHAPE, variance_GLCM, correlation_GLCM, LZE_GLZLM, and GLNU_GLZLM. The ROC analysis for molecular alteration prediction using this model found an AUC equal to 0.866 (p < 0.0001). A cut-off value set to 0.48 led to a sensitivity of 90.6% and a positive likelihood ratio (LR+) value equal to 2.4. After application of this cut-off value in the unseen test dataset of patients (n = 22), the test presented a sensitivity equal to 90.0% and an LR+ value of 1.35. A clinico-metabolic 18 F-FDG PET phenotype allows the detection of key molecular target alterations with high sensitivity and negative predictive value. Hence, it opens the way to the selection of patients for molecular analysis. [ABSTRACT FROM AUTHOR]

Purpose: We investigated whether artificial intelligence (AI)-based denoising halves PET acquisition time in digital PET/CT. Methods: One hundred ninety-five patients referred for [18F]FDG PET/CT were prospectively included. Body PET acquisitions were performed in list mode. Original "PET90" (90 s/bed position) was compared to reconstructed ½-duration PET (45 s/bed position) with and without AI-denoising, "PET45AI and PET45". Denoising was performed by SubtlePET™ using deep convolutional neural networks. Visual global image quality (IQ) 3-point scores and lesion detectability were evaluated. Lesion maximal and peak standardized uptake values using lean body mass (SULmax and SULpeak), metabolic volumes (MV), and liver SULmean were measured, including both standard and EARL1 (European Association of Nuclear Medicine Research Ltd) compliant SUL. Lesion-to-liver SUL ratios (LLR) and liver coefficients of variation (CVliv) were calculated. Results: PET45 showed mediocre IQ (scored poor in 8% and moderate in 68%) and lesion concordance rate with PET90 (88.7%). In PET45AI, IQ scores were similar to PET90 (P = 0.80), good in 92% and moderate in 8% for both. The lesion concordance rate between PET90 and PET45AI was 836/856 (97.7%), with 7 lesions (0.8%) only detected in PET90 and 13 (1.5%) exclusively in PET45AI. Lesion EARL1 SULpeak was not significantly different between both PET (P = 0.09). Lesion standard SULpeak, standard and EARL1 SULmax, LLR and CVliv were lower in PET45AI than in PET90 (P < 0.0001), while lesion MV and liver SULmean were higher (P < 0.0001). Good to excellent intraclass correlation coefficients (ICC) between PET90 and PET45AI were observed for lesion SUL and MV (ICC ≥ 0.97) and for liver SULmean (ICC ≥ 0.87). Conclusion: AI allows [18F]FDG PET duration in digital PET/CT to be halved, while restoring degraded ½-duration PET image quality. Future multicentric studies, including other PET radiopharmaceuticals, are warranted. [ABSTRACT FROM AUTHOR]

In DLBCL, the Deauville scoring system (DS) is the standard for PET/CT response assessment. An alternative system, based on the semi-quantitative change in standardized uptake values, namely ΔSUVmax, has been reported to be more objective than the DS. We aimed to compare ΔSUVmax and DS for risk stratification of DLBCL patients on end-of-treatment (EoT) PET. 108 consecutive patients were included. 2-year EFS Kaplan–Meier survival analyses and Cox regression models were performed. 2-year EFS was significantly different between favorable ΔSUVmax (favΔ < −86.5%) and unfavorable ΔSUVmax (unfavΔ ≥ −86.5%) patients: 100.0% ± 0.0 versus 58.3% ± 14.2 (p = 0.001). On Cox multivariable regression, ΔSUVmax status was the only independent predictor of 2-year EFS, outperforming DS. Therefore, ΔSUVmax should be computed for non-responder patients, especially DS4, as the 2-year EFS is not different between responders and non-responders in the case of favΔ. Further studies are needed in order to confirm this hypothesis. [ABSTRACT FROM AUTHOR]

Aim: Our aim was (1) to evaluate the prevalence of steatosis in lymphoma patients and its evolution during treatment; (2) to evaluate the impact of hepatic steatosis on 18F-FDG liver uptake; and (3) to study how hepatic steatosis affects the Deauville score (DS) for discriminating between responders and non-responders.Methods: Over a 1-year period, 358 PET scans from 227 patients [122 diffuse large B cell lymphoma (DLBCL), 57 Hodgkin lymphoma (HL) and 48 Follicular lymphoma (FL)] referred for baseline (n = 143), interim (n = 79) and end-of-treatment (EoT, n = 136) PET scans were reviewed. Steatosis was diagnosed on the unenhanced CT part of PET/CT examinations using a cut-off value of 42 Hounsfield units (HU). EARL-compliant SULmax were recorded on the liver and the tumour target lesion. DS were then computed.Results: Prevalence of steatosis at baseline, interim and EoT PET was 15/143 (10.5%), 6/79 (7.6%) and 16/136 (11.8%), respectively (p = 0.62).Ten out of 27 steatotic patients (37.0%) displayed a steatotic liver on all examinations. Six patients (22.2%) had a disappearance of hepatic steatosis during their time-course of treatment. Only one patient developed steatosis during his course of treatment. Liver SULmax values were significantly lower in the steatosis versus non-steatotic groups of patients for interim (1.66 ± 0.36 versus 2.15 ± 0.27) and EoT (1.67 ± 0.29 versus 2.17 ± 0.30) PET. CT density was found to be an independent factor that correlated with liver SULmax, while BMI, blood glucose level and the type of chemotherapy regimen were not. Using a method based on this correlation to correct liver SULmax, all DS4 steatotic patients on interim (n = 1) and EoT (n = 2) PET moved to DS3.Conclusions: Steatosis is actually a theoretical but not practical issue in most patients but should be recognised and corrected in appropriate cases, namely, for those patients scored DS4 with a percentage difference between the target lesion and the liver background lower than 30%. [ABSTRACT FROM AUTHOR]

Purpose: To seek for the minimal duration per bed position with a digital PET system without compromising image quality and lesion detection in patients requiring fast 18F-FDG PET imaging.Materials and Methods: 19 cancer patients experiencing pain or dyspnea and 9 pediatric patients were scanned on a Vereos system. List mode data were reconstructed with decreasing time frame down to 10 s per bed position. Noise was evaluated in the liver, blood pool and muscle, and using target-to-background ratios. Five PET readers recorded image quality, number of clinically relevant foci and of involved anatomical sites in reconstructions ranging from 60 to 10 s per bed position, compared to the standard 90 s reconstruction.Results: The following reconstructions, which harboured a noise not significantly higher than that of the standard reconstruction, were selected for clinical evaluation: 1iterations/10 subsets/20sec (1i10 s20sec), 1i10 s30sec, and 2i10 sPSF60sec. Only the 60 s per bed acquisition displayed similar target-to-background ratios compared to the standard reconstruction, but mean ratios were still higher than 2.0 for the 30 s reconstruction. Inter-rater agreement for the number of involved anatomical sites and detected lesion was good or almost perfect (Kappa: 0.64-0.91) for all acquisitions. In particular, kappa for the 30 s per bed acquisition was 0.78 and 0.91 for lesion and anatomical sites number, respectively. Intra-rater agreement was also excellent for the 30 s reconstruction (kappa = 0.72). Median estimated total PET acquisition time for the 1i10 s30sec, and the standard reconstruction were 4 and 12 min, respectively.Conclusions: Fast imaging is feasible with state-of-the-art PET systems. Acquisitions of 30 s per bed position are feasible with the Vereos system, requiring optimization of reconstruction parameters. [ABSTRACT FROM AUTHOR]