Dual-Energy CT-Derived Electron Density for Diagnosing Metastatic Mediastinal Lymph Nodes in Non-Small Cell Lung Cancer: Comparison With Conventional CT and FDG PET/CT Findings.

BACKGROUND. Accurate nodal staging is essential to guide treatment selection in patients with non-small cell lung cancer (NSCLC). To our knowledge, measurement of electron density (ED) using dual-energy CT (DECT) is unexplored for this purpose. OBJECTIVE. The purpose of our study was to assess the utility of ED from DECT in diagnosing metastatic mediastinal lymph nodes in patients with NSCLC in comparison with conventional CT and FDG PET/CT. METHODS. This retrospective study included 57 patients (36 men, 21 women; mean age, 68.4 ± 8.9 [SD] years) with NSCLC and surgically resected mediastinal lymph nodes who underwent preoperative DECT and FDG PET/CT. The patients had a total of 117 resected mediastinal lymph nodes (33 metastatic, 84 nonmetastatic). Two radiologists independently reviewed the morphologic features of nodes on the 120-kVp images and also measured the iodine concentration (IC) and ED of nodes using maps generated from DECT data; consensus was reached for discrepancies. Two different radiologists assessed FDG PET/CT examinations in consensus for positive node uptake. Diagnostic performance was evaluated for individual and pairwise combinations of features. RESULTS. The sensitivity, specificity, and accuracy for nodal metastasis were 15.2%, 98.8%, and 75.2% for the presence of necrosis, respectively; 54.5%, 85.7%, and 76.9% for short-axis diameter greater than 8.5 mm; 63.6%, 73.8%, and 70.9% for long-axis diameter greater than 13.0 mm; 51.5%, 79.8%, and 71.8% for attenuation on 120-kVp images of 95.8 HU or less; 87.9%, 58.3%, and 66.7% for ED of 3.48 × 1023/cm3 or less; and 66.7%, 75.0%, and 72.6% for positive FDG uptake. Among pairwise combinations of features, accuracy was highest for the combination of ED and short-axis diameter (accuracy, 82.9%; sensitivity, 54.5%; specificity, 94.0%) and the combination of ED and positive FDG uptake (accuracy, 82.1%; sensitivity, 60.6%; specificity, 90.5%); these accuracies were greater than those for the individual features (p < .05). The remaining combinations exhibited accuracies ranging from 74.4% to 77.8%. Interobserver agreement analysis showed an intraclass correlation coefficient of 0.90 for ED. IC was not significantly different between metastatic and nonmetastatic nodes (p = .18) and was excluded from the diagnostic performance analysis. CONCLUSION. ED derived from DECT may help diagnose metastatic lymph nodes in NSCLC given decreased ED in metastatic nodes. CLINICAL IMPACT. ED may complement conventional CT findings and FDG uptake on PET/CT in diagnosing metastatic nodes.

The clinical value of texture analysis of dual-time-point 18F-FDG-PET/CT imaging to differentiate between 18F-FDG-avid benign and malignant pulmonary lesions.

OBJECTIVES: To examine whether the texture analysis of dual-time-point (DTP) F-18-fluorodeoxyglucose (18F-FDG)-PET/CT imaging can differentiate between 18F-FDG-avid benign and malignant pulmonary lesions. METHODS: We compared standardized uptake value (SUV)-related (SUVmax [g/ml] and SUVmean [g/ml]), volumetric (metabolic tumor volume [MTV] [cm3] and total lesion glycolysis [TLG] [g]), and texture (entropy, homogeneity, dissimilarity, intensity variability [IV], size-zone variability [SZV], and zone percentage [ZP]) (MTV ≥ 5.0 cm3 and SUV ≥ 2.5 g/ml) parameters between 13 benign and 46 malignant lesions using the Mann-Whitney U test. Diagnostic performance was evaluated by receiver operating characteristic (ROC) analysis. Stepwise logistic regression analysis was performed to identify and use the independent variables that correctly differentiate between benign and malignant lesions. RESULTS: Malignant pulmonary lesions showed significantly higher SUVmax, SUVmean, MTV, TLG, entropy, dissimilarity, IV, and SZV and significantly lower homogeneity and ZP than benign pulmonary lesions (all p < 0.05) in both early and delayed images. Their areas under the ROC curves (AUCs) ranged between 0.69 and 0.94, and diagnostic accuracies between 64.4% and 93.2%. Entropy-early (p = 0.014), SUVmean-delay (p = 0.039), and dissimilarity-delay (p = 0.027) were independent parameters, and combined use of them yielded the highest AUC (0.98) with 100% sensitivity (46/46), 84.6% specificity (11/13), and 96.7% (57/59) accuracy for distinguishing between benign and malignant lesions. CONCLUSIONS: The individual early and delayed SUV-related, volumetric, and texture parameters showed a wide range of accuracy. Combined use of independent parameters extracted from DTP imaging might yield a high diagnostic accuracy with balanced sensitivity and specificity to differentiate between benign and malignant 18F-FDG-avid pulmonary lesions. KEY POINTS: • Malignant pulmonary lesions showed significantly higher SUV-related (SUVmax and SUVmean) and volumetric (MTV and TLG) parameters than benign pulmonary lesions in both early and delayed images. • Malignant pulmonary lesions showed significantly more heterogeneous18F-FDG uptake than benign pulmonary lesions in both early and delayed images. • Combined use of independent parameters extracted from DTP imaging might yield a high diagnostic accuracy to differentiate between benign and malignant18F-FDG-avid pulmonary lesions.

Impairment of Iodine-123-Metaiodobenzylguanidine (123I-MIBG) Uptake in Patients with Pulmonary Artery Hypertension.

According to recent studies, lung uptake of iodine-123-metaiodobenzylguanidine (123I-MIBG) is impaired in many lung diseases and low lung uptake of 123I-MIBG suggests endothelial dysfunction of the pulmonary artery. 123I-MIBG scintigraphy in patients with pulmonary hypertension (PH) has not yet been clinically evaluated. We hypothesized that the lung uptake of 123I-MIBG is reduced in patients with PH and differs among PH subtypes. The purpose of the present study was to analyze the lung uptake of 123I-MIBG in patients with PH and compare it with the data obtained by echocardiography or right heart catheterization. 123I-MIBG scintigraphy was performed in 286 consecutive patients from 2003 to 2014. We enrolled 21 patients with PH and 8 control patients. The 21 patients with PH were categorized into those with pulmonary artery hypertension (PAH, n = 12) and those with chronic thromboembolic pulmonary hypertension (CTEPH, n = 9). The mean pulmonary artery pressure was not significantly different between patients with CTEPH and PAH (37.7 ± 6.8 versus 32.3 ± 5.3 mmHg respectively; P = 0.054). There were no significant differences in any other hemodynamic parameters between the two groups. The lung uptake of 123I-MIBG in PAH patients (early image: 1.54 ± 0.18, delayed image: 1.41 ± 0.16) was significantly lower than that of CTEPH patients (early image: 2.17 ± 0.25, P < 0.0001; delayed image: 1.99 ± 0.20, P = 0.0001, adjusted for age and World Health Organization classification) and controls (early image: 2.32 ± 0.27, P = 0.0007; delayed image: 1.92 ± 0.19, P = 0.0007). In conclusion, we found for the first time that the lung uptake of 123I-MIBG in patients with PAH is lower than that in patients with CTEPH and controls.

A pilot study for texture analysis of 18F-FDG and 18F-FLT-PET/CT to predict tumor recurrence of patients with colorectal cancer who received surgery.

PURPOSE: This retrospective study was done to examine whether the heterogeneity in primary tumor F-18-fluorodeoxyglucose (18F-FDG) and 18F-3'-fluoro-3'-deoxythymidine (18F-FLT) distribution can predict prognosis of patients with colorectal cancer who received surgery. METHODS: The enrolled 32 patients with colorectal cancer underwent both 18F-FDG- and 18F-FLT-PET/CT studies before surgery. Clinicopathological factors, stage, SUVmax, SUVmean, metabolic tumor volume (SUV ≥ 2.5), total lesion glycolysis, total lesion proliferation and seven texture heterogeneity parameters (coefficient of variation, local parameters: entropy, homogeneity, and dissimilarity; and regional parameters: intensity variability [IV], size-zone variability [SZV], and zone percentage [ZP]) were obtained. Progression free survival (PFS) was calculated by the Kaplan-Meier method. Prognostic significance was assessed by Cox proportional hazards analysis. RESULTS: Eight patients had eventually come to progression, and 24 patients were alive without progression during clinical follow-up [mean follow-up PFS; 55.9 months (range, 1-72)]. High stage (p = 0.004), high 18F-FDG-IV (p = 0.015), high 18F-FDG-SZV (p = 0.013) and high 18F-FLT-entropy (p = 0.015) were significant in predicting poor 5-year PFS. Other parameters did not predict the disease outcome. At bivariate analysis, disease event hazards ratios for 18F-FDG-IV and 18F-FDG-SZV remained significant when adjusted for stage and 18F-FLT-entropy (18F-FDG-IV; p = 0.004 [adjusted for stage], 0.007 [adjusted for 18F-FLT-entropy]; 18F-FDG-SZV; p = 0.028 [adjusted for stage], 0.040 [adjusted for 18F-FLT-entropy]). CONCLUSION: 18F-FDG PET heterogeneity parameters, IV and SZV, have a potential to be strong prognostic factors to predict PFS of patients with surgically resected colorectal cancer and are more useful than 18F-FLT-PET/CT heterogeneity parameters.

Texture analysis of 18F-FDG PET/CT to predict tumour response and prognosis of patients with esophageal cancer treated by chemoradiotherapy.

