Comparison of MRI, [18F]FDG PET/CT, and 99mTc-UBI 29-41 scintigraphy for postoperative spondylodiscitis-a prospective multicenter study.

PURPOSE: Postoperative infection still constitutes an important complication of spine surgery, and the optimal imaging modality for diagnosing postoperative spine infection has not yet been established. The aim of this prospective multicenter study was to assess the diagnostic performance of three imaging modalities in patients with suspected postoperative spine infection: MRI, [18F]FDG PET/CT, and SPECT/CT with 99mTc-UBI 29-41. METHODS: Patients had to undergo at least 2 out of the 3 imaging modalities investigated. Sixty-three patients enrolled fulfilled such criteria and were included in the final analysis: 15 patients underwent all 3 imaging modalities, while 48 patients underwent at least 2 imaging modalities (MRI + PET/CT, MRI + SPECT/CT, or PET/CT + SPECT/CT). Final diagnosis of postoperative spinal infection was based either on biopsy or on follow-up for at least 6 months. The MRI, PET/CT, and SPECT/CT scans were read blindly by experts at designated core laboratories. Spine surgery included metallic implants in 46/63 patients (73%); postoperative spine infection was diagnosed in 30/63 patients (48%). RESULTS: Significant discriminants between infection and no infection included fever (P = 0.041), discharge at the wound site (P < 0.0001), and elevated CRP (P = 0.042). There was no difference in the frequency of infection between patients who underwent surgery involving spinal implants versus those who did not. The diagnostic performances of MRI and [18F]FDG PET/CT analyzed as independent groups were equivalent, with values of the area under the ROC curve equal to 0.78 (95% CI: 0.64-0.92) and 0.80 (95% CI: 0.64-0.98), respectively. SPECT/CT with 99mTc-UBI 29-41 yielded either unacceptably low sensitivity (44%) or unacceptably low specificity (41%) when adopting more or less stringent interpretation criteria. The best diagnostic performance was observed when combining the results of MRI with those of [18F]FDG PET/CT, with an area under the ROC curve equal to 0.938 (95% CI: 0.80-1.00). CONCLUSION: [18F]FDG PET/CT and MRI both possess equally satisfactory diagnostic performance in patients with suspected postoperative spine infection, the best diagnostic performance being obtained by combining MRI with [18F]FDG PET/CT. The diagnostic performance of SPECT/CT with 99mTc-UBI 29-41 was suboptimal in the postoperative clinical setting explored with the present study.

SPECT-MPI for Coronary Artery Disease: A Deep Learning Approach.

Background: Worldwide, coronary artery disease (CAD) is a leading cause of mortality and morbidity and remains to be a top health priority in many countries. A non-invasive imaging modality for diagnosis of CAD such as single photon emission computed tomography-myocardial perfusion imaging (SPECT-MPI) is usually requested by cardiologists as it displays radiotracer distribution in the heart reflecting myocardial perfusion. The interpretation of SPECT-MPI is done visually by a nuclear medicine physician and is largely dependent on his clinical experience and showing significant inter-observer variability. Objective: The aim of the study is to apply a deep learning approach in the classification of SPECT-MPI for perfusion abnormalities using convolutional neural networks (CNN). Methods: A publicly available anonymized SPECT-MPI from a machine learning repository (https://www.kaggle.com/selcankaplan/spect-mpi) was used in this study involving 192 patients who underwent stress-test-rest Tc99m MPI. An exploratory approach of CNN hyperparameter selection to search for optimum neural network model was utilized with particular focus on various dropouts (0.2, 0.5, 0.7), batch sizes (8, 16, 32, 64), and number of dense nodes (32, 64, 128, 256). The base CNN model was also compared with the commonly used pre-trained CNNs in medical images such as VGG16, InceptionV3, DenseNet121 and ResNet50. All simulations experiments were performed in Kaggle using TensorFlow 2.6.0., Keras 2.6.0, and Python language 3.7.10. Results: The best performing base CNN model with parameters consisting of 0.7 dropout, batch size 8, and 32 dense nodes generated the highest normalized Matthews Correlation Coefficient at 0.909 and obtained 93.75% accuracy, 96.00% sensitivity, 96.00% precision, and 96.00% F1-score. It also obtained higher classification performance as compared to the pre-trained architectures. Conclusions: The results suggest that deep learning approaches through the use of CNN models can be deployed by nuclear medicine physicians in their clinical practice to further augment their decision skills in the interpretation of SPECT-MPI tests. These CNN models can also be used as a dependable and valid second opinion that can aid physicians as a decision-support tool as well as serve as teaching or learning materials for the less-experienced physicians particularly those still in their training career. These highlights the clinical utility of deep learning approaches through CNN models in the practice of nuclear cardiology.

Validation of Convolutional Neural Networks for Fast Determination of Whole-Body Metabolic Tumor Burden in Pediatric Lymphoma.

18F-FDG PET/CT quantification of whole-body tumor burden in lymphoma is not routinely performed because of the lack of fast methods. Although the semiautomatic method is fast, it is not fast enough to quantify tumor burden in daily clinical practice. Our purpose was to evaluate the performance of convolutional neural network (CNN) software in localizing neoplastic lesions in whole-body 18F-FDG PET/CT images of pediatric lymphoma patients. Methods: The retrospective image dataset, derived from the data pool of the International Atomic Energy Agency (coordinated research project E12017), included 102 baseline staging 18F-FDG PET/CT studies of pediatric lymphoma patients (mean age, 11 y). The images were quantified to determine the whole-body tumor burden (whole-body metabolic tumor volume [wbMTV] and whole-body total lesion glycolysis [wbTLG]) using semiautomatic software and CNN-based software. Both were displayed as semiautomatic wbMTV and wbTLG and as CNN wbMTV and wbTLG. The intraclass correlation coefficient (ICC) was applied to evaluate concordance between the CNN-based software and the semiautomatic software. Results: Twenty-six patients were excluded from the analysis because the software was unable to perform calculations for them. In the remaining 76 patients, CNN and semiautomatic wbMTV tumor burden metrics correlated strongly (ICC, 0.993; 95% CI, 0.989 - 0.996; P < 0.0001), as did CNN and semiautomatic wbTLG (ICC, 0.999; 95% CI, 0.998-0.999; P < 0.0001). However, the time spent calculating these metrics was significantly (<0.0001) less by CNN (mean, 19 s; range, 11-50 s) than by the semiautomatic method (mean, 21.6 min; range, 3.2-62.1 min), especially in patients with advanced disease. Conclusion: Determining whole-body tumor burden in pediatric lymphoma patients using CNN is fast and feasible in clinical practice.

Worldwide Diagnostic Reference Levels for Single-Photon Emission Computed Tomography Myocardial Perfusion Imaging: Findings From INCAPS.

OBJECTIVES: This study sought to establish worldwide and regional diagnostic reference levels (DRLs) and achievable administered activities (AAAs) for single-photon emission computed tomography (SPECT) myocardial perfusion imaging (MPI). BACKGROUND: Reference levels serve as radiation dose benchmarks to compare individual laboratories against aggregated data, helping to identify sites in greatest need of dose reduction interventions. DRLs for SPECT MPI have previously been derived from national or regional registries. To date there have been no multiregional reports of DRLs for SPECT MPI from a single standardized dataset. METHODS: Data were submitted voluntarily to the INCAPS (International Atomic Energy Agency Nuclear Cardiology Protocols Study), a cross-sectional, multinational registry of MPI protocols. A total of 7,103 studies were included. DRLs and AAAs were calculated by protocol for each world region and for aggregated worldwide data. RESULTS: The aggregated worldwide DRLs for rest-stress or stress-rest studies employing technetium Tc 99m-labeled radiopharmaceuticals were 11.2 mCi (first dose) and 32.0 mCi (second dose) for 1-day protocols, and 23.0 mCi (first dose) and 24.0 mCi (second dose) for multiday protocols. Corresponding AAAs were 10.1 mCi (first dose) and 28.0 mCi (second dose) for 1-day protocols, and 17.8 mCi (first dose) and 18.7 mCi (second dose) for multiday protocols. For stress-only technetium Tc 99m studies, the worldwide DRL and AAA were 18.0 mCi and 12.5 mCi, respectively. Stress-first imaging was used in 26% to 92% of regional studies except in North America where it was used in just 7% of cases. Significant differences in DRLs and AAAs were observed between regions. CONCLUSIONS: This study reports reference levels for SPECT MPI for each major world region from one of the largest international registries of clinical MPI studies. Regional DRLs may be useful in establishing or revising guidelines or simply comparing individual laboratory protocols to regional trends. Organizations should continue to focus on establishing standardized reporting methods to improve the validity and comparability of regional DRLs.

Results of a Prospective Multicenter International Atomic Energy Agency Sentinel Node Trial on the Value of SPECT/CT Over Planar Imaging in Various Malignancies.

UNLABELLED: We aimed to assess the additional value of SPECT/CT over planar lymphoscintigraphy (PI) in sentinel node (SN) detection in malignancies with different lymphatic drainage such as breast cancer, melanoma, and pelvic tumors. METHODS: From 2010 to 2013, 1,508 patients were recruited in a multicenter study: 1,182 breast cancer, 262 melanoma, and 64 pelvic malignancies (prostate, cervix, penis, vulva). PI was followed by SPECT/CT 1-3 h after injection of (99m)Tc-colloid particles. Surgery was performed the same or next day. RESULTS: Significantly more SNs were detected by SPECT/CT for breast cancer (2,165 vs. 1,892), melanoma (602 vs. 532), and pelvic cancer (195 vs. 138), all P < 0.001. The drainage basin mismatch between PI and SPECT/CT was 16.5% for breast cancer, 11.1% for melanoma, and 51.6% for pelvic cancers. Surgical adjustment was 17% for breast cancer, 37% for melanoma, and 65.6% for pelvic cancer. CONCLUSION: SPECT/CT detected more SNs and changed the drainage territory, leading to surgical adjustments in a considerable number of patients in all malignancies studied but especially in the pelvic cancer group because of this group's deep lymphatic drainage. We recommend SPECT/CT in all breast cancer patients with no SN visualized on PI, all patients with melanoma of the head and neck or trunk, all patients with pelvic malignancies, and those breast cancer and melanoma patients with unexpected drainage on PI.