Computational competitions are the standard for benchmarking medical image analysis algorithms, but they typically use small curated test datasets acquired at a few centers, leaving a gap to the reality of diverse multicentric patient data. To this end, the Federated Tumor Segmentation (FeTS) Challenge represents the paradigm for real-world algorithmic performance evaluation. The FeTS challenge is a competition to benchmark (i) federated learning aggregation algorithms and (ii) state-of-the-art segmentation algorithms, across multiple international sites. Weight aggregation and client selection techniques were compared using a multicentric brain tumor dataset in realistic federated learning simulations, yielding benefits for adaptive weight aggregation, and efficiency gains through client sampling. Quantitative performance evaluation of state-of-the-art segmentation algorithms on data distributed internationally across 32 institutions yielded good generalization on average, albeit the worst-case performance revealed data-specific modes of failure. Similar multi-site setups can help validate the real-world utility of healthcare AI algorithms in the future.

Der systemische Lupus erythematodes wird mit einer hohen Mortalitätsrate in Verbindung gebracht. Umso wichtiger ist die Kenntnis der zugrunde liegenden Pathomechanismen, der neuroimmunen Schnittstellen und der Komplikationen dieser Erkrankung für die adäquate Diagnostik und Behandlung.

Objective: Preoperative imaging assessment influences the decision to perform mastoidectomy for the mastoid extension of middle ear cholesteatoma. This study compared the performance of temporal subtraction CT (TSCT) and non-echoplanar diffusion-weighted imaging (non-EP DWI) in evaluating such mastoid extensions. Methods: We retrospectively evaluated 239 consecutive patients with surgically proven mid-dle ear cholesteatoma between April 2016 and April 2021. The diagnostic performance of TSCT, wherein the presence of black color indicated progressive bone erosion, and non-EP DWI, wherein high signal intensity in the mastoid region suggested mastoid extension, was compared using Fisher's exact test.Results: In 34 patients with evaluable TSCT images, black color was significantly more common in patients with mastoid extension than in those without; the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of TSCT were 1.00, 0.95, 0.94, 1.00, and 0.97, respectively. In 90 patients with evaluable non-EP DWI, high signal intensity was significantly more common in patients with mastoid extension than in those without; the sensitivity, specificity, PPV, NPV, and accuracy of non-EP DWI were 0.88, 0.85, 0.91, 0.81, and 0.87, respectively. In 16 patients with both evaluable TSCT and non-EP DWI, the diagnostic performance of the TSCT was slightly superior to that of the non-EP DWI for predicting mastoid extension, although the difference was not significant.Conclusions: TSCT images generated using consecutively acquired preoperative high-resolution CT images are useful for predicting mastoid extension of middle ear cholesteatoma, and the diagnostic performance of TSCT is non-inferior to that of non-EP DWI.(c) 2022 Japanese Society of Otorhinolaryngology-Head and Neck Surgery, Inc. Published by Elsevier B.V. All rights reserved.

Ectopic tissue is an anatomic abnormality in which tissue develops in an area outside its normal location. It is primarily caused by abnormalities during the process of embryologic development. Although the majority of individuals with ectopic tissues remain asymptomatic, various symptoms and associated complications can occur. Failure in normal embryologic development leads to loss of normal physiologic function or may result in harmful functions such as ectopic hormonal secretion in the ectopic pituitary adenoma. Ectopic tissues may also frequently mimic tumors. For example, developmental abnormalities in the pharyngeal pouches may result in an ectopic parathyroid gland and ectopic thymus, both of which are frequently misdiagnosed as tumors. Adequate knowledge of embryology is essential for understanding the differential diagnoses of ectopic tissues and facilitating appropriate management. The authors summarize the embryologic development and pathogenesis of ectopic tissues by using illustrations to facilitate a deeper understanding of embryologic development and anatomy. Characteristic imaging findings (US, CT, MRI, and scintigraphy) are described for ectopic tissues of the brain, head, neck, thorax, abdomen, and pelvis by focusing on common conditions that radiologists may encounter in daily practice and their differential diagnoses. ©RSNA, 2023 Quiz questions for this article are available through the Online Learning Center.

When head and neck infection is suspected, appropriate imaging contributes to treatment decisions and prognosis. While contrast‐enhanced CT is the standard imaging modality for evaluating head and neck infections, MRI can better characterize the skull base, intracranial involvement, and osteomyelitis, implying that these are complementary techniques for a comprehensive assessment. Both CT and MRI are useful in the evaluation of abscesses and thrombophlebitis, while MRI is especially useful in the evaluation of intracranial inflammatory spread/abscess formation, differentiation of abscess from other conditions, evaluation of the presence and activity of inflammation and osteomyelitis, evaluation of mastoid extension in middle ear cholesteatoma, and evaluation of facial neuritis and labyrinthitis. Apparent diffusion coefficient derived from diffusion‐weighted imaging is useful for differential diagnosis and treatment response of head and neck infections in various anatomical sites. Dynamic contrast‐enhanced MRI perfusion may be useful in assessing the activity of skull base osteomyelitis. MR bone imaging may be of additional value in evaluating bony structures of the skull base and jaw. Dual‐energy CT is helpful in reducing metal artifacts, evaluating deep neck abscess, and detecting salivary stones. Subtraction CT techniques are used to detect progressive bone‐destructive changes and to reduce dental amalgam artