Intra-reader reliability of FDG PET volumetric tumor parameters: effects of primary tumor size and segmentation methods

Machine learning predictive performance evaluation of conventional and fuzzy radiomics in clinical cancer imaging cohorts

BACKGROUND

Hybrid imaging became an instrumental part of medical imaging, particularly cancer imaging processes in clinical routine. To date, several radiomic and machine learning studies investigated the feasibility of in vivo tumor characterization with variable outcomes. This study aims to investigate the effect of recently proposed fuzzy radiomics and compare its predictive performance to conventional radiomics in cancer imaging cohorts. In addition, lesion vs. lesion+surrounding fuzzy and conventional radiomic analysis was conducted.

METHODS

Previously published 11C Methionine (MET) positron emission tomography (PET) glioma, 18F-FDG PET/computed tomography (CT) lung, and 68GA-PSMA-11 PET/magneto-resonance imaging (MRI) prostate cancer retrospective cohorts were included in the analysis to predict their respective clinical endpoints. Four delineation methods including manually defined reference binary (Ref-B), its smoothed, fuzzified version (Ref-F), as well as extended binary (Ext-B) and its fuzzified version (Ext-F) were incorporated to extract imaging biomarker standardization initiative (IBSI)-conform radiomic features from each cohort. Machine learning for the four delineation approaches was performed utilizing a Monte Carlo cross-validation scheme to estimate the predictive performance of the four delineation methods.

RESULTS

Reference fuzzy (Ref-F) delineation outperformed its binary delineation (Ref-B) counterpart in all cohorts within a volume range of 938-354987 mm³ with relative cross-validation area under the receiver operator characteristics curve (AUC) of  +4.7-10.4. Compared to Ref-B, the highest AUC performance difference was observed by the Ref-F delineation in the glioma cohort (Ref-F: 0.74 vs. Ref-B: 0.70) and in the prostate cohort by Ref-F and Ext-F (Ref-F: 0.84, Ext-F: 0.86 vs. Ref-B: 0.80). In addition, fuzzy radiomics decreased feature redundancy by approx. 20%.

CONCLUSIONS

Fuzzy radiomics has the potential to increase predictive performance particularly in small lesion sizes compared to conventional binary radiomics in PET. We hypothesize that this effect is due to the ability of fuzzy radiomics to model partial volume effects and delineation uncertainties at small lesion boundaries. In addition, we consider that the lower redundancy of fuzzy radiomic features supports the identification of imaging biomarkers in future studies. Future studies shall consider systematically analyzing lesions and their surroundings with fuzzy and binary radiomics.

Application of 18 F-fluorodeoxyglucose PET/CT radiomic features and machine learning to predict early recurrence of non-small cell lung cancer after curative-intent therapy

OBJECTIVE

To predict the recurrence of non-small cell lung cancer (NSCLC) within 2 years after curative-intent treatment using a machine-learning approach with PET/CT-based radiomics.

PATIENTS AND METHODS

A total of 77 NSCLC patients who underwent pretreatment 18 F-fluorodeoxyglucose PET/CT were retrospectively analyzed. Five clinical features (age, sex, tumor stage, tumor histology, and smoking status) and 48 radiomic features extracted from primary tumors on PET were used for binary classifications. These were ranked, and a subset of useful features was selected based on Gini coefficient scores in terms of associations with relapsed status. Areas under the receiver operating characteristics curves (AUC) were yielded by six machine-learning algorithms (support vector machine, random forest, neural network, naive Bayes, logistic regression, and gradient boosting). Model performances were compared and validated via random sampling.

RESULTS

A PET/CT-based radiomic model was developed and validated for predicting the recurrence of NSCLC during the first 2 years after curation. The most important features were SD and variance of standardized uptake value, followed by low-intensity short-zone emphasis and high-intensity zone emphasis. The naive Bayes model with the 15 best-ranked features displayed the best performance (AUC: 0.816). Prediction models using the five best PET-derived features outperformed those using five clinical variables.

CONCLUSION

The machine learning model using PET-derived radiomic features showed good performance for predicting the recurrence of NSCLC during the first 2 years after a curative intent therapy. PET/CT-based radiomic features may help clinicians improve the risk stratification of relapsed NSCLC.

18F-FDG PET/CT Staging of Head and Neck Cancer: Interobserver Agreement and Accuracy-Results from Multicenter ACRIN 6685 Clinical Trial

To our knowledge, no prior multicenter clinical trial has reported interobserver agreement of 18F-FDG PET/CT scans for staging of clinical N0 neck in head and neck cancer. Methods: A total of 287 participants were recruited. For visual analysis, positive nodal uptake of 18F-FDG was defined as uptake visually greater than activity seen in the blood pool. Results: The negative predictive value of the 18F-FDG PET/CT for N0 clinical neck was 86% or above for visual assessment (95% CI, 86%-88%) for the 2 central readers and above 90% (95% CI, 90%-95%) for SUVmax for central reads and site reads dichotomized at the optimal cutoff value of 1.8 and the prespecified cutoff value of 3.5, respectively. The κ coefficients between the 2 expert readers and between central reads and site reads varied between 0.53 and 0.78. Conclusion: The NPV of the 18F-FDG PET/CT for N0 clinical neck was 86% or above for visual assessment and above 90% for SUVmax cut points of 1.8 and 3.5 with moderate to substantial agreements.

Standardised lesion segmentation for imaging biomarker quantitation: a consensus recommendation from ESR and EORTC

BACKGROUND

Lesion/tissue segmentation on digital medical images enables biomarker extraction, image-guided therapy delivery, treatment response measurement, and training/validation for developing artificial intelligence algorithms and workflows. To ensure data reproducibility, criteria for standardised segmentation are critical but currently unavailable.

METHODS

A modified Delphi process initiated by the European Imaging Biomarker Alliance (EIBALL) of the European Society of Radiology (ESR) and the European Organisation for Research and Treatment of Cancer (EORTC) Imaging Group was undertaken. Three multidisciplinary task forces addressed modality and image acquisition, segmentation methodology itself, and standards and logistics. Devised survey questions were fed via a facilitator to expert participants. The 58 respondents to Round 1 were invited to participate in Rounds 2-4. Subsequent rounds were informed by responses of previous rounds.

RESULTS/CONCLUSIONS

Items with ≥ 75% consensus are considered a recommendation. These include system performance certification, thresholds for image signal-to-noise, contrast-to-noise and tumour-to-background ratios, spatial resolution, and artefact levels. Direct, iterative, and machine or deep learning reconstruction methods, use of a mixture of CE marked and verified research tools were agreed and use of specified reference standards and validation processes considered essential. Operator training and refreshment were considered mandatory for clinical trials and clinical research. Items with a 60-74% agreement require reporting (site-specific accreditation for clinical research, minimal pixel number within lesion segmented, use of post-reconstruction algorithms, operator training refreshment for clinical practice). Items with ≤ 60% agreement are outside current recommendations for segmentation (frequency of system performance tests, use of only CE-marked tools, board certification of operators, frequency of operator refresher training). Recommendations by anatomical area are also specified.

Reliability of gradient-based segmentation for measuring metabolic parameters influenced by uptake time on 18F-PSMA-1007 PET/CT for prostate cancer

Purpose

To determine an optimal setting for functional contouring and quantification of prostate cancer lesions with minimal variation by evaluating metabolic parameters on 18F-PSMA-1007 PET/CT measured by threshold-based and gradient-based methods under the influence of varying uptake time.

Methods and materials

Dual time point PET/CT was chosen to mimic varying uptake time in clinical setting. Positive lesions of patients who presented with newly diagnosed disease or biochemical recurrence after total prostatectomy were reviewed retrospectively. Gradient-based and threshold-based tools at 40%, 50% and 60% of lesion SUVmax (MIM 6.9) were used to create contours on PET. Contouring was considered completed if the target lesion, with its hottest voxel, was delineated from background tissues and nearby lesions under criteria specific to their operations. The changes in functional tumour volume (FTV) and metabolic tumour burden (MTB, defined as the product of SUVmean and FTV) were analysed. Lesion uptake patterns (increase/decrease/stable) were determined by the percentage change in tumour SUVmax at ±10% limit.

Results

A total of 275 lesions (135 intra-prostatic lesions, 65 lymph nodes, 45 bone lesions and 30 soft tissue lesions in pelvic region) in 68 patients were included. Mean uptake time of early and delayed imaging were 94 and 144 minutes respectively. Threshold-based method using 40% to 60% delineated only 85 (31%), 110 (40%) and 137 (50%) of lesions which all were contoured by gradient-based method. Although the overall percentage change using threshold at 50% was the smallest among other threshold levels in FTV measurement, it was still larger than gradient-based method (median: 50%=-7.6% vs gradient=0%). The overall percentage increase in MTB of gradient-based method (median: 6.3%) was compatible with the increase in tumour SUVmax. Only a small proportion of intra-prostatic lesions (<2%), LN (<4%), bone lesions (0%) and soft tissue lesions (<4%) demonstrated decrease uptake patterns.

Conclusions

With a high completion rate, gradient-based method is reliable for prostate cancer lesion contouring on 18F-PSMA-1007 PET/CT. Under the influence of varying uptake time, it has smaller variation than threshold-based method for measuring volumetric parameters. Therefore, gradient-based method is recommended for tumour delineation and quantification on 18F-PSMA-1007 PET/CT.

The role of 18F-FDG PET/CT in patients with diffuse large B-cell lymphoma after radioimmunotherapy using 131I-rituximab as consolidation therapy

Purpose

To evaluate the prognostic value of pretreatment ¹⁸F-FDG PET/CT after consolidation therapy of ¹³¹I-rituximab in patients with diffuse large B-cell lymphoma (DLBCL) who had acquired complete remission after receiving chemotherapy.

Methods

Patients who were diagnosed with DLBCL via histologic confirmation were retrospectively reviewed. All patients had achieved complete remission after 6 to 8 cycles of R-CHOP (rituximab, cyclophosphamide, vincristine, doxorubicin, and prednisolone) chemotherapy after which they underwent consolidation treatment with ¹³¹I-rituximab. ¹⁸F-FDG PET/CT scans were performed before R-CHOP for initial staging. The largest diameter of tumor, maximum standardized uptake value (SUVmax), metabolic tumor volume (MTV), and total lesion glycolysis (TLG) were obtained from pretreatment ¹⁸F-FDG PET/CT scans. Receiver-operating characteristic curves analysis was introduced for assessing the optimal criteria. Kaplan-Meier curve survival analysis was performed to evaluate both relapse free survival (RFS) and overall survival (OS).

Results

A total of 15 patients (12 males and 3 females) with a mean age of 56 (range, 30–73) years were enrolled. The median follow-up period of these patients was 73 months (range, 11–108 months). Four (27%) patients relapsed. Of them, three died during follow-up. Median values of the largest tumor size, highest SUVmax, MTV, and TLG were 5.3 cm (range, 2.0–16.4 cm), 20.2 (range, 11.1–67.4), 231.51 (range, 15–38.34), and 1277.95 (range, 238.37–10341.04), respectively. Patients with SUVmax less than or equal to 16.9 showed significantly worse RFS than patients with SUVmax greater than 16.9 (5-year RFS rate: 60% vs. 100%, p = 0.008). Patients with SUVmax less than or equal to 16.9 showed significantly worse OS than patients with SUVmax greater than 16.9 (5-year OS rate: 80% vs. 100% p = 0.042).

Conclusion

Higher SUVmax at pretreatment ¹⁸F-FDG PET/CT was associated with better relapse free survival and overall survival in DLBCL patients after consolidation therapy with ¹³¹I-rituximab. However, because this study has a small number of patients, a phase 3 study with a larger number of patients is needed for clinical application in the future.

Diffusion-weighted MRI and PET/CT reproducibility in epithelial ovarian cancers during neoadjuvant chemotherapy

PURPOSE

To investigate the reproducibility of diffusion-weighted (DW) MRI and ¹⁸F-Fluorodeoxyglucose (¹⁸F-FDG)-Positron emission tomography/CT (PET/CT) in monitoring response to neoadjuvant chemotherapy in epithelial ovarian cancer.

MATERIALS AND METHODS

Ten women (median age, 67 years; range: 41.8-77.3 years) with stage IIIC-IV epithelial ovarian cancers were included in this prospective trial (NCT02792959) between 2014 and 2016. All underwent initial laparoscopic staging, four cycles of carboplatine-paclitaxel-based chemotherapy and interval debulking surgery. PET/CT and DW-MRI were performed at baseline (C0), after one cycle (C1) and before surgery (C4). Two nuclear physicians and two radiologists assessed five anatomic sites for the presence of ≥1 lesion. Target lesions in each site were defined and their apparent diffusion coefficient (ADC), maximal standardized uptake value (SUV-max), SUV-mean, SUL-peak, metabolic tumor volume (MTV) and total lesion glycolysis (TLG) were monitored (i.e., 10 patients ×5 sites ×3 time-points). Their relative early and late changes were calculated. Intra/inter-observer reproducibilities of qualitative and quantitative analysis were estimated with Kappa and intra-class correlation coefficients (ICCs).

RESULTS

For both modalities, inter- and intra-observer agreement percentages were excellent for initial staging but declined later for DW-MRI, leading to lower Kappa values for inter- and intra-observer variability (0.949 and 1 at C0, vs. 0.633 and 0.643 at C4, respectively) while Kappa values remained>0.8 for PET/CT. Inter- and intra-observer ICCs were>0.75 for SUV-max, SUL-peak, SUV-mean and their change regardless the time-point. ADC showed lower ICCs (range: 0.013-0.811). ANOVA found significant influences of the evaluation time, the measurement used (ADC, SUV-max, SUV-mean, SUV-max, SUL-peak, MTV or TLG) and their interaction on ICC values (P=0.0023, P<0.0001 and P =0.0028, respectively).

CONCLUSION

While both modalities demonstrated high reproducibility at baseline, only SUV-max, SUL-peak, SUV-mean and their changes maintained high reproducibility during chemotherapy.

Quantification of Metastatic Prostate Cancer Whole-Body Tumor Burden with 18F-FDG PET Parameters and Associations with Overall Survival After First-Line Abiraterone or Enzalutamide: A Single-Center Retrospective Cohort Study

New biomarkers for metastatic prostate cancer are needed. The aim of this study was to evaluate the prognostic value of ¹⁸F-FDG PET whole-body tumor burden parameters in patients with metastatic prostate cancer who received first-line abiraterone or enzalutamide therapy. Methods: This was a retrospective study of patients with metastatic castration-sensitive prostate cancer (mCSPC, n = 25) and metastatic castration-resistant prostate cancer (mCRPC, n = 71) who underwent ¹⁸F-FDG PET/CT within 90 d before first-line treatment with abiraterone or enzalutamide at a tertiary-care academic cancer center. Whole-body tumor burden on PET/CT was quantified as metabolic tumor volume (MTV) and total lesion glycolysis (TLG) and correlated with overall survival (OS) probabilities using Kaplan-Meier curves and Cox models. Results: The median follow-up in survivors was 56.3 mo (interquartile range, 37.7-66.8 mo); the median OSs for patients with mCRPC and mCSPC were 27.8 and 76.1 mo, respectively (P < 0.001). On univariate analysis, the OS probability of mCRPC patients was significantly associated with plasma levels of alkaline phosphatase (hazard ratio [HR], 1.90; P < 0.001), plasma levels of lactate dehydrogenase (HR, 1.01; P < 0.001), hemoglobin levels (HR, 0.80; P = 0.013), whole-body SUV_max (HR, 1.14; P < 0.001), the number of ¹⁸F-FDG-avid metastases (HR, 1.08; P < 0.001), whole-body metabolic tumor volume (HR, 1.86; P < 0.001), and TLG (HR, 1.84; P < 0.001). On multivariable analysis with stepwise variable selection, hemoglobin levels (HR, 0.81; P = 0.013) and whole-body TLG (HR, 1.88; P < 0.001) were independently associated with OS. In mCSPC patients, no significant association was observed between these variables and OS. Conclusion: In patients with mCRPC receiving first-line treatment with abiraterone or enzalutamide, ¹⁸F-FDG PET WB TLG is independently associated with OS and might be used as a quantitative prognostic imaging biomarker.

A physics-guided modular deep-learning based automated framework for tumor segmentation in PET

An important need exists for reliable positron emission tomography (PET) tumor-segmentation methods for tasks such as PET-based radiation-therapy planning and reliable quantification of volumetric and radiomic features. To address this need, we propose an automated physics-guided deep-learning-based three-module framework to segment PET images on a per-slice basis. The framework is designed to help address the challenges of limited spatial resolution and lack of clinical training data with known ground-truth tumor boundaries in PET. The first module generates PET images containing highly realistic tumors with known ground-truth using a new stochastic and physics-based approach, addressing lack of training data. The second module trains a modified U-net using these images, helping it learn the tumor-segmentation task. The third module fine-tunes this network using a small-sized clinical dataset with radiologist-defined delineations as surrogate ground-truth, helping the framework learn features potentially missed in simulated tumors. The framework was evaluated in the context of segmenting primary tumors in 18F-fluorodeoxyglucose (FDG)-PET images of patients with lung cancer. The framework's accuracy, generalizability to different scanners, sensitivity to partial volume effects (PVEs) and efficacy in reducing the number of training images were quantitatively evaluated using Dice similarity coefficient (DSC) and several other metrics. The framework yielded reliable performance in both simulated (DSC: 0.87 (95% confidence interval (CI): 0.86, 0.88)) and patient images (DSC: 0.73 (95% CI: 0.71, 0.76)), outperformed several widely used semi-automated approaches, accurately segmented relatively small tumors (smallest segmented cross-section was 1.83 cm2), generalized across five PET scanners (DSC: 0.74 (95% CI: 0.71, 0.76)), was relatively unaffected by PVEs, and required low training data (training with data from even 30 patients yielded DSC of 0.70 (95% CI: 0.68, 0.71)). In conclusion, the proposed automated physics-guided deep-learning-based PET-segmentation framework yielded reliable performance in delineating tumors in FDG-PET images of patients with lung cancer.

Imaging phenotype using 18F-fluorodeoxyglucose positron emission tomography-based radiomics and genetic alterations of pancreatic ductal adenocarcinoma

PURPOSE

This study aimed to determine if major gene mutations including in KRAS, SMAD4, TP53, and CDKN2A were related to imaging phenotype using ¹⁸F-fluorodeoxyglucose (FDG) positron emission tomography (PET)-based radiomics in patients with pancreatic ductal adenocarcinoma (PDAC).

METHODS

Data on 48 PDAC patients with pretreatment FDG PET/CT who underwent genomic analysis of their tumor tissue were retrospectively analyzed. A total of 35 unique quantitative radiomic features were extracted from PET images, including imaging phenotypes such as pixel intensity, shape, and textural features. Targeted exome sequencing using a customized cancer panel was used for genomic analysis. To assess the predictive performance of genetic alteration using PET-based radiomics, areas under the receiver operating characteristic curve (AUC) were used.

RESULTS

Mutation frequencies were KRAS 87.5%, TP53 70.8%, SMAD4 25.0%, and CDKN2A 18.8%. KRAS gene mutations were significantly associated with low-intensity textural features, including long-run emphasis (AUC = 0.806), zone emphasis (AUC = 0.794), and large-zone emphasis (AUC = 0.829). SMAD4 gene mutations showed significant relationships with standardized uptake value skewness (AUC = 0.727), long-run emphasis (AUC = 0.692), and high-intensity textural features such as run emphasis (AUC = 0.775), short-run emphasis (AUC = 0.736), zone emphasis (AUC = 0.750), and short-zone emphasis (AUC = 0.725). No significant associations were seen between the imaging phenotypes and genetic alterations in TP53 and CDKN2A.

CONCLUSION

Genetic alterations of KRAS and SMAD4 had significant associations with FDG PET-based radiomic features in PDAC. PET-based radiomics may help clinicians predict genetic alteration status in a noninvasive way.

Pre-treatment 18F-FDG PET-based radiomics predict survival in resected non-small cell lung cancer

AIM

To assess the prognostic value of 2-[¹⁸F]-fluoro-2-deoxy-d-glucose (FDG) positron-emission tomography (PET)-based radiomics using a machine learning approach in patients with non-small cell lung cancer (NSCLC).

MATERIALS AND METHODS

Ninety-three patients with stage I-III NSCLC who underwent combined PET/computed tomography (CT) followed by curative resection. A total of 35 unique quantitative radiomic features was extracted from the PET images, which included imaging phenotypes such as pixel intensity, shape, and texture. Radiomic features were ranked based on score according to their correlation with disease recurrence status within a 3-year follow-up. The recurrence risk classification performances of machine learning algorithms (random forest, neural network, naive Bayes, logistic regression, and support vector machine) using the 20 best-ranked features were compared using the areas under the receiver operating characteristic curve (AUC) and validated by the random sampling method.

RESULTS

Contrast and busyness texture features from neighbourhood grey-level difference matrix were found to be the two best predictors of disease recurrence. The random forest model obtained the best performance (AUC: 0.956, accuracy: 0.901, F1 score: 0.872, precision: 0.905, recall: 0.842), followed by the neural network model (AUC: 0.871, accuracy: 0.780, F1 score: 0.708, precision: 0.755, recall: 0.666).

CONCLUSION

A PET-based radiomic model was developed and validated for risk classification in NSCLC. The machine learning approach with random forest classifier exhibited good performance in predicting the recurrence risk. Radiomic features may help clinicians to improve the risk stratification for clinical practice.

Quantitative PET/CT in clinical practice: assessing the agreement of PET tumor indices using different clinical reading platforms

OBJECTIVE

The aim of this study was to determine whether various fluorine-18-fluorodeoxyglucose PET/CT-derived parameters used in oncology vary significantly depending on the interpretation software systems used in clinical practice for multiple human solid tumors.

PATIENTS AND METHODS

A total of 120 fluorine-18-fluorodeoxyglucose PET/CT studies carried out in patients with pancreatic, lung, colorectal, and head and neck cancers were evaluated retrospectively on two different vendor software platforms including Mirada and MIMVista. Regions of interest were placed on the liver to determine the liver mean standardized uptake value at lean body mass (SUL) and on each tumor to determine the SULmax, SULpeak. Total lesion glycolysis (TLG) and metabolic tumor volume (MTV) were determined using fixed thresholds of 50% of SULmax and SULpeak. Inter-reader, intersystem intraclass correlations, systematic bias, and variability reflected by the 95% limits of agreement, and precision were determined.

RESULTS

There was excellent inter-reader reliability between the readers and the two software systems, with intraclass correlations more than 0.9 for all PET metrics, with P values less than 0.0001. The bias and SD on Bland-Altman analysis between the two software platforms for tumor SULmax, SULpeak, Max50MTV, and Peak50MTV, respectively, for Reader 1 were -1.52±2.24, 0.80±3.67, -0.80±13.01, and -4.49±20.6. For Reader 2, the biases were -1.62±1.95, 0.18±3.60, -0.27±4.64, and -3.13±8.30. The precision between the two systems was better for SULmax and SULpeak, with less variance observed, than for volume-based metrics such as Max50MTV and Peak50MTV or TLG.

CONCLUSION

Excellent correlation has been found between two tested software reading platforms for all PET-derived metrics in a dual-reader analysis. Overall, the SULmax and SULpeak values had less bias and better precision compared with the MTV and TLG.

Practical no-gold-standard evaluation framework for quantitative imaging methods: application to lesion segmentation in positron emission tomography

Recently, a class of no-gold-standard (NGS) techniques have been proposed to evaluate quantitative imaging methods using patient data. These techniques provide figures of merit (FoMs) quantifying the precision of the estimated quantitative value without requiring repeated measurements and without requiring a gold standard. However, applying these techniques to patient data presents several practical difficulties including assessing the underlying assumptions, accounting for patient-sampling-related uncertainty, and assessing the reliability of the estimated FoMs. To address these issues, we propose statistical tests that provide confidence in the underlying assumptions and in the reliability of the estimated FoMs. Furthermore, the NGS technique is integrated within a bootstrap-based methodology to account for patient-sampling-related uncertainty. The developed NGS framework was applied to evaluate four methods for segmenting lesions from F-Fluoro-2-deoxyglucose positron emission tomography images of patients with head-and-neck cancer on the task of precisely measuring the metabolic tumor volume. The NGS technique consistently predicted the same segmentation method as the most precise method. The proposed framework provided confidence in these results, even when gold-standard data were not available. The bootstrap-based methodology indicated improved performance of the NGS technique with larger numbers of patient studies, as was expected, and yielded consistent results as long as data from more than 80 lesions were available for the analysis.

Combined fuzzy logic and random walker algorithm for PET image tumor delineation

PURPOSE

The random walk (RW) technique serves as a powerful tool for PET tumor delineation, which typically involves significant noise and/or blurring. One challenging step is hard decision-making in pixel labeling. Fuzzy logic techniques have achieved increasing application in edge detection. We aimed to combine the advantages of fuzzy edge detection with the RW technique to improve PET tumor delineation.

METHODS

A fuzzy inference system was designed for tumor edge detection from RW probabilities. Three clinical PET/computed tomography datasets containing 12 liver, 13 lung, and 18 abdomen tumors were analyzed, with manual expert tumor contouring as ground truth. The standard RW and proposed combined method were compared quantitatively using the dice similarity coefficient, the Hausdorff distance, and the mean standard uptake value.

RESULTS

The dice similarity coefficient of the proposed method versus standard RW showed significant mean improvements of 21.0±7.2, 12.3±5.8, and 18.4%±6.1% for liver, lung, and abdominal tumors, respectively, whereas the mean improvements in the Hausdorff distance were 3.6±1.4, 1.3±0.4, 1.8±0.8 mm, and the mean improvements in SUVmean error were 15.5±6.3, 11.7±8.6, and 14.1±6.8% (all P's<0.001). For all tumor sizes, the proposed method outperformed the RW algorithm. Furthermore, tumor edge analysis demonstrated further enhancement of the performance of the algorithm, relative to the RW method, with decreasing edge gradients.

CONCLUSION

The proposed technique improves PET lesion delineation at different tumor sites. It depicts greater effectiveness in tumors with smaller size and/or low edge gradients, wherein most PET segmentation algorithms encounter serious challenges. Favorable execution time and accurate performance of the algorithm make it a great tool for clinical applications.

Reproducibility and uptake time dependency of volume-based parameters on FDG-PET for lung cancer

Background

Volume-based parameters, such as metabolic tumor volume (MTV) and total lesion glycolysis (TLG), on F-18 fluorodeoxyglucose (FDG) positron emission tomography (PET) are useful for predicting treatment response in nonsmall cell lung cancer (NSCLC). We aimed to examine intra- and inter-operator reproducibility to measure the MTV and TLG, and to estimate their dependency on the uptake time.

Methods

Fifty NSCLC patients underwent preoperative FDG-PET. After an injection of FDG, the whole body was scanned twice: at the early phase (61.4 ± 2.8 min) and delayed phase (117.7 ± 1.6 min). Two operators independently defined the tumor boundary using three different delineation methods: (1) the absolute SUV threshold method (MTV_p and TLG_p; p = 2.0, 2.5, 3.0, 3.5), (2) the fixed% SUVmax threshold method (MTV_q% and TLG_q%; q = 35, 40, 45), and (3) the adaptive region-growing method (MTV_ARG and TLG_ARG). Parameters were compared between operators and between phases.

Results

Both the intra- and inter-operator reproducibility were high for all parameters using any method (intra-class correlation > 0.99 each). MTV_3.0 and MTV_3.5 resulted in a significant increase from the early to delayed phase (P < 0.05 for both), whereas MTV_2.0 and MTV_2.5 neither increased nor decreased (P = n.s.). All of the MTV_q% values significantly decreased over time (P < 0.01), whereas MTV_ARG and TLG with any delineation method increased significantly (P < 0.05).

Conclusions

High reproducibility of MTV and TLG was obtained by all of the methods used. MTV_2.0 and MTV_2.5 were the least sensitive to uptake time, and may be good alternatives when we compare images acquired with different uptake times, although applying constant uptake time is important for volume measurement.

Evaluation of diagnostic performance of whole-body simultaneous PET/MRI in pediatric lymphoma

BACKGROUND

Whole-body (18)F-fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) is the standard of care for lymphoma. Simultaneous PET/MRI (magnetic resonance imaging) is a promising new modality that combines the metabolic information of PET with superior soft-tissue resolution and functional imaging capabilities of MRI while decreasing radiation dose. There is limited information on the clinical performance of PET/MRI in the pediatric setting.

OBJECTIVE

This study evaluated the feasibility, dosimetry, and qualitative and quantitative diagnostic performance of simultaneous whole-body FDG-PET/MRI in children with lymphoma compared to PET/CT.

MATERIALS AND METHODS

Children with lymphoma undergoing standard of care FDG-PET/CT were prospectively recruited for PET/MRI performed immediately after the PET/CT. Images were evaluated for quality, lesion detection and anatomical localization of FDG uptake. Maximum and mean standardized uptake values (SUVmax/mean) of normal organs and SUVmax of the most FDG-avid lesions were measured for PET/MRI and PET/CT. Estimation of radiation exposure was calculated using specific age-related factors.

RESULTS

Nine PET/MRI scans were performed in eight patients (mean age: 15.3 years). The mean time interval between PET/CT and PET/MRI was 51 ± 10 min. Both the PET/CT and PET/MRI exams had good image quality and alignment with complete (9/9) concordance in response assessment. The SUVs from PET/MRI and PET/CT were highly correlated for normal organs (SUVmean r(2): 0.88, P<0.0001) and very highly for FDG-avid lesions (SUVmax r(2): 0.94, P=0.0002). PET/MRI demonstrated an average percent radiation exposure reduction of 39% ± 13% compared with PET/CT.

CONCLUSION

Simultaneous whole-body PET/MRI is clinically feasible in pediatric lymphoma. PET/MRI performance is comparable to PET/CT for lesion detection and SUV measurements. Replacement of PET/CT with PET/MRI can significantly decrease radiation dose from diagnostic imaging in children.

Intratumoral heterogeneity of (18)F-FDG uptake predicts survival in patients with pancreatic ductal adenocarcinoma

PURPOSE

To assess whether intratumoral heterogeneity measured by (18)F-FDG PET texture analysis has potential as a prognostic imaging biomarker in patients with pancreatic ductal adenocarcinoma (PDAC).

METHODS

We evaluated a cohort of 137 patients with newly diagnosed PDAC who underwent pretreatment (18)F-FDG PET/CT from January 2008 to December 2010. First-order (histogram indices) and higher-order (grey-level run length, difference, size zone matrices) textural features of primary tumours were extracted by PET texture analysis. Conventional PET parameters including metabolic tumour volume (MTV), total lesion glycolysis (TLG), and standardized uptake value (SUV) were also measured. To assess and compare the predictive performance of imaging biomarkers, time-dependent receiver operating characteristic (ROC) curves for censored survival data and areas under the ROC curve (AUC) at 2 years after diagnosis were used. Associations between imaging biomarkers and overall survival were assessed using Cox proportional hazards regression models.

RESULTS

The best imaging biomarker for overall survival prediction was first-order entropy (AUC = 0.720), followed by TLG (AUC = 0.697), MTV (AUC = 0.692), and maximum SUV (AUC = 0.625). After adjusting for age, sex, clinical stage, tumour size and serum CA19-9 level, multivariable Cox analysis demonstrated that higher entropy (hazard ratio, HR, 5.59; P = 0.028) was independently associated with worse survival, whereas TLG (HR 0.98; P = 0.875) was not an independent prognostic factor.

CONCLUSION

Intratumoral heterogeneity of (18)F-FDG uptake measured by PET texture analysis is an independent predictor of survival along with tumour stage and serum CA19-9 level in patients with PDAC. In addition, first-order entropy as a measure of intratumoral metabolic heterogeneity is a better quantitative imaging biomarker of prognosis than conventional PET parameters.

Prognostic Value of FDG PET/CT-Derived Parameters in Pancreatic Adenocarcinoma at Initial PET/CT Staging

OBJECTIVE

The purpose of this study is to evaluate the performance of PET-derived parameters as prognostic markers for overall survival (OS) and progression-free survival (PFS) outcome in patients with pancreatic adenocarcinoma.

MATERIALS AND METHODS

We conducted a retrospective study of 106 patients (62 men and 44 women) with histologically proven pancreatic adenocarcinoma who underwent initial staging FDG PET/CT before treatment. Peak standardized uptake value (SUV), maximum SUV (SUVmax), metabolic tumor volume, and tumor glycolytic activity of the primary pancreatic tumor were measured. Two segmentation methods were performed to obtain the metabolic tumor volume and tumor glycolytic activity for all tumors: a gradient-based segmentation model (metabolic tumor volume and tumor glycolytic activity by gradient edge detection) and a fixed-threshold model with a threshold of 50% of the lesion's SUVmax and peak SUV. Univariate and multivariate Cox regression models were developed including clinical and imaging parameters for OS and PFS.

RESULTS

Multivariate Cox regression analysis showed a statistically significant association between PFS and age, SUVmax, peak SUV, and tumor glycolytic activity by gradient edge detection. There was a statistically significant difference in PFS for patients with values above and below the median cutoff points for SUVmax (hazard ratio [HR], 1.12; p < 0.01), peak SUV (HR, 1.25; p < 0.02), and tumor glycolytic activity measured by gradient edge detection (HR, 1.00; p < 0.02) of the primary tumor. However, multivariate Cox regression analysis showed a statistically significant association only between tumor glycolytic activity by gradient edge detection and OS (p = 0.04), and there was a statistically significant difference in OS between patients with values above and below the median cutoff point for the tumor glycolytic activity by gradient edge detection of the primary tumor (HR, 1.42; p = 0.05).

CONCLUSION

Age, SUVmax, peak SUV, and total lesion glycolysis (i.e., tumor glycolytic activity) of the primary tumor are associated with PFS, and tumor glycolytic activity is associated with OS in patients with pancreatic adenocarcinoma.

Value of FDG PET/CT in Patient Management and Outcome of Skeletal and Soft Tissue Sarcomas

Fluorodeoxyglucose (FDG)-PET/computed tomography (CT) has been increasingly used in bone and soft tissue sarcomas and provides advantages in the initial tumor staging, tumor grading, therapy assessment, and recurrence detection. FDG-PET/CT metabolic parameters are reliable predictors of survival in sarcomas and could be implemented in risk stratification models along with other prognostic factors in these patients.

Quantitative interpretation of FDG PET/CT with myocardial perfusion imaging increases diagnostic information in the evaluation of cardiac sarcoidosis

BACKGROUND

FDG PET/CT with myocardial perfusion imaging is a useful method for evaluating cardiac sarcoidosis (CS), but interpretation is not standardized. We developed a method for quantification of cardiac FDG PET/CT and evaluated its relationship to conventional interpretation, perfusion defects, clinical events, and immunosuppressive treatment.

METHODS AND RESULTS

FDG PET/CT with MPI studies performed for CS (n = 38) were retrospectively compared to negative control studies acquired for oncologic indications (n = 10). Quantitative measures of FDG volume-intensity (Cardiac Metabolic Activity, CMA) was performed using standardized uptake values (SUVs). CMA (477.7 ± 909 vs 0.55 ± 2.1 vs 0.3 ± 0.3 g glucose, P = .02) was significantly greater in visually FDG-positive studies compared to visually negative and oncologic negative studies. Among patients with CS, CMA was greater in studies with an EF < 50% (760.3 ± 1,148 vs 87.4 ± 161 g glucose, P = .03) and preceding an adverse clinical event (1,095 ± 1,253 vs 73 ± 144 g glucose, P = .006). CMA was the only independent predictor of events by multivariate analysis. In patients with repeat examinations (n = 7), CMA decreased with prednisone treatment in 5 of 6 patients.

CONCLUSIONS

Quantification of FDG uptake in CS correlates with lower EFs, clinical events, and immunosuppression treatment.

A review on segmentation of positron emission tomography images

Positron Emission Tomography (PET), a non-invasive functional imaging method at the molecular level, images the distribution of biologically targeted radiotracers with high sensitivity. PET imaging provides detailed quantitative information about many diseases and is often used to evaluate inflammation, infection, and cancer by detecting emitted photons from a radiotracer localized to abnormal cells. In order to differentiate abnormal tissue from surrounding areas in PET images, image segmentation methods play a vital role; therefore, accurate image segmentation is often necessary for proper disease detection, diagnosis, treatment planning, and follow-ups. In this review paper, we present state-of-the-art PET image segmentation methods, as well as the recent advances in image segmentation techniques. In order to make this manuscript self-contained, we also briefly explain the fundamentals of PET imaging, the challenges of diagnostic PET image analysis, and the effects of these challenges on the segmentation results.

Head and neck squamous cell cancer (stages III and IV) induction chemotherapy assessment: value of FDG volumetric imaging parameters

INTRODUCTION

To evaluate whether the change in the metabolic tumour volume (MTV) or total lesion glycolysis (TLG) of the primary tumour, before and after induction chemotherapy, predicts outcome for patients with advanced head and neck squamous cell cancer (SCC).

METHODS

Twenty-eight patients with advanced (American Joint Committee on Cancer stage III and IV) head and neck SCC who underwent positron emission tomography (PET)/CT were included in this retrospective study. Primary tumour MTV and TLG were measured using gradient and fixed percentage threshold segmentations. Outcome endpoint was disease progression or mortality. Pearson correlation, Bland-Altman and receiver operator characteristic analysis were performed.

RESULTS

The Pearson's correlation coefficients between percentage changes (pre- and post-induction chemotherapy) from gradient MTV (MTVG) and the 38% SUVmax threshold MTV (MTV38) was 0.96 and between MTVG and the 50% threshold MTV (MTV50) was 0.95 (P < 0.0001). The corresponding Pearson r between TLGG and TLG38 was 0.94 and between TLGG and TLG50 was 0.96 (P < 0.0001). The least bias was 1.89% (standard deviation = 25.30%) between the percentage changes of MTVG and MTV50. The areas under the curve for predicting progression or mortality were 0.76 (P = 0.03) for MTVG and 0.82 for TLGG (P = 0.009). Optimum cut points of a 42% reduction in MTVG and a 55% reduction in the TLGG predict event-free survival with a sensitivity of 62.5% and a specificity of 90% and a hazards ratio of 6.25.

CONCLUSION

A reduction in primary tumour MTV of at least 42% or in TLG of at least 55% after induction chemotherapy may predict event-free survival in patients with advanced head and neck SCC.

Evolving role of FDG PET/CT in multiple myeloma imaging and management

OBJECTIVE

Multiple myeloma is the most common cause of primary malignancy in bones. Radiography has been the imaging reference standard for decades. However, the growing use of CT, MRI, and PET technology has led to earlier diagnosis of multiple myeloma, more accurate therapy assessment, and better prediction of patient outcome. This article is focused on the evolving role of (18)F-FDG PET/CT in multiple myeloma diagnosis, therapy assessment, and prognosis.

CONCLUSION

FDG PET/CT is a valuable imaging modality in diagnosis, therapy assessment, and prognosis of multiple myeloma.