PURPOSE: This retrospective study was done to examine whether the heterogeneity in primary tumour F-18-fluorodeoxyglucose (18F-FDG) distribution can predict tumour response and prognosis of patients with esophageal cancer treated by chemoradiotherapy (CRT). METHODS: The enrolled 52 patients with esophageal cancer underwent 18F-FDG-PET/CT studies before CRT. SUVmax, SUVmean, metabolic tumour volume (MTV, SUV ≥ 2.5), total lesion glycolysis (TLG) and six heterogeneity parameters assessed by texture analysis were obtained. Patients were classified as responders or non-responders according to Response Evaluation Criteria in Solid Tumors. Progression-free survival (PFS) and overall survival (OS) were calculated by the Kaplan-Meier method. Prognostic significance was assessed by Cox proportional hazards analysis. RESULTS: Thirty four non-responders showed significantly higher MTV (p = 0.006), TLG (p = 0.007), intensity variability (IV; p = 0.003) and size-zone variability (SZV; p = 0.004) than 18 responders. The positive and negative predictive values for non-responders were 77 % and 69 % in MTV, 76 % and 100 % in TLG, 78 % and 67 % in IV and 78 % and 82 % in SZV, respectively. Although PFS and OS were significantly shorter in patients with high MTV (PFS, p = 0.018; OS, p = 0.014), TLG (PFS, p = 0.009; OS, p = 0.025), IV (PFS, p = 0.013; OS, p = 0.007) and SZV (PFS, p = 0.010; OS, p = 0.007) at univariate analysis, none of them was an independent factor, while lymph node status, stage and tumour response status were independent factors at multivariate analysis. CONCLUSION: Texture features IV and SZV, and volumetric parameters MTV and TLG can predict tumour response, but all of them have limited value in prediction of prognosis of patients with esophageal cancer treated by CRT.

Dexamethasone Suppression FDG PET/CT for Differentiating between True- and False-Positive Pulmonary and Mediastinal Lymph Node Metastases in Non-Small Cell Lung Cancer: A Pilot Study of FDG PET/CT after Oral Administration of Dexamethasone.

PURPOSE: To examine whether dexamethasone suppression can reduce fluorine 18 fluorodeoxyglucose (FDG) uptake in false-positive (FP) findings in pulmonary and mediastinal lymph nodes in non-small cell lung cancer (NSCLC). MATERIALS AND METHODS: Institutional ethics review board approved this prospective study with written informed consent. The study population was composed of 17 patients with NSCLC who underwent both baseline and dexamethasone suppression (24 hours after oral administration of 8 mg dexamethasone) FDG positron emission tomography/computed tomography and surgery. FDG uptake was evaluated by using a five-point visual scoring system (negative findings, score of 0-1; positive findings, score of 2-4) and maximum standardized uptake value (SUVmax). The Mann-Whitney U, Wilcoxon signed-rank, Kruskal-Wallis, or Spearman rank correlation tests were used as necessary for statistical evaluations. RESULTS: In 17 primary lesions, no significant difference was noted in visual score between baseline (mean, 3.4 ± 1.2) and dexamethasone suppression scans (mean, 3.3 ± 1.2; P = .16), although SUVmax was significantly lower on dexamethasone suppression scans (mean, 7.1 ± 5.2) than on baseline scans (mean, 8.6 ± 6.6; P = .005). In eight nodes with true-positive (TP) findings, there were no significant differences in visual score (mean for both, 3.8 ± 0.5) and SUVmax (mean, 5.3 ± 2.3 vs 5.5 ± 2.5, respectively; P = .81) between baseline and dexamethasone suppression scans. In 19 nodes with FP findings at baseline, dexamethasone suppression resulted in significantly lowered visual score (mean, 3.4 ± 0.6 vs 2.4 ± 0.8, respectively; P < .001) and SUVmax (mean, 3.5 ± 0.8 vs 2.7 ± 0.7, respectively; P < .001), and four nodes with FP findings were rated as true-negative findings on dexamethasone suppression scans, which resulted in a significant difference in SUVmax between nodal lesions with TP and FP findings (P = .014). CONCLUSION: Oral dexamethasone has the potential to reduce FDG uptake in pulmonary and mediastinal nodes with FP findings in NSCLC.

FDG-PET/CT and FLT-PET/CT for differentiating between lipid-poor benign and malignant adrenal tumours.

OBJECTIVE: To compare F-18-fluorodeoxyglucose (FDG) and F-18-fluorothymidine (FLT) PET/CT examinations for differentiating between benign and malignant adrenal tumours. METHODS: Thirty lipid-poor benign and 11 malignant tumours of 40 patients were included. FDG- and FLT-based indices including visual score, maximum standardized uptake value (SUVmax) and FDG adrenal lesion/liver SUVmax (A/L SUVmax) or FLT adrenal lesion/back muscle SUVmax (A/B SUVmax) ratio were compared between benign and malignant tumours using the Mann-Whitney's U or Wilcoxon signed-rank test, and their diagnostic performances were evaluated by means of the area under the curve (AUC) values derived from the receiver operating characteristic analysis. RESULTS: All indices were significantly higher in malignant than benign tumours on both images (p < 0.05 each). On FDG-PET/CT, the sensitivity, specificity, and accuracy were 91 %, 63 % and 71 % for visual score, 91 %, 67 % and 73 % for SUVmax, and 100 %, 70 % and 78 % for A/L SUVmax ratio, respectively. On FLT-PET/CT, they were 100 %, 97 % and 98 % for visual score, SUVmax and A/B SUVmax ratio, respectively. All FLT indices were significantly higher than those of FDG in AUC (p < 0.05 each). CONCLUSION: FLT-PET/CT may be superior to FDG-PET/CT in differentiating lipid-poor benign from malignant adrenal tumours because of higher specificity and accuracy. KEY POINTS: • All FDG indices were significantly higher in malignant than in benign tumours. • All FLT indices were significantly higher in malignant than in benign tumours. • All FLT indices were significantly higher than those of FDG in AUC.

Diagnostic performances of FDG-PET/CT and diffusion-weighted imaging indices for differentiating benign pheochromocytoma from other benign adrenal tumors.

PURPOSE: The purpose of this study is to compare diagnostic performances of (18)F-fluorodeoxyglucose (FDG) visual score, maximum standardized uptake value (SUVmax), ratio of adrenal SUVmax to liver SUVmax (A/L SUVmax), apparent diffusion coefficient (ADC) from diffusion-weighted imaging, and SUVmax/ADC ratio to differentiate adrenal pheochromocytoma from other benign tumors. METHODS: Eleven pheochromocytomas and 22 other benign tumors in 30 patients were included. FDG-based indices, ADC, and SUVmax/ADC ratio were compared between groups using the Mann-Whitney U test, and sensitivity, specificity, accuracy, and area under the curve (AUC) for diagnosing pheochromocytoma by receiver operating characteristic analyses. The correlation between SUVmax and ADC was analyzed using the Spearman's rank test. RESULTS: Pheochromocytoma showed significantly higher visual score (2.8 ± 0.4 vs. 1.3 ± 0.9), SUVmax (11.0 ± 8.9 vs. 3.2 ± 1.4), A/L SUVmax ratio (3.96 ± 3.48 vs. 0.96 ± 0.51), and SUVmax/ADC ratio (10.6 ± 8.09 vs. 2.28 ± 0.98) (each P < 0.001) and significantly lower ADC (1.08 ± 0.23 × 10(-3) mm(2)/s vs. 1.43 ± 0.29 × 10(-3) mm(2)/s, P = 0.003) than other benign tumors. Sensitivity, specificity, and accuracy for diagnosing pheochromocytoma were 100, 73, and 82% for visual score, 100, 86, and 91% for both SUVmax and A/L SUVmax ratio, and 64, 100, and 88% for ADC and 82, 95, and 91% for SUVmax/ADC ratio. No significant differences in AUC were found between FDG-based indices, ADC, and SUVmax/ADC ratio. A significant negative correlation was noted between SUVmax and ADC (ρ = -0.36, P = 0.039). CONCLUSION: FDG-based indices and ADC appear comparably useful for differentiating pheochromocytoma from other benign adrenal tumors.

[Draft guideline regarding appropriate use of 131I-MIBG radiotherapy for neuroendocrine tumors Drafting Committee for Guideline of Radiotherapy with 131I-MIBG, Committee for Nuclear Oncology and Immunology, The Japanese Society of Nuclear Medicine].

131I-MIBG radiotherapy has been used for unresectable nueroendocrine tumors including malignant pheochromocytomas and neuroblastomas in foreign countries since the '80s when clinical therapeutic trials were initiated. In Japan, 131I-MIBG radiotherapy has not been approved by Ministry of Health, Labor and Welfare, however, personally imported 131I-MIBG is now available in limited institutions for therapeutic purpose. This updated guideline draft aims to provide useful information concerning 131I-MIBG radiotherapy, to prevent side effects, and to protect physicians, nurses, other health care professionals, patients and their families from radiation exposure. The committee also provides appendices including practical guidance for attending physicians, patient management and referring physicians for their conveniences.

Correlations of (18)F-fluorothymidine uptake with pathological tumour size, Ki-67 and thymidine kinase 1 expressions in primary and metastatic lymph node colorectal cancer foci.

OBJECTIVE: To examine correlations of (18)F-fluorothymidine (FLT) uptake with pathological tumour size and immunohistochemical Ki-67, and thymidine kinase 1 (TK-1) expressions in primary and metastatic node colorectal cancer foci. METHODS: Thirty primary cancers (PCs) and 37 metastatic nodes (MNs) were included. FLT uptake was assessed by visual scores (non-visible: 0-1 and visible: 2-4), standardized uptake value (SUV), and correlated with size, Ki-67, and TK-1. SUV was measured in visible lesions. FLT heterogeneity was assessed by visual scores (no heterogeneous uptake: 0 and heterogeneous uptake: 1-4). RESULTS: Forty-two lesions were visible. The visible group showed significantly higher values than the non-visible group in size, Ki-67, and TK-1 (each p < 0.05). Size correlated significantly with visual score (PC; ρ = 0.74 and MN; ρ = 0.63), SUVmax (PC; ρ = 0.49, and MN; ρ = 0.76), and SUVmean (PC; ρ = 0.40 and MN; ρ = 0.76) (each p < 0.05). Visual score correlated significantly with size (ρ = 0.86), Ki-67max (ρ = 0.35), Ki-67mean (ρ = 0.38), TK-1max (ρ = 0.35) and TK-1mean (ρ = 0.25) (each p < 0.05). No significant correlations were found between FLT uptake and Ki-67 or TK-1 in 42 visible lesions (each p > 0.05). Heterogeneous FLT uptake was noted in 73 % (22/30) of PCs. CONCLUSION: FLT uptake correlated with size. Heterogeneous FLT distribution in colorectal cancers may be one of the causes of weak or lack of FLT uptake/Ki-67 or TK-1 correlation. KEY POINTS: FLT uptake correlated well with tumour size in colorectal cancer. Weak or lack of FLT uptake/Ki-67 and TK-1 correlations were observed. Immunohistochemical Ki-67 and TK-1 expressions are not always correlated with FLT uptake.

Effects and safety of ¹³¹I-metaiodobenzylguanidine (MIBG) radiotherapy in malignant neuroendocrine tumors: results from a multicenter observational registry.

Effective treatments for malignant neuroendocrine tumors are under development. While iodine-131 metaiodobenzylguanidine (¹³¹I-MIBG) radiotherapy has been used in the treatment of malignant neuroendocrine tumors, there are few studies evaluating its therapeutic effects and safety in a multicenter cohort. In the current study, we sought to evaluate the effects and safety of ¹³¹I-MIBG therapy for conditions including malignant pheochromocytoma and paraganglioma within a multicenter cohort. Forty-eight malignant neuroendocrine tumors (37 pheochromocytoma and 11 paraganglioma) from four centers underwent clinical ¹³¹I-MIBG radiotherapy. The tumor responses were observed before and 3 to 6 months after the ¹³¹I-MIBG radiotherapy in accordance with RECIST criteria. We also evaluated the data for any adverse effects. The four centers performed a total of 87 ¹³¹I-MIBG treatments on 48 patients between January 2000 and March 2009. Of the treatments, 65 were evaluable using RECIST criteria. One partial response (PR), 40 stable disease (SD), and 9 progressive disease (PD) in malignant pheochromocytoma were observed after each treatment. Fourteen SD and one PD-were observed in paraganglioma. Patients with normal hypertension (systolic blood pressure (BP) > 130 mmHg) showed significantly reduced systolic BP after the initial follow-up (n=10, 138.1±8.2 to 129.5±13.5 mmHg, P=0.03). In adult neuroendocrine tumors with a treatment-basis analysis, there were side effects following 41 treatments (47.1%) and most of them (90.2%) were minor. In this multicenter registry, PR or SD was achieved in 84.6% of the treatment occasions in adult neuroendocrine tumors through ¹³¹I-MIBG radiotherapy. This indicated that most of the ¹³¹I-MIBG radiotherapy was performed safely without significant side effects.

Clinical application of 18F-fluorodeoxyglucose positron emission tomography/computed tomography radiomics-based machine learning analyses in the field of oncology.

Machine learning (ML) analyses using 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography (PET)/computed tomography (CT) radiomics features have been applied in the field of oncology. The current review aimed to summarize the current clinical articles about 18F-FDG PET/CT radiomics-based ML analyses to solve issues in classifying or constructing prediction models for several types of tumors. In these studies, lung and mediastinal tumors were the most commonly evaluated lesions, followed by lymphatic, abdominal, head and neck, breast, gynecological, and other types of tumors. Previous studies have commonly shown that 18F-FDG PET radiomics-based ML analysis has good performance in differentiating benign from malignant tumors, predicting tumor characteristics and stage, therapeutic response, and prognosis by examining significant differences in the area under the receiver operating characteristic curves, accuracies, or concordance indices (> 0.70). However, these studies have reported several ML algorithms. Moreover, different ML models have been applied for the same purpose. Thus, various procedures were used in 18F-FDG PET/CT radiomics-based ML analysis in oncology, and 18F-FDG PET/CT radiomics-based ML models, which are easy and universally applied in clinical practice, would be expected to be established.

Machine learning approach using 18F-FDG-PET-radiomic features and the visibility of right ventricle 18F-FDG uptake for predicting clinical events in patients with cardiac sarcoidosis.

OBJECTIVES: To investigate the usefulness of machine learning (ML) models using pretreatment 18F-FDG-PET-based radiomic features for predicting adverse clinical events (ACEs) in patients with cardiac sarcoidosis (CS). MATERIALS AND METHODS: This retrospective study included 47 patients with CS who underwent 18F-FDG-PET/CT scan before treatment. The lesions were assigned to the training (n = 38) and testing (n = 9) cohorts. In total, 49 18F-FDG-PET-based radiomic features and the visibility of right ventricle 18F-FDG uptake were used to predict ACEs using seven different ML algorithms (namely, decision tree, random forest [RF], neural network, k-nearest neighbors, Naïve Bayes, logistic regression, and support vector machine [SVM]) with tenfold cross-validation and the synthetic minority over-sampling technique. The ML models were constructed using the top four features ranked by the decrease in Gini impurity. The AUCs and accuracies were used to compare predictive performances. RESULTS: Patients who developed ACEs presented with a significantly higher surface area and gray level run length matrix run length non-uniformity (GLRLM_RLNU), and lower neighborhood gray-tone difference matrix_coarseness and sphericity than those without ACEs (each, p < 0.05). In the training cohort, all seven ML algorithms had a good classification performance with AUC values of > 0.80 (range: 0.841-0.944). In the testing cohort, the RF algorithm had the highest AUC and accuracy (88.9% [8/9]) with a similar classification performance between training and testing cohorts (AUC: 0.945 vs 0.889). GLRLM_RLNU was the most important feature of the modeling process of this RF algorithm. CONCLUSION: ML analyses using 18F-FDG-PET-based radiomic features may be useful for predicting ACEs in patients with CS.

Applying deep learning-based ensemble model to [18F]-FDG-PET-radiomic features for differentiating benign from malignant parotid gland diseases.

OBJECTIVES: To develop and identify machine learning (ML) models using pretreatment 2-deoxy-2-[18F]fluoro-D-glucose ([18F]-FDG)-positron emission tomography (PET)-based radiomic features to differentiate benign from malignant parotid gland diseases (PGDs). MATERIALS AND METHODS: This retrospective study included 62 patients with 63 PGDs who underwent pretreatment [18F]-FDG-PET/computed tomography (CT). The lesions were assigned to the training (n = 44) and testing (n = 19) cohorts. In total, 49 [18F]-FDG-PET-based radiomic features were utilized to differentiate benign from malignant PGDs using five different conventional ML algorithmic models (random forest, neural network, k-nearest neighbors, logistic regression, and support vector machine) and the deep learning (DL)-based ensemble ML model. In the training cohort, each conventional ML model was constructed using the five most important features selected by the recursive feature elimination method with the tenfold cross-validation and synthetic minority oversampling technique. The DL-based ensemble ML model was constructed using the five most important features of the bagging and multilayer stacking methods. The area under the receiver operating characteristic curves (AUCs) and accuracies were used to compare predictive performances. RESULTS: In total, 24 benign and 39 malignant PGDs were identified. Metabolic tumor volume and four GLSZM features (GLSZM_ZSE, GLSZM_SZE, GLSZM_GLNU, and GLSZM_ZSNU) were the five most important radiomic features. All five features except GLSZM_SZE were significantly higher in malignant PGDs than in benign ones (each p < 0.05). The DL-based ensemble ML model had the best performing classifier in the training and testing cohorts (AUC = 1.000, accuracy = 1.000 vs AUC = 0.976, accuracy = 0.947). CONCLUSIONS: The DL-based ensemble ML model using [18F]-FDG-PET-based radiomic features can be useful for differentiating benign from malignant PGDs. The DL-based ensemble ML model using [18F]-FDG-PET-based radiomic features can overcome the previously reported limitation of [18F]-FDG-PET/CT scan for differentiating benign from malignant PGDs. The DL-based ensemble ML approach using [18F]-FDG-PET-based radiomic features can provide useful information for managing PGD.

Usefulness of the Primary Tumor Standardized Uptake Value of Iodine-123 Metaiodobenzylguanidine for Predicting Metastatic Potential in Pheochromocytoma and Paraganglioma.

PURPOSE: To examine the usefulness of semi-quantitative analysis using the standardized uptake value (SUV) of iodine-123 metaiodobenzylguanidine ([123I]-MIBG) for predicting metastatic potential in patients with pheochromocytoma (PHEO) and paraganglioma (PGL). PROCEDURES: This study included 18 PHEO and 2 PGL patients. [123I]-MIBG visibility and SUV-related parameters (SUVmax, SUVmean, tumor volume of [123I]-MIBG uptake [TV_MIBG], and total lesion [123I]-MIBG uptake) were compared with the pathological grading obtained using the Pheochromocytoma of the Adrenal Gland Scaled Score (PASS) and the Grading System for Adrenal Pheochromocytoma and Paraganglioma (GAPP), which are used to predict metastatic potential. The PASS scores were categorized as < 4 and ≥ 4. Based on the GAPP scores, PHEOs/PGLs were categorized as follows: well, moderately, and poorly differentiated tumors. The Mann-Whitney U test or Spearman's rank correlation was used to assess differences or associations between two quantitative variables. RESULTS: All PHEOs/PGLs were visualized on [123I]-MIBG scintigraphy. There were 16 PASS < 4 and 4 PASS ≥ 4 tumors. Moreover, 11 and 9 tumors were well and moderately differentiated, respectively. The uptake scores and SUV-related parameters significantly differed between tumors with a PASS score of < 4 and those with a PASS score of ≥ 4 (each, p > 0.05). Moderately differentiated tumors had significantly higher uptake scores and SUV-related parameters except TV_MIBG than well-differentiated tumors (each, p < 0.05). The GAPP score was positively correlated with the uptake scores and SUV-related parameters (each, p < 0.05) except TV_MIBG. CONCLUSIONS: The primary tumor [123I]-MIBG uptake assessed using SUV-related parameters can be an imaging tool for predicting metastatic potential in patients with PHEO/PGL.

Diagnosis of metastases from postoperative differentiated thyroid cancer: comparison between FDG and FLT PET/CT studies.

