Information fusion for fully automated segmentation of head and neck tumors from PET and CT images.

BACKGROUND: PET/CT images combining anatomic and metabolic data provide complementary information that can improve clinical task performance. PET image segmentation algorithms exploiting the multi-modal information available are still lacking. PURPOSE: Our study aimed to assess the performance of PET and CT image fusion for gross tumor volume (GTV) segmentations of head and neck cancers (HNCs) utilizing conventional, deep learning (DL), and output-level voting-based fusions. METHODS: The current study is based on a total of 328 histologically confirmed HNCs from six different centers. The images were automatically cropped to a 200 × 200 head and neck region box, and CT and PET images were normalized for further processing. Eighteen conventional image-level fusions were implemented. In addition, a modified U2-Net architecture as DL fusion model baseline was used. Three different input, layer, and decision-level information fusions were used. Simultaneous truth and performance level estimation (STAPLE) and majority voting to merge different segmentation outputs (from PET and image-level and network-level fusions), that is, output-level information fusion (voting-based fusions) were employed. Different networks were trained in a 2D manner with a batch size of 64. Twenty percent of the dataset with stratification concerning the centers (20% in each center) were used for final result reporting. Different standard segmentation metrics and conventional PET metrics, such as SUV, were calculated. RESULTS: In single modalities, PET had a reasonable performance with a Dice score of 0.77 ± 0.09, while CT did not perform acceptably and reached a Dice score of only 0.38 ± 0.22. Conventional fusion algorithms obtained a Dice score range of [0.76-0.81] with guided-filter-based context enhancement (GFCE) at the low-end, and anisotropic diffusion and Karhunen-Loeve transform fusion (ADF), multi-resolution singular value decomposition (MSVD), and multi-level image decomposition based on latent low-rank representation (MDLatLRR) at the high-end. All DL fusion models achieved Dice scores of 0.80. Output-level voting-based models outperformed all other models, achieving superior results with a Dice score of 0.84 for Majority_ImgFus, Majority_All, and Majority_Fast. A mean error of almost zero was achieved for all fusions using SUVpeak , SUVmean and SUVmedian . CONCLUSION: PET/CT information fusion adds significant value to segmentation tasks, considerably outperforming PET-only and CT-only methods. In addition, both conventional image-level and DL fusions achieve competitive results. Meanwhile, output-level voting-based fusion using majority voting of several algorithms results in statistically significant improvements in the segmentation of HNC.

Unnecessary thyroid surgery rate for suspicious nodule in the absence of molecular testing.

Background: Molecular tests for suspicious thyroid nodules decrease rates of unnecessary surgeries but are not widely used due to reimbursement issues. The aim of this study was to assess the rate of unnecessary surgery performed in real-life setting for Bethesda III, IV and V nodules in the absence of molecular testing. Method: This is a single-center retrospective study of consecutive patients undergoing fine needle aspiration cytology (FNAC) with rapid on-site evaluation between January 2017 and December 2021. Unnecessary surgery was defined as surgery performed because of Bethesda III, IV, or V results in the absence of local compressive symptoms with final benign pathology and as second surgery for completion thyroidectomy. Results: In the 862 patients (640 females, mean age: 54.2 years), 1010 nodules (median size: 24.4 mm) underwent 1189 FNAC. Nodules were EU-TIRADS 2, 3, 4, and 5 in 3%, 34%, 42%, and 22% of cases, respectively. FNAC was Bethesda I, II, III, IV, V, and VI in 8%, 48%, 17%, 17%, 3%, and 6%, respectively. Surgery was performed in 36% of Bethesda III nodules (benign on pathology: 81%), in 74% of Bethesda IV nodules (benign on pathology: 76%) and in 97% of Bethesda V nodules (benign on pathology: 21%). Surgery was considered unnecessary in 56%, 68%, and 21% of patients with Bethesda III, IV, and V nodules, respectively. Conclusion: In this real data cohort surgery was unnecessary in more than half of patients with Bethesda III and IV nodules and in 21% of patients with Bethesda V nodules.

How to prepare for a bright future of radiology in Europe.

