An important need exists for reliable PET tumor-segmentation methods for tasks such as PET-based radiation-therapy planning and reliable quantification of volumetric and radiomic features. The purpose of this study was to develop an automated physics-guided deeplearning-based PET tumor-segmentation framework that addresses the challenges of limited spatial resolution, high image noise, and lack of clinical training data with ground-truth tumor boundaries in PET imaging. We propose a three-module PET-segmentation framework in the context of segmenting primary tumors in 3D 18 F-fluorodeoxyglucose (FDG)-PET images of patients with lung cancer on a per-slice basis. 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'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% 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 cm 2 ), 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)). A modular deep-learning-based framework yielded reliable automated tumor delineation in FDG-PET images of patients with lung cancer using a small-sized clinical training dataset, generalized across scanners, and demonstrated ability to segment small tumors.

The observed close correlation between a longitudinal MBF gradient during hyperaemic flows and invasively measured FFR suggests the longitudinal flow gradient as an emerging non-invasive index of flow-limiting CAD.

18F-fludeoxyglucose-positron emission tomography/computed tomography (FDG-PET/CT) has been performed widely in diagnosis and management of patients with oropharyngeal squamous cell carcinoma (OPSCC). This review summarizes the literature on this tool in the management of these patients. The use of FDG-PET/CT helps in accurate staging of primary tumor, nodal involvement, and distant metastasis of patients with OPSCC. Contrast-enhanced FDG-PET/CT combines high-resolution CT and functional FDG-PET, providing the optimum imaging information for patient management. Using contrast-enhanced PET/CT leads to a combined anatomic and metabolic approach to radiation therapy planning in OPSCC. Moreover, PET/CT not only is a good modality for therapy assessment but also is a powerful tool in early recurrence detection of OPSCC. Finally, the PET/CT parameters provide survival information in patients with OPSCC; however, further studies are needed to introduce a scoring system to use clinically for prognosis prediction.

SummaryIntroduction Diagnostic I-123 scans have been shown to underestimate the disease burden in differentiated thyroid cancer (DTC) when compared to I-131 post-treatment scans, especially in children and patients who have had prior radioiodine (RAI) therapy and/or distant metastasis. I-124 PET/CT has been shown to be highly effective in imaging DTC-related metastatic disease. Methods We performed a systematic review and meta-analysis of studies investigating the sensitivity and specificity of I-124 PET/CT in identifying lesions amenable to RAI therapy as confirmed by I-131 post-treatment scanning. Results There were 141 patients and 415 lesions of DTC identified altogether. There was significant heterogeneity in the individual studies. The pooled sensitivity of the I-124 PET/CT in detecting lesions of differentiated thyroid cancer amenable to I-131 therapy was 94Á2% (91Á3-96Á4% CI, P < 0Á01), and the pooled specificity was 49Á0% (34Á8-63Á4% CI, P < 0Á01). The pooled positive likelihood ratio (LR) was 1Á43 (1Á05-1Á94 CI), and the pooled negative LR was 0Á28 (0Á15-0Á53 CI). Overall, the diagnostic odds ratio was 7Á90 (3Á39-18Á48 CI). There were a small but increased number of lesions identified by I-124 PET/CT that was not detected on post-treatment scan. Conclusion I-124 PET/CT is a sensitive tool to diagnose RAI avid DTC lesions, but also detects some new lesions that are not visualized on the post-treatment I-131 scan. Further, carefully designed dosimetric studies may be required to fully establish the role of I-124 PET CT for identifying potential lesions for I-131 therapy. I-124 PET/CT in patients with DTC may have other applications in specific clinical situations.