Application of radiomics and machine learning to thyroid diseases in nuclear medicine: a systematic review.

BACKGROUND: In the last years growing evidences on the role of radiomics and machine learning (ML) applied to different nuclear medicine imaging modalities for the assessment of thyroid diseases are starting to emerge. The aim of this systematic review was therefore to analyze the diagnostic performances of these technologies in this setting. METHODS: A wide literature search of the PubMed/MEDLINE, Scopus and Web of Science databases was made in order to find relevant published articles about the role of radiomics or ML on nuclear medicine imaging for the evaluation of different thyroid diseases. RESULTS: Seventeen studies were included in the systematic review. Radiomics and ML were applied for assessment of thyroid incidentalomas at 18 F-FDG PET, evaluation of cytologically indeterminate thyroid nodules, assessment of thyroid cancer and classification of thyroid diseases using nuclear medicine techniques. CONCLUSION: Despite some intrinsic limitations of radiomics and ML may have affect the results of this review, these technologies seem to have a promising role in the assessment of thyroid diseases. Validation of preliminary findings in multicentric studies is needed to translate radiomics and ML approaches in the clinical setting.

Multicenter Evaluation of the Feasibility of Clinical Implementation of SPECT Myocardial Blood Flow Measurement: Intersite Variability and Imaging Time.

BACKGROUND: Single-center studies have shown that single photon emission computed tomography myocardial blood flow (MBF) measurement is accurate compared with MBF measured with microspheres in a porcine model, positron emission tomography, and angiography. Clinical implementation requires consistency across multiple sites. The study goal is to determine the intersite processing repeatability of single photon emission computed tomography MBF and the additional camera time required. METHODS: Five sites (Canada, Italy, Japan, Germany, and Singapore) each acquired 25 to 35 MBF studies at rest and with pharmacological stress using technetium-99m-tetrofosmin on a pinhole-collimated cadmium-zinc-telluride-based cardiac single photon emission computed tomography camera with standardized list-mode imaging and processing protocols. Patients had intermediate to high pretest probability of coronary artery disease. MBF was measured locally and at a core laboratory using commercially available software. The time a room was occupied for an MBF study was compared with that for a standard rest/stress myocardial perfusion study. RESULTS: With motion correction, the overall correlation in MBF between core laboratory and local site was 0.93 (range, 0.87-0.97) at rest, 0.90 (range, 0.84-0.96) at stress, and 0.84 (range, 0.70-0.92) for myocardial flow reserve. The local-to-core difference in global MBF (bias-MBF) was 5.4% (-3.8% to 14.8%; median [interquartile range]) at rest and 5.4% (-6.2% to 19.4%) at stress. Between the 5 sites, bias-MBF ranged from -1.6% to 11.0% at rest and from -1.9% to 16.3% at stress; the interquartile range in bias-MBF was between 9.3% (4.8%-14.0%) and 22.3% (-10.3% to 12.0%) at rest and between 17.0% (-11.3% to 5.6%) and 33.3% (-10.4% to 22.9%) at stress and was not significantly different between most sites. Both bias and interquartile range were like previously reported interobserver variability and less than the SD of the test-retest difference of 30%. The overall difference in myocardial flow reserve was 1.52% (-10.6% to 11.3%). There were no significant differences between with and without motion correction. The average additional acquisition time varied between sites from 44 to 79 minutes. CONCLUSIONS: The average bias-MBF and bias-MFR values were small with standard deviations substantially less than the test-retest variability. This demonstrates that MBF can be measured consistently across multiple sites and further supports that this technique can be reliably implemented. REGISTRATION: URL: https://www. CLINICALTRIALS: gov; Unique identifier: NCT03427749.

PET/CT and Conventional Imaging for the Assessment of Neuroendocrine Prostate Cancer: A Systematic Review.

BACKGROUND: Neuroendocrine prostate cancer (NEPC) is a rare neoplasm, and the role of both conventional imaging (CI) and positron emission tomography/computed tomography (PET/CT) for its assessment has not been clearly evaluated and demonstrated. The aim of this systematic review was to analyze the diagnostic performances of these imaging modalities in this setting. METHODS: A wide literature search of the PubMed/MEDLINE, Scopus, and Web of Science databases was made to find relevant published articles about the role of CI and PET/CT for the evaluation of NEPC. RESULTS: 13 studies were included in the systematic review. PET/CT imaging with different radiopharmaceuticals has been evaluated in many studies (10) compared to CI (3 studies), which has only a limited role in NEPC. Focusing on PET/CT, a study used [18F]FDG, labeled somatostatin analogs were used in 5 cases, a study used [68Ga]Ga-FAPI-04, [68Ga]Ga-PSMA-11 was evaluated in a single case, and two works used different tracers. CONCLUSION: Published data on the role of PET/CT for the assessment of NEPC are limited. At present, it is still uncertain which tracer performs best, and although [18F]FDG has been evaluated and seems to offer some advantages in availability and clinical staging, other tracers may be more useful to understand tumor biology or identify targets for subsequent radioligand therapy. Further research is therefore desirable. In contrast, data are still limited to draw a final conclusion on the role and the specific characteristics of CI in this rare form of neoplasm, and therefore, more studies are needed in this setting.

EANM procedural recommendations for managing the paediatric patient in diagnostic nuclear medicine.

PURPOSE: The manuscript aims to characterize the principles of best practice in performing nuclear medicine procedures in paediatric patients. The paper describes all necessary technical skills that should be developed by the healthcare professionals to ensure the best possible care in paediatric patients, as it is particularly challenging due to psychological and physical conditions of children. METHODS: We performed a comprehensive literature review to establish the most relevant elements of nuclear medicine studies in paediatric patients. We focused the attention to the technical aspects of the study, such as patient preparation, imaging protocols, and immobilization techniques, that adhere to best practice principles. Furthermore, we considered the psychological elements of working with children, including comforting and distraction strategies. RESULTS: The extensive literature review combined with practical conclusions and recommendations presented and explained by the authors summarizes the most important principles of the care for paediatric patient in the nuclear medicine field. CONCLUSION: Nuclear medicine applied to the paediatric patient is a very special and challenging area, requiring proper education and experience in order to be performed at the highest level and with the maximum safety for the child.

Role of splenic and bone marrow uptake at 18 F-FDG PET/CT for the assessment of large vessels vasculitis and the influence of glucocorticoids therapy on their values.

INTRODUCTION: To assess the relationship between splenic and bone marrow (BM) uptake with the presence of large vessel vasculitis (LVV) at 18 F-FDG PET/CT and to evaluate the influence of glucocorticoid (GC) therapy on these uptakes. METHODS: One hundred and one subjects with LVV and 18 F-FDG PET/CT were included in the study. Clinical features, including blood samples and duration of GC therapy, were collected. Standardized uptake value body weight max (SUVmax) of the spleen, BM, liver and arterial walls were extracted; spleen/liver (SL) and BM/liver (BML) ratios were calculated. Chi-square and T-test were used to assess the relationship between PET/CT parameters and clinical features with the presence of LVV. Rank correlation was used to evaluate the correlation between PET/CT parameters and clinical parameters. Receiver operating curve (ROC) analysis was used to find the best parameter able to discriminate between positive and negative PET/CT. All analyses were performed considering the duration of GC therapy. RESULTS: Significant correlation for PET/CT results with spleen uptake (P-value = 0.001), SL (P-value < 0.001) and BML (P-value = 0.005) were reported in patients with no more than 3 days of therapy; the correlation with SL was confirmed in the total cohort of patients. A value of 0.92 for SL had an AUC of 0.959, a sensitivity of 92.6% and a specificity of 96.6% (P-value < 0.001) in predicting PET/CT results. CONCLUSION: Higher splenic and BM uptake in patients with positive PET/CT for LVV were reported. A long duration of GC therapy is able to reduce such uptakes.

Relationship between Baseline [18F]FDG PET/CT Semiquantitative Parameters and BRCA Mutational Status and Their Prognostic Role in Patients with Invasive Ductal Breast Carcinoma.

AIM: To assess the relationship between [18F]FDG PET/CT, breast cancer gene (BRCA) status, and their prognostic role in patients with ductal breast cancer (DBC). METHODS: Forty-one women were included. PET/CT semiquantitative parameters such as standardized uptake value (SUV) body weight max (SUVmax), SUV body weight mean (SUVmean), SUV lean body mass (SUVlbm), SUV body surface area (SUVbsa), metabolic tumor volume (MTV), total lesion glycolysis (TLG), ratio SUVmax/blood-pool (S-BP), and ratio SUVmax/liver (S-L) were also extracted. The relationship between these parameters, BRCA, and other clinicopathological features were evaluated. Kaplan-Meier, univariate, and multivariate analyses were performed to find independent prognosticators for progression free (PFS) and overall survival (OS). RESULTS: Significant positive correlations between BRCA status and SUVmax (p-value 0.025), SUVlbm (p-value 0.016), and SUVbsa (p-value 0.018) were reported. Mean PFS was 53.90 months with relapse/progression of disease occurring in nine (22.0%) patients; mean OS was 57.48 months with death occurring in two (4.9%) patients. Survival curves revealed TLG, MTV, and BRCA status as prognosticator for PFS; BRCA was also a prognosticator for OS. Univariate and multivariate analyses did not confirm such insights. CONCLUSION: We reported a correlation between some PET/CT parameters and BRCA status; some insights on their prognostic role have been underlined.

Correlation between brain glucose metabolism (18F-FDG) and cerebral blood flow with amyloid tracers (18F-Florbetapir) in clinical routine: Preliminary evidences.

OBJECTIVE: This study compared the performance of 18F-Florbetapir PET/CT early acquisitions to 18F-FDG PET/CT. METHODS: We included 12 patients who underwent 18F-FDG PET/CT and a dual-time 18F-Florbetapir PET/CT (1-6 min early-scan and 50 min late-scan). PET/CT were analyzed visually by three nuclear medicine physicians with different experience using a four-point scale (0 = no reduction, 1 = slight, 2 = moderate, 3 = severe reduction) for 18F-Florbetapir early-phase and 18F-FDG images in 10 cortical regions (bilateral frontal, temporal, parietal, occipital, posterior cingulate/precuneus), and 18F-Florbetapir late-phase in the same cortical regions using a three-point scale (0 = normal, 1 = abnormal with minor plaques, 2 = abnormal with major plaques). We used SPM12 for semiquantitative analysis applying a ROI-based correlation analysis (considering precuneus as target region and normalized for the mean global binding), a covariance-analysis taking precuneus as target and a comparison of global DMN (default mode network). RESULTS: Inter-reader agreement was high (Cohen's kappa 0.762 for 18F-FDG, 0.775 for 18F-Florbetapir early-phase and 0.794 for late-phase). Regional visual scores of early-phase and 18F-FDG were significantly correlated (ρ = 0.867). Also ROI-based analysis, global brain visual analysis and DMN comparison revealed concordant results, especially at parietal and precuneus (p < 0.001). CONCLUSIONS: 18F-Florbetapir early-phase scans significantly correlate on quantitative and visual images with 18F-FDG-PET/CT scans, suggesting that amyloid tracer could be instead of 18F-FDG.

Role of Radiomics Features and Machine Learning for the Histological Classification of Stage I and Stage II NSCLC at [18F]FDG PET/CT: A Comparison between Two PET/CT Scanners.

The aim of this study was to compare two different PET/CT tomographs for the evaluation of the role of radiomics features (RaF) and machine learning (ML) in the prediction of the histological classification of stage I and II non-small-cell lung cancer (NSCLC) at baseline [18F]FDG PET/CT. A total of 227 patients were retrospectively included and, after volumetric segmentation, RaF were extracted. All of the features were tested for significant differences between the two scanners and considering both the scanners together, and their performances in predicting the histology of NSCLC were analyzed by testing of different ML approaches: Logistic Regressor (LR), k-Nearest Neighbors (kNN), Decision Tree (DT) and Random Forest (RF). In general, the models with best performances for all the scanners were kNN and LR and moreover the kNN model had better performances compared to the other. The impact of the PET/CT scanner used for the acquisition of the scans on the performances of RaF was evident: mean area under the curve (AUC) values for scanner 2 were lower compared to scanner 1 and both the scanner considered together. In conclusion, our study enabled the selection of some [18F]FDG PET/CT RaF and ML models that are able to predict with good performances the histological subtype of NSCLC. Furthermore, the type of PET/CT scanner may influence these performances.

Clinical and prognostic 18F-FDG PET/CT role in recurrent vulvar cancer: a multicentric experience.

PURPOSE: The aim of this retrospective multicentric study was to investigate the diagnostic performance, the prognostic value and the impact of 18F-FDG PET/CT on treatment decision-making in patients with suspected recurrent vulvar cancer (VC). MATERIALS AND METHODS: Sixty-three patients affected by VC performed 18F-FDG-PET/CT for restaging purposes in case of suspected clinical and/or radiological recurrence. Histopatology results if available and/or clinical-imaging follow-up for at least 12 months were considered as reference standard. The diagnostic accuracy and the clinical impact of 18F-FDG PET/CT were investigated. Progression free survival (PFS) and overall survival (OS) were calculated using Kaplan-Meier curves. RESULTS: Fifty-two (82.5%) PET/CT showed the presence of recurrence, while the remaining 11 (17.5%) were negative. Sensitivity, specificity, positive predictive value, negative predictive value and accuracy of PET/CT were 100% (95%CI 93-100%), 92% (95%CI 62-100%), 98% (95%CI 89-99%), 100% and 98% (95%CI 92-100%). A relevant impact of 18F-FDG PET/CT imaging was registered in 28 cases: in 12 cases moving from local therapy to chemotherapy due to the recognition of disseminate localizations; in 10 showing the site of recurrence in presence of negative conventional imaging, and in 6 cases confirming to be true negative and avoiding unnecessary therapies. Beside advanced age and HPV status, a positive restaging 18F-FDG PET/CT scan was significantly correlated with shorter PFS and OS compared to negative scan (p < 0.001). CONCLUSIONS: 18F-FDG PET/CT demonstrated to be an accurate tool in the assessing of recurrent VC with high sensitivity and specificity and with a significant impact on clinical decision-making. Restaging 18F-FDG PET/CT findings were associated with survival.

Role of 18F-FDG PET/CT Radiomics Features in the Differential Diagnosis of Solitary Pulmonary Nodules: Diagnostic Accuracy and Comparison between Two Different PET/CT Scanners.

The aim of this retrospective study was to investigate the ability of 18 fluorine-fluorodeoxyglucose positron emission tomography/CT (18F-FDG-PET/CT) metrics and radiomics features (RFs) in predicting the final diagnosis of solitary pulmonary nodules (SPN). We retrospectively recruited 202 patients who underwent a 18F-FDG-PET/CT before any treatment in two PET scanners. After volumetric segmentation of each lung nodule, 8 PET metrics and 42 RFs were extracted. All the features were tested for significant differences between the two PET scanners. The performances of all features in predicting the nature of SPN were analyzed by testing three classes of final logistic regression predictive models: two were built/trained through exploiting the separate data from the two scanners, and the other joined the data together. One hundred and twenty-seven patients had a final diagnosis of malignancy, while 64 were of a benign nature. Comparing the two PET scanners, we found that all metabolic features and most of RFs were significantly different, despite the cross correlation being quite similar. For scanner 1, a combination between grey level co-occurrence matrix (GLCM), histogram, and grey-level zone length matrix (GLZLM) related features presented the best performances to predict the diagnosis; for scanner 2, it was GLCM and histogram-related features and metabolic tumour volume (MTV); and for scanner 1 + 2, it was histogram features, standardized uptake value (SUV) metrics, and MTV. RFs had a significant role in predicting the diagnosis of SPN, but their accuracies were directly related to the scanner.

Comparison between skeletal muscle and adipose tissue measurements with high-dose CT and low-dose attenuation correction CT of 18F-FDG PET/CT in elderly Hodgkin lymphoma patients: a two-centre validation.

OBJECTIVES: High-dose CT (HDCT) is considered the gold-standard imaging for the measurements of skeletal muscle area (SMA), visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT) and intramuscular adipose tissue (IMAT) areas in the abdomen. These parameters may reflect sarcopenia, which can have a prognostic impact in several oncological diseases. The aim of this study was to compare the agreement of measurements of SMA, VAT, SAT and IMAT areas between HDCT and low-dose CT (LDCT) of 18-fludeoxyglucose positron emission tomography (18F-FDG PET)/CT in elderly patients affected by Hodgkin lymphoma (HL). METHODS: We retrospectively included 90 patients affected by HL who underwent baseline 18F-FDG-PET/CT and HDCT within a mean interval of 7 days. HDCT and LDCT images were analysed by two blinded observers using segmentation software (Slice-O-Matic, Tomovision) to quantify the areas. HDCT and LDCT measurements were compared using Bland-Altman plots and Passing-Bablock regression analyses. Pearson correlation coefficient (r) was used to correlate measurements from the two imaging modalities. RESULTS: Comparison of HDCT and LDCT data demonstrated a strong correlation for measurement of VAT(r = 0.942, p < 0.0001), SAT (r = 0.894, p < 0.0001) and SMA (r = 0.934, p < 0.0001). Instead considering IMAT, correlation was good but less significant (r = 0.742). The mean difference between the two methods was found to be very small, with a difference of 1% for SAT,+6.1% for VAT,+2.5% for SMA and -1.9% for IMAT. CONCLUSION: LDCT of PET/CT is a safe, accurate and precise method for the measurements of skeletal muscle area, visceral and subcutaneous adipose tissue. Their measurements are reproducible and correlate closely with HDCT. ADVANCES IN KNOWLEDGE: LLDCT of PET/CT is a safe and accurate method for the measurements of SMA, VAT and SAT; their measurements are closely correlated with HDCT. LDCT can be considered an accurate alternative tool for measuring abdominal fat and muscles in the clinical practice.

2-[18F]-FDG PET/CT Role in Detecting Richter Transformation of Chronic Lymphocytic Leukemia and Predicting Overall Survival.

INTRODUCTION: Chronic lymphocytic leukemia (CLL) is an indolent, low-grade B-cell lymphoproliferative disorder, which may evolve into aggressive lymphoma, a phenomenon called Richter syndrome (RS). Our aim was to study the accuracy of 2-deoxy-2-[18F]fluoro-D-glucose positron emission tomography/computed tomography (2-[18F]-FDG PET/CT) and its semiquantitative parameters for the detection of RS and the impact on overall survival (OS). MATERIALS AND METHODS: Eighty patients with histologically proven CLL were retrospectively included. PET/CT images were qualitatively and semiquantitatively examined by estimating the maximum standardized uptake value body weight (SUVbw), lean body mass (SUVlbm), body surface area (SUVbsa), lesion-to-blood-pool SUV ratio (L-BP SUV R), lesion-to-liver SUV ratio (L-L SUV R), total lesion glycolysis (TLG), and metabolic tumor volume (MTV) and comparing them with the main clinical-histologic variables. OS curves were plotted according to the Kaplan-Meier method. RESULTS: Seventy-eight patients had positive 2-[18F]-FDG PET/CT, whereas the remaining 2 were not FDG-avid. All PET/CT metabolic parameters were significantly higher in the RS group compared with the no-RS group, except for MTV and TLG. The best thresholds identified were 9 for SUVbw, 5.3 for SUVlbm, 1.7 for SUVbsa, 2 for L-L SUV R, and 4.8 for L-BP SUV R. After a median follow-up of 32 months, 24 patients died; OS was significantly shorter in patients with RS than patients without RS (16.5 vs. 27.8 months; P = .001). Binet-stage, B symptoms, SUVbw, SUVlbm, SUVbsa, L-L SUV R, and L-BP SUV R were shown to be independent prognostic features. CONCLUSIONS: Semiquantitative PET/CT parameters that are SUV-related may be useful in discriminating patients with a high risk of developing RS and also for predicting OS.

Comparison between N13NH3-PET and 99mTc-Tetrofosmin-CZT SPECT in the evaluation of absolute myocardial blood flow and flow reserve.

BACKGROUND: PET/CT is the standard for quantitative assessments of myocardial blood flow (MBF), but it requires short-lived-tracers, costly, and not widely available. SPECT with Cadmium Zinc Telluride (CZT) detectors allows dynamic acquisition and quantitation of MBF. The study aims were to compare MBF measurements by 99mTc-tetrofosmin-CZT to N13NH3 PET/CT after regadenoson-induced coronary hyperemia and to evaluate the effect of attenuation correction (AC). METHODS: 54 patients were evaluated at rest and during vasodilation by 99mTc-tetrofosmin-CZT and N13NH3 PET/CT within 2 weeks. MBF and MBF reserve (MFR) were measured by CZT with or without AC (NAC). RESULTS: The global rest MBF was 0.76 ± 0.19 mL/min/gr by PET and 0.76 ± 0.24 by AC-CZT (P = NS) and 1.14 ± 0.4 by NAC-CZT (P < 0.001 vs PET and AC-CZT). Stress MBF was higher when measured by PET than AC-CZT (1.87 ± 0.45 vs 1.62 ± 0.68 mL/min/gr, P < 0.0008), but lower than NAC-CZT (2.36 ± 1.1, P < 0.0003). The MBF reserve ratio (MFR) was higher by PET than AC-CZT (2.52 ± 0.56 vs 2.22 ± 1 (P < 0.009) and NAC-CZT (2.18 ± 1.0, P < 0.004). Linear regression was better between PET (MFR and stress MBF) and AC-CZT than between PET and NAC-CZT. ROC curve analysis showed the significant ability of AC-CZT to predict MFR < 2 and stress MBF < 1.7 (AUC = 0.75 and 0.82 respectively) and to differentiate between normal and CAD patients (AUC = 0.747 and 0.892 for MFR and stress MBF, respectively). CONCLUSIONS: Our data show a reasonable correlation between MBF and MFR measured by N13NH3-PET and 99mTc-Tetrofosmin-CZT SPECT. NAC-CZT overestimates MBF. AC is recommended when using CZT for measuring MBF.

Performing an Additional Lateral Decubitus PET/CT Scan to Resolve a Respiratory Motion Artifact.

Respiratory motion artifacts may affect interpretation of whole-body 18F-FDG PET/CT scans, especially when lesions are localized between the liver and the lung. We report a case of a patient with breast cancer who underwent PET/CT after therapy and in whom focal 18F-FDG uptake of equivocal interpretation was observed between the liver and the pleura. A subsequent imaging acquisition of the right lateral decubitus showed that the lesion had a pleural location, thus improving the diagnostic accuracy of the PET/CT finding.

Prognostic Value of 18F-FDG PET/CT Metabolic Parameters in Splenic Marginal Zone Lymphoma.

INTRODUCTION: Splenic marginal zone lymphoma (SMZL) is an indolent non-Hodgkin lymphoma usually with a good prognosis, but no clear metabolic behavior at fluorine-18-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT). The aim of our analysis was to investigate the prognostic role of baseline 18F-FDG PET/CT parameters in SMZL. MATERIALS AND METHODS: We retrospectively included 42 patients who received 18F-FDG-PET/CT before any treatments, and PET images were evaluated visually and semi-quantitatively by measuring lesion to liver (L-L) maximum standardized uptake volume (SUVmax) ratio (L-L SUV R), lesion to blood-pool SUVmax ratio (L-BP SUV R), metabolic tumor volume, and total lesion glycolysis. Progression-free (PFS) and overall survival (OS) curves were plotted according to the Kaplan-Meier method. RESULTS: In all patients, an increased splenic FDG uptake (higher than the background) was identified, showing the presence of diffuse spleen uptake in 35 patients and focal uptake in the remaining 7 patients. At a median follow-up of 51 months, relapse or progression of disease occurred in 23 patients with an average time of 38.1 months from the baseline 18F-FDG PET/CT, and death occurred in 4 patients with an average time of 26.8 months. The estimated 2-year PFS and OS rates were 78% and 90%, respectively, whereas 5-year PFS and OS rates were 63% and 82%, respectively. At multivariate analysis, only L-L SUV R and L-BP SUV R were independent prognostic factors for PFS. In addition, no significant association was discovered for OS, considering all features. CONCLUSIONS: L-L SUV R and L-BP SUV R were independently correlated with PFS.

18F-FDG PET/CT Metabolic Behavior of COVID-19 Pneumonia: A Series of 4 Patients With RT-PCR Confirmation.

Between March 26 and April 6, among 80 patients who underwent F-FDG PET/CT in our department (Brescia, Italy), 4 showed the presence of an interstitial pneumonia suspected for COVID-19 with reverse transcriptase polymerase chain reaction confirmation. All patients except one had bilateral ground-glass opacities and/or lung consolidations in at least 2 pulmonary lobes. Inferior lobes and basal segments were the most frequent site of disease. All lung lesions had an increased FDG uptake corresponding to the interstitial pneumonia, and in one case, mediastinal nodal involvement was registered.