Because artificial intelligence (AI)-powered algorithms allow automated image analysis in a growing number of diagnostic scenarios, some healthcare stakeholders have raised doubts about the future of the entire radiologic profession. Their view disregards not only the role of radiologists in the diagnostic service chain beyond reporting, but also the many multidisciplinary and patient-related consulting tasks for which radiologists are solicited. The time commitment for these non-reporting tasks is considerable but difficult to quantify and often impossible to fulfil considering the current mismatch between workload and workforce in many countries. Nonetheless, multidisciplinary, and patient-centred consulting activities could move up on radiologists' agendas as soon as AI-based tools can save time in daily routine. Although there are many reasons why AI will assist and not replace radiologists as imaging experts in the future, it is important to position the next generation of European radiologists in view of this expected trend. To ensure radiologists' personal professional recognition and fulfilment in multidisciplinary environments, the focus of training should go beyond diagnostic reporting, concentrating on clinical backgrounds, specific communication skills with referrers and patients, and integration of imaging findings with those of other disciplines. Close collaboration between the European Society of Radiology (ESR) and European national radiologic societies can help to achieve these goals. Although each adequate treatment begins with a correct diagnosis, many health politicians see radiologic procedures mainly as a cost factor. Radiologic research should, therefore, increasingly investigate the imaging impact on treatment and outcome rather than focusing mainly on technical improvements and diagnostic accuracy alone.Critical relevance statement Strategies are presented to prepare for a successful future of the radiologic profession in Europe, if AI-powered tools can alleviate the current reporting overload: engaging in multidisciplinary activities (clinical and integrative diagnostics), enhancing the value and recognition of radiologists' role through clinical expertise, focusing radiological research on the impact on diagnosis and outcome, and promoting patient-centred radiology by enhancing communication skills.Key points • AI-powered tools will not replace radiologists but hold promise to reduce the current reporting burden, enabling them to reinvest liberated time in multidisciplinary clinical and patient-related tasks.• The skills and resources for these tasks should be considered when recruiting and teaching the next generation of radiologists, when organising departments and planning staffing.• Communication skills will play an increasing role in both multidisciplinary activities and patient-centred radiology.• The value and importance of a correct and integrative diagnosis and the cost of an incorrect imaging diagnosis should be emphasised when discussing with non-medical stakeholders in healthcare.• The radiologic community in Europe should start now to prepare for a bright future of the profession for the benefit of patients and medical colleagues alike.

PET/MR Imaging in Head and Neck Cancer.

Head and neck squamous cell carcinoma (HNSCC) can either be examined with hybrid PET/MR imaging systems or sequentially, using PET/CT and MR imaging. Regardless of the acquisition technique, the superiority of MR imaging compared to CT lies in its potential to interrogate tumor and surrounding tissues with different sequences, including perfusion and diffusion. For this reason, PET/MR imaging is preferable for the detection and assessment of locoregional residual/recurrent HNSCC after therapy. In addition, MR imaging interpretation is facilitated when combined with PET. Nevertheless, distant metastases and distant second primary tumors are detected equally well with PET/MR imaging and PET/CT.

Does the type of craniomaxillofacial fracture (CMF) differ between patients with intracranial hemorrhage (ICH) and those with blunt cerebrovascular injury (BCVI)? A retrospective study.

The purpose of the present study was to investigate and compare craniomaxillofacial fracture (CMF) type in patients with intracranial hemorrhage (ICH) versus blunt cerebrovascular injury (BCVI). A retrospective cohort study was performed. The predictor variables were the types of CMF. The primary outcomes variables were ICH and BCVI. Secondary outcomes were death and survival with or without neurological sequelae. Descriptive, bivariate, and multiple logistic regression statistics were computed, and the significance level was set at P ≤ 0.05. The sample was composed of 1440 patients with a mean age of 46.6 years ±24 years, and 71% were men. Pure orbital wall (odds ratio [OR]), 3.62; 95% confidence interval [CI], 1.32-12.69; P < 0.022), Le Fort III (OR, 16.08; 95% CI, 5.89-43.50; P < 0.001), cranial vault (OR, 9.74; 95% CI, 3.83.24.32; P < 0.001), skull base (OR, 9.42; 95% CI, 3.86-24.02; P < 0.001) and cervical fractures (OR, 5.50; 95% CI, 1.65-15.97; P = 0.003) were significantly associated with BCVI. All of the CMFs (P < 0.001), except for Le Fort I (OR, 0.79; 95% CI, 0.18-2.63; P = 0.731), nasal (OR, 1.05; 95% CI, 0.77-1.42; P = 0.758), and mandibular (OR, 0.68; 95% CI, 0.45-1.01; P = 0.066) fractures, were significantly associated with ICH. Secondary outcomes were negatively influenced by ICH and BCVI (P < 0.001). Within the limitations of the study it seems that Le Fort I and nasal fractures could be protective of cerebrovascular injuries, by cushioning impact forces. On the other hand it seems that patients with pure orbital wall, Le Fort III and cranio-cervical fractures are more prone to having concomitant life-threatening cerebrovascular injuries. This category of patients should have an immediate and comprehensive neurological assessment and CT angiography to rule out BCVI and to determine its severity.