PURPOSE: To compare positron emission tomography (PET)/computed tomography (CT) studies performed with the glucose analog fluorine 18 ((18)F) fluorodeoxyglucose (FDG) and the cell proliferation tracer (18)F fluorothymidine (FLT) in the diagnosis of metastases from postoperative differentiated thyroid cancer. MATERIALS AND METHODS: The institutional ethics review board approved this prospective study. From March 2010 to February 2012, 20 patients (mean age, 53 years; age range, 22-79 years) with postoperative differentiated thyroid cancer underwent both FDG and FLT PET/CT as a staging work-up before radioiodine therapy. In each patient, 28 anatomic areas were set and analyzed for lymph node and distant metastases. The McNemar exact or χ(2) test was used to examine differences in diagnostic indexes in the detection of lymph node and distant metastases between both tracer PET/CT studies. RESULTS: There were 34 lymph node metastases and/or 73 distant metastases (70 metastases in lung and one each in bone, nasopharynx, and brain) in 13 patients. At patient-based analysis, the sensitivity, specificity, and accuracy were 92% (12 of 13 patients), 86% (six of seven patients), and 90% (18 of 20 patients), respectively, for FDG PET/CT and 69% (nine of 13 patients), 29% (two of seven patients), and 55% (11 of 20 patients) for FLT PET/CT. The accuracy of FDG PET/CT was significantly better than that of FLT PET/CT (P = .023). At lesion-based analysis, the sensitivity, specificity, and accuracy for diagnosing lymph node metastases were 85% (29 of 34 lesions), 99.6% (245 of 246 lesions), and 97.9% (274 of 280 lesions), respectively, for FDG PET/CT and 50% (17 of 34 lesions), 90.7% (223 of 246 lesions), and 85.7% (240 of 280 lesions) for FLT PET/CT. The sensitivity, specificity, and accuracy for diagnosing distant metastases were 45% (33 of 73 lesions), 100% (207 of 207 lesions), and 85.7% (240 of 280 lesions), respectively, for FDG PET/CT and 6.8% (five of 73 lesions), 100% (207 of 207 lesions), and 75.7% (212 of 280 lesions) for FLT PET/CT. The sensitivity (P = .002), specificity (P < .001), and accuracy (P < .001) of FDG PET/CT in the diagnosis of lymph node metastases were superior to those of FLT PET, as were the sensitivity (P < .001) and accuracy (P < .001) in the diagnosis of distant metastases. CONCLUSION: FDG PET/CT is superior to FLT PET/CT in the diagnosis of postoperative differentiated thyroid cancer lymph node and distant metastases. Thus, FDG PET/CT is more suitable than FLT PET/CT for examining recurrence of postoperative differentiated thyroid cancer.

Diagnostic performance of ¹8F-fluorothymidine PET/CT for primary colorectal cancer and its lymph node metastasis: comparison with ¹8F-fluorodeoxyglucose PET/CT.

PURPOSE: To examine the diagnostic performance of (18)F-fluorothymidine (FLT) PET/CT in primary and metastatic lymph node colorectal cancer foci in comparison with (18)F-fluorodeoxyglucose (FDG) PET/CT. METHODS: The study population comprised 28 patients with 30 newly diagnosed colorectal cancers who underwent surgical resection of the primary lesion and regional lymph nodes after both FLT and FDG PET/CT. The associations between SUVmax levels and pathological factors were evaluated using the Mann-Whitney U or Kruskal-Wallis test. Differences in diagnostic indexes for detecting nodal metastasis between the two tracers were estimated using the McNemar exact or χ(2) test. RESULTS: All 30 primary cancers (43.0 ± 20.0 mm, range 14 - 85 mm) were visualized by both tracers, but none of the FLT SUVmax values exceeded the FDG SUVmax values in any of the primary cancers (6.6 ± 2.4 vs. 13.6 ± 5.8, p < 0.001). The sensitivity, specificity and accuracy for detecting nodal metastasis were 41% (15/37), 98.8% (493/499) and 94.8% (508/536) for FDG PET/CT, and 32% (12/37), 98.8% (493/499) and 94.2% (505/536) for FLT PET/CT, respectively. The sensitivity (p = 0.45), specificity (p = 0.68) and accuracy (p = 0.58) were not different between the tracers. Nodal uptake of FLT and FDG was discordant in 7 (19%) of 37 metastatic nodes. There were ten concordant true-positive nodes of which six showed higher FDG SUVmax and four showed higher FLT SUVmax, but the difference between FDG and FLT SUVmax was not significant (5.56 ± 3.55 and 3.62 ± 1.45, respectively; p = 0.22). CONCLUSION: FLT has the same potential as FDG in PET/CT for the diagnosis of primary and nodal foci of colorectal cancer despite significantly lower FLT uptake in primary foci.

Application of 123I-MIBG myocardial maximum standardized uptake value to characterize cardiac function in patients with pheochromocytoma: comparison with echocardiography.

PURPOSE: This study examined the usefulness of the maximum standardized uptake value (SUVmax) of myocardial [123I]-metaiodobenzylguanidine ([123I]-MIBG) to characterize myocardial function by comparing it with echocardiographic parameters in patients with pheochromocytoma. MATERIALS AND METHODS: This study included 18 patients with pheochromocytoma who underwent both planar and [123I]-MIBG single-photon emission computed tomography/computed tomography scans and echocardiography before surgery. Myocardial [123I]-MIBG visibility and SUVmax were compared with echocardiographic parameters related to systolic and diastolic functions. The Mann-Whitney U test, Fisher exact test, or Spearman rank correlation assessed differences or relationships between two quantitative variables. RESULTS: On visual analysis, 6 patients showed normal myocardial [123I]-MIBG uptake, whereas 12 patients showed decreased myocardial [123I]-MIBG uptake. No patients showed systolic dysfunction. A significant difference was observed in the incidence of diastolic dysfunction between the groups with normal and decreased uptake (p = 0.009), and left ventricular (LV) diastolic dysfunction was observed in 9 (75%) of 12 patients with decreased myocardial uptake. The myocardial SUVmax was significantly lower in 9 patients with LV diastolic dysfunction than in 9 patients with normal cardiac function (1.67 ± 0.37 vs. 3.03 ± 1.38, p = 0.047). Myocardial SUVmax was positively correlated with septal e' (early diastolic velocity of septal mitral annulus) (ρ = 0.51, p = 0.031) and negatively correlated with the septal E/e' ratio (early mitral E-velocity to early diastolic velocity of septal mitral annulus; ρ = - 0.64, p = 0.004), respectively. CONCLUSIONS: LV diastolic dysfunction was inversely related to myocardial [123I]-MIBG uptake. Myocardial [123I]-MIBG SUVmax may be useful for characterizing cardiac function in patients with pheochromocytoma. Second abstract. The semiquantitative analysis using the myocardial SUVmax in 123I-MIBG SPECT/CT was found to be potentially useful for characterizing cardiac function in patients with pheochromocytoma.

Application of Machine Learning Analyses Using Clinical and [18F]-FDG-PET/CT Radiomic Characteristics to Predict Recurrence in Patients with Breast Cancer.

PURPOSE: To develop and identify machine learning (ML) models using pretreatment clinical and 2-deoxy-2-[18F]fluoro-D-glucose positron emission tomography ([18F]-FDG-PET)-based radiomic characteristics to predict disease recurrences in patients with breast cancers who underwent surgery. PROCEDURES: This retrospective study included 112 patients with 118 breast cancer lesions who underwent [18F]-FDG-PET/ X-ray computed tomography (CT) preoperatively, and these lesions were assigned to training (n=95) and testing (n=23) cohorts. A total of 12 clinical and 40 [18F]-FDG-PET-based radiomic characteristics were used to predict recurrences using 7 different ML algorithms, namely, decision tree, random forest (RF), neural network, k-nearest neighbors, naive Bayes, logistic regression, and support vector machine (SVM) with a 10-fold cross-validation and synthetic minority over-sampling technique. Three different ML models were created using clinical characteristics (clinical ML models), radiomic characteristics (radiomic ML models), and both clinical and radiomic characteristics (combined ML models). Each ML model was constructed using the top ten characteristics ranked by the decrease in Gini impurity. The areas under ROC curves (AUCs) and accuracies were used to compare predictive performances. RESULTS: In training cohorts, all 7 ML algorithms except for logistic regression algorithm in the radiomics ML model (AUC = 0.760) achieved AUC values of >0.80 for predicting recurrences with clinical (range, 0.892-0.999), radiomic (range, 0.809-0.984), and combined (range, 0.897-0.999) ML models. In testing cohorts, the RF algorithm of combined ML model achieved the highest AUC and accuracy (95.7% (22/23)) with similar classification performance between training and testing cohorts (AUC: training cohort, 0.999; testing cohort, 0.992). The important characteristics for modeling process of this RF algorithm were radiomic GLZLM_ZLNU and AJCC stage. CONCLUSIONS: ML analyses using both clinical and [18F]-FDG-PET-based radiomic characteristics may be useful for predicting recurrence in patients with breast cancers who underwent surgery.

The usefulness of machine-learning-based evaluation of clinical and pretreatment 18F-FDG-PET/CT radiomic features for predicting prognosis in patients with laryngeal cancer.

OBJECTIVE: To examine whether machine learning (ML) analyses involving clinical and 18F-FDG-PET-based radiomic features are helpful in predicting prognosis in patients with laryngeal cancer. METHODS: This retrospective study included 49 patients with laryngeal cancer who underwent18F-FDG-PET/CT before treatment, and these patients were divided into the training (n = 34) and testing (n = 15) cohorts.Seven clinical (age, sex, tumor size, T stage, N stage, Union for International Cancer Control stage, and treatment) and 40 18F-FDG-PET-based radiomic features were used to predict disease progression and survival. Six ML algorithms (random forest, neural network, k-nearest neighbors, naïve Bayes, logistic regression, and support vector machine) were used for predicting disease progression. Two ML algorithms (cox proportional hazard and random survival forest [RSF] model) considering for time-to-event outcomes were used to assess progression-free survival (PFS), and prediction performance was assessed by the concordance index (C-index). RESULTS: Tumor size, T stage, N stage, GLZLM_ZLNU, and GLCM_Entropy were the five most important features for predicting disease progression.In both cohorts, the naïve Bayes model constructed by these five features was the best performing classifier (training: AUC = 0.805; testing: AUC = 0.842). The RSF model using the five features (tumor size, GLZLM_ZLNU, GLCM_Entropy, GLRLM_LRHGE and GLRLM_SRHGE) exhibited the highest performance in predicting PFS (training: C-index = 0.840; testing: C-index = 0.808). CONCLUSION: ML analyses involving clinical and 18F-FDG-PET-based radiomic features may help predict disease progression and survival in patients with laryngeal cancer. ADVANCES IN KNOWLEDGE: ML approach using clinical and 18F-FDG-PET-based radiomic features has the potential to predict prognosis of laryngeal cancer.