Clinical audit in European radiology: current status and recommendations for improvement endorsed by the European Society of Radiology (ESR).

Clinical audit is an important quality improvement activity and has significant benefits for patients in terms of enhanced care, safety, experience and outcomes. Clinical audit in support of radiation protection is mandated within the European Council Basic Safety Standards Directive (BSSD), 2013/59/Euratom. The European Society of Radiology (ESR) has recognised clinical audit as an area of particular importance in the delivery of safe and effective health care. The ESR, alongside other European organisations and professional bodies, has developed a range of clinical audit-related initiatives to support European radiology departments in developing a clinical audit infrastructure and fulfilling their legal obligations. However, work by the European Commission, the ESR and other agencies has demonstrated a persisting variability in clinical audit uptake and implementation across Europe and a lack of awareness of the BSSD clinical audit requirements. In recognition of these findings, the European Commission supported the QuADRANT project, led by the ESR and in partnership with ESTRO (European Association of Radiotherapy and Oncology) and EANM (European Association of Nuclear Medicine). QuADRANT was a 30-month project which completed in Summer 2022, aiming to provide an overview of the status of European clinical audit and identifying barriers and challenges to clinical audit uptake and implementation. This paper summarises the current position of European radiological clinical audit and considers the barriers and challenges that exist. Reference is made to the QuADRANT project, and a range of potential solutions are suggested to enhance radiological clinical audit across Europe.

Increased Petrous Bone Uptake on 18 F-PSMA-1007 PET/CT Due to Otospongiosis.

ABSTRACT: We report the case of a 74-year-old man who had undergone radical prostatectomy for prostatic cancer 6 months earlier. Elevated prostate-specific antigen during follow-up prompted 18 F-prostate-specific membrane antigen (PSMA) ligand PET/CT ( 18 F-PSMA-1007 PET/CT) to search for new manifestations of prostate cancer, revealing an increased focal uptake (SUV max , 5.9) in the left cochlear/pericochlear temporal bone and equivocal PSMA-RADS-3a external iliac nodes. Comparison with cone-beam CT and MRI showed that the focal temporal bone uptake corresponded to the typical morphological features of active otospongiosis (otosclerosis) in the context of a previously known long-standing otospongiosis.

Isolated Internal Ophthalmoplegia from Posterior Cerebral Artery Neurovascular Conflict.

We report a rare case of recurrent isolated internal ophthalmoplegia attributed to oculomotor nerve (CN III) compression by the posterior cerebral artery (PCA). A 30-year-old female patient presented with recurrent right-sided headaches, right periorbital pain, and slight anisocoria. Slit-lamp examination revealed normal anterior and posterior segments except for vermiform movements of the right pupil with a temporal hyporeactive flat area. Tonic pupils were ruled out with pilocarpine 0.1% testing. Suspecting an internal ophthalmoplegia, magnetic resonance imaging was ordered which demonstrated the right CN III indented by the PCA, fulfilling the criteria of a neurovascular conflict. The evaluation of unilateral mydriasis from internal ophthalmoplegia should prompt neuroimaging with exclusion of aneurysmal or compressive lesions. CN III palsy can rarely be caused by vascular anatomical variants because of the proximity of the posterior intracranial circulation and CN III. Newer, more precise imaging techniques will better help characterize neurovascular conflicts presenting as cranial nerve palsies.

Early MRI termination with major impact on the radiological interpretation: The experience of a large university hospital.

PURPOSE: To report the incidence of early magnetic resonance imaging (MRI) terminations and analyse their risk factors in a large university hospital. METHOD: All consecutive patients aged > 16 years who underwent an MRI over a 14-month period were included. The following parameters were collected: demographics, in- or outpatient, history of claustrophobia, anatomical region investigated, and early MRI termination along with its cause. The potential link between these parameters and early MRI termination was statistically analysed. RESULTS: Overall, 22,566MRIs were performed (10,792 [48%] men and 11,774[52%] women, mean age: 57 [range: 16-103] years). Early MRI termination was reported in 183 (0.8%) patients (99 men and 84 women, mean age: 63 years). Of these early terminations, 103 (56%) were due to claustrophobia and 80 (44%) to other causes. Early terminations were more common in inpatients than outpatients (1.2% vs. 0.6%, p < 0.001), for both claustrophobia- and non-claustrophobia-related reasons. A prior history of claustrophobia was strongly associated with claustrophobia-related early termination (6.6% vs. 0.2%, p = 0.0001). Non-claustrophobia-related early terminations were significantly more common (0.6% vs. 0.2%) in elderly patients (>65 years old) than in younger ones. No other parameter was significantly associated with early termination. CONCLUSIONS: Early MRI termination is currently rare. The main risk factors for claustrophobia-related terminations comprised a prior history of claustrophobia, and examinations in inpatients. Non-claustrophobia-related early terminations were more common in both elderly patients and inpatients.

Rosai-Dorfman disease presenting as stridor and hoarseness in a young female patient.

Rosai-Dorfman disease is a non-Langherans cell histiocytosis typically revealed by lymphadenopathy. Extranodal involvement occurs in 43% and most commonly involves the head and neck, skin, and bones. Few reports have described laryngeal lesions. We report the case of a Rosai-Dorfman disease in a 27-year-old female, presenting as an obstructing transglottic mass. We provide the results of the MRI and PET-scanner examination. As the treatment relies on surgical excision and the diagnosis depends on pathological examination, we also detail the analysis results that followed the surgical resection. This report highlights the necessity to consider Rosai-Dorfman disease as a potential diagnosis in case of a laryngeal submucosal mass.

Diagnostic value and prognostic significance of MRI findings in sudden sensorineural hearing loss.

Purpose: We evaluated the clinical significance of magnetic resonance imaging (MRI) findings and their prognostic value for initial hearing loss and recovery in patients with sudden sensorineural hearing loss (SSNHL). Materials and Methods: This retrospective study included consecutive adult patients with unilateral SSNHL, contrast-enhanced MRI and audiometric testing evaluated in our institution between 2005 and 2017. MRI reports, patient data, treatment, and audiometric tests were reviewed, with the relationship between MRI findings and hearing loss/recovery analyzed. Results: Overall, 266 patients were included. Additional symptoms comprised tinnitus (114/266; 43%), vertigo (45/266; 17%), ear pain (26/266; 10%), and ear pressure (6/266; 2%). At least one cardiovascular risk factor (hypertension, diabetes, hypercholesterolemia, cardiopathy, and active smoking) existed in 167/266 (63%) patients. Corticosteroid treatment was followed by 198/266 (74%) patients while contraindications/refusal/compliance precluded treatment in 68/266(26%). Complete, partial or slight hearing recovery occurred in 167/266 (63%) patients. Three MRI patient groups were identified: a group with normal MRI examinations or incidentalomas (128/266; 48%), a group with peripheral auditory system (PAS) lesions (95/266; 36%), and a group with central nervous system (CNS) lesions (43/266; 16%). PAS lesions included lesions from the cochlea to the brain stem (e.g., schwannoma, meningioma, labyrinthitis, intracochlear hemorrhage, vestibulocochlear neuritis), whereas CNS lesions corresponded in 42/43(98%) of cases to leukoaraiosis and other vascular lesions (e.g., stroke, hemorrhage, aneurysm, venous sinus thrombosis, and cavernoma). Belonging to one of the three MRI groups did not influence the degree of initial hearing loss, affected frequencies or treatment, p > .05. Gender and cardiovascular risk factors did neither affect initial hearing loss nor recovery. However, age > 70 years negatively affected initial hearing loss in all frequencies, as well as recovery in all frequencies except 1000 Hz. Also, poor recovery of initial high-frequency hearing loss (>1000 Hz) was significantly associated with CNS lesions. Conclusion: Age > 70 years and CNS lesions depicted by MRI independently predicted poor auditory recovery, albeit in different frequencies. Lay Summary: In patients with sudden hearing loss, older age (above 70 years) predicts poorer hearing recovery than in younger patients in most hearing frequencies. In addition, abnormalities of brain tissue revealed by MRI predict poorer hearing recovery at high frequencies. Level of Evidence: Level III.

What morphological MRI features enable differentiation of low-grade from high-grade soft tissue sarcoma?

Objective: To assess the diagnostic performance of morphological MRI features separately and in combination for distinguishing low- from high-grade soft tissue sarcoma (STS). Methods and materials: We retrospectively analysed pre-treatment MRI examinations with T1, T2 with and without fat suppression (FS) and contrast-enhanced T1 obtained in 64 patients with STS categorized histologically as low (n = 21) versus high grade (n = 43). Two musculoskeletal radiologists blinded to histology evaluated MRI features. Diagnostic performance was calculated for each reader and for MRI features showing significant association with histology (p < 0.05). Logistic regression analysis was performed to develop a diagnostic model to identify high-grade STS. Results: Among all evaluated MRI features, only six features had adequate interobserver reproducibility (kappa>0.5). Multivariate logistic regression analysis revealed a significant association with tumour grade for lesion heterogeneity on FS images, intratumoural enhancement≥51% of tumour volume and peritumoural enhancement for both readers (p < 0.05). For both readers, the presence of each of the three features yielded odds ratios for high grade versus low grade from 4.4 to 9.1 (p < 0.05). The sum of the positive features for each reader independent of reader expertise yielded areas under the curve (AUCs) > 0.8. The presence of ≥2 positive features indicated a high risk for high-grade sarcoma, whereas ≤1 positive feature indicated a low-to-moderate risk. Conclusion: A diagnostic MRI score based on tumour heterogeneity, intratumoural and peritumoural enhancement enables identification of lesions that are likely to be high-grade as opposed to low-grade STS. Advances in knowledge: Tumour heterogeneity in Fat Suppression sequence, intratumoural and peritumoural enhancement is identified as signs of high-grade sarcoma.

Detection of distant metastases and distant second primary cancers in head and neck squamous cell carcinoma: comparison of [18F]FDG PET/MRI and [18F]FDG PET/CT.

PURPOSE: This prospective study aimed to compare the diagnostic performance of [18]FDG PET/MRI and PET/CT for the detection of distant metastases and distant second primary cancers in patients with head and neck squamous cell carcinoma (HNSCC). METHODS: A total of 103 [18F]FDG PET/MRI examinations immediately followed by PET/CT were obtained in 82 consecutive patients for staging of primary HNSCC (n = 38), suspected loco-regional recurrence/follow-up (n = 41) or unknown primary HNSCC (n = 3). Histology and follow-up > 2 years formed the standard of reference. Blinded readers evaluated the anonymized PET/MRI and PET/CT examinations separately using a 5-point Likert score. Statistical analysis included: receiver operating characteristic (ROC) analysis, jackknife alternative free-response ROC (JAFROC) and region-of-interest (ROI)-based ROC to account for data clustering and sensitivity/specificity/accuracy comparisons for a score ≥ 3. RESULTS: Distant metastases and distant second primary cancers were present in 23/103 (22%) examinations in 16/82 (19.5%) patients, and they were more common in the post-treatment group (11/41, 27%) than in the primary HNSCC group (3/38, 8%), p = 0.039. The area under the curve (AUC) per patient/examination/lesion was 0.947 [0.927-1]/0.965 [0.917-1]/0.957 [0.928-0.987] for PET/MRI and 0.975 [0.950-1]/0.968 [0.920-1]/0.944 [0.910-0.979] for PET/CT, respectively (p > 0.05). The diagnostic performance of PET/MRI and PET/CT was similar according to JAFROC (p = 0.919) and ROI-based ROC analysis (p = 0.574). Sensitivity/specificity/accuracy for PET/MRI and PET/CT for a score ≥ 3 was 94%/88%/89% and 94%/91%/91% per patient, 96%/90%/91% and 96%/93%/93% per examination and 95%/85%/90% and 90%/86%/88% per lesion, respectively, p > 0.05. CONCLUSIONS: In HNSCC patients, PET/MRI and PET/CT had a high and similar diagnostic performance for detecting distant metastases and distant second primary cancers.

Decentralized Distributed Multi-institutional PET Image Segmentation Using a Federated Deep Learning Framework.

PURPOSE: The generalizability and trustworthiness of deep learning (DL)-based algorithms depend on the size and heterogeneity of training datasets. However, because of patient privacy concerns and ethical and legal issues, sharing medical images between different centers is restricted. Our objective is to build a federated DL-based framework for PET image segmentation utilizing a multicentric dataset and to compare its performance with the centralized DL approach. METHODS: PET images from 405 head and neck cancer patients from 9 different centers formed the basis of this study. All tumors were segmented manually. PET images converted to SUV maps were resampled to isotropic voxels (3 × 3 × 3 mm3) and then normalized. PET image subvolumes (12 × 12 × 12 cm3) consisting of whole tumors and background were analyzed. Data from each center were divided into train/validation (80% of patients) and test sets (20% of patients). The modified R2U-Net was used as core DL model. A parallel federated DL model was developed and compared with the centralized approach where the data sets are pooled to one server. Segmentation metrics, including Dice similarity and Jaccard coefficients, percent relative errors (RE%) of SUVpeak, SUVmean, SUVmedian, SUVmax, metabolic tumor volume, and total lesion glycolysis were computed and compared with manual delineations. RESULTS: The performance of the centralized versus federated DL methods was nearly identical for segmentation metrics: Dice (0.84 ± 0.06 vs 0.84 ± 0.05) and Jaccard (0.73 ± 0.08 vs 0.73 ± 0.07). For quantitative PET parameters, we obtained comparable RE% for SUVmean (6.43% ± 4.72% vs 6.61% ± 5.42%), metabolic tumor volume (12.2% ± 16.2% vs 12.1% ± 15.89%), and total lesion glycolysis (6.93% ± 9.6% vs 7.07% ± 9.85%) and negligible RE% for SUVmax and SUVpeak. No significant differences in performance (P > 0.05) between the 2 frameworks (centralized vs federated) were observed. CONCLUSION: The developed federated DL model achieved comparable quantitative performance with respect to the centralized DL model. Federated DL models could provide robust and generalizable segmentation, while addressing patient privacy and legal and ethical issues in clinical data sharing.

MR Imaging of Laryngeal and Hypopharyngeal Cancer.

State-of-the-art MR imaging of the larynx and hypopharynx with high-resolution surface coils, parallel imaging techniques, and DWI has several advantages over CT for assessing submucosal tumor spread, in particular neoplastic involvement of the paraglottic space, laryngeal cartilages, and extralaryngeal soft tissues. Current diagnostic MR imaging criteria based on a combination of distinct imaging features on morphologic sequences combined with DWI allow improved discrimination between tumor, peritumoral inflammation, and fibrosis and, ultimately, an increased precision for submucosal tumor delineation, which is a key prerequisite for tailored treatment options. Multiparametric MR imaging with DWI has a higher diagnostic performance than CT.

Fully Automated Gross Tumor Volume Delineation From PET in Head and Neck Cancer Using Deep Learning Algorithms.

PURPOSE: The availability of automated, accurate, and robust gross tumor volume (GTV) segmentation algorithms is critical for the management of head and neck cancer (HNC) patients. In this work, we evaluated 3 state-of-the-art deep learning algorithms combined with 8 different loss functions for PET image segmentation using a comprehensive training set and evaluated its performance on an external validation set of HNC patients. PATIENTS AND METHODS: 18F-FDG PET/CT images of 470 patients presenting with HNC on which manually defined GTVs serving as standard of reference were used for training (340 patients), evaluation (30 patients), and testing (100 patients from different centers) of these algorithms. PET image intensity was converted to SUVs and normalized in the range (0-1) using the SUVmax of the whole data set. PET images were cropped to 12 × 12 × 12 cm3 subvolumes using isotropic voxel spacing of 3 × 3 × 3 mm3 containing the whole tumor and neighboring background including lymph nodes. We used different approaches for data augmentation, including rotation (-15 degrees, +15 degrees), scaling (-20%, 20%), random flipping (3 axes), and elastic deformation (sigma = 1 and proportion to deform = 0.7) to increase the number of training sets. Three state-of-the-art networks, including Dense-VNet, NN-UNet, and Res-Net, with 8 different loss functions, including Dice, generalized Wasserstein Dice loss, Dice plus XEnt loss, generalized Dice loss, cross-entropy, sensitivity-specificity, and Tversky, were used. Overall, 28 different networks were built. Standard image segmentation metrics, including Dice similarity, image-derived PET metrics, first-order, and shape radiomic features, were used for performance assessment of these algorithms. RESULTS: The best results in terms of Dice coefficient (mean ± SD) were achieved by cross-entropy for Res-Net (0.86 ± 0.05; 95% confidence interval [CI], 0.85-0.87), Dense-VNet (0.85 ± 0.058; 95% CI, 0.84-0.86), and Dice plus XEnt for NN-UNet (0.87 ± 0.05; 95% CI, 0.86-0.88). The difference between the 3 networks was not statistically significant (P > 0.05). The percent relative error (RE%) of SUVmax quantification was less than 5% in networks with a Dice coefficient more than 0.84, whereas a lower RE% (0.41%) was achieved by Res-Net with cross-entropy loss. For maximum 3-dimensional diameter and sphericity shape features, all networks achieved a RE ≤ 5% and ≤10%, respectively, reflecting a small variability. CONCLUSIONS: Deep learning algorithms exhibited promising performance for automated GTV delineation on HNC PET images. Different loss functions performed competitively when using different networks and cross-entropy for Res-Net, Dense-VNet, and Dice plus XEnt for NN-UNet emerged as reliable networks for GTV delineation. Caution should be exercised for clinical deployment owing to the occurrence of outliers in deep learning-based algorithms.

Introducing the Node Reporting and Data System 1.0 (Node-RADS): a concept for standardized assessment of lymph nodes in cancer.

"Node-RADS" addresses the lack of consensus in the radiologic assessment of lymph node involvement by cancer and meets the increasing demand for structured reporting on the likelihood of disease involvement. Node Reporting and Data System 1.0 (Node-RADS) systematically classifies the degree of suspicion of lymph node involvement based on the synthesis of established imaging findings. Straightforward definitions of imaging findings for two proposed scoring categories "size" and "configuration" are combined into assessment categories between 1 ("very low likelihood") and 5 ("very high likelihood"). This scoring system is suitable for assessing likely involvement of lymph nodes on CT and MRI scans. It can be applied at any anatomical site, and to regional and non-regional lymph nodes in relation to a primary tumor location. Node-RADS will improve communication with referring physicians and promote the consistency of reporting for primary staging and in response assessment settings. KEY POINTS: • Node-RADS standardizes reporting of possible cancer involvement of regional and distant lymph nodes on CT and MRI. • Node-RADS proposes the scoring categories "size" and "configuration" for assigning the 5-point Node-RADS score from 1 ("very low likelihood") to 5 ("very high likelihood"). • Node-RADS aims to increase consensus among radiologists for primary staging and in response assessment settings.

Emergency Computed Tomography: How Misinterpretations Vary According to the Periods of the Nightshift?

OBJECTIVE: To evaluate the accuracy of initial computed tomography (CT) interpretations made by radiology residents during nightshifts in the emergency department. METHODS: Preliminary CT reports performed by radiology residents during 120 consecutive nightshifts (08:30 pm to 08:30 am) were reviewed, attendings' final interpretation being the reference standard. Nightshifts were divided into four consecutive periods of 3 hours. Major misinterpretations were related to potentially life-threatening conditions if not treated immediately after CT. The rate of misinterpretations was calculated for all CT examinations, separately for nightshift's periods and for residents' training years. RESULTS: Misinterpretations were recorded in 155 (7.4%) of 2102 CT examinations, 0.6% (13/2102) were major. There were 2.2% (4/186) major misinterpretations that occurred during the last period of the nightshift versus 0.4% (9/1916) during the first periods of the night (P < 0.05). Of all misinterpretations, 8.5% (130/1526) were made by third- and fourth-year residents and 4.3% (25/576) by fifth-year residents (P < 0.005). CONCLUSIONS: Major misinterpretations occur at the end of the nightshift, which may be explained by the fatigue effect. The rate of misinterpretations is lower among fifth-year residents, which may be related to their prior experience in reading emergency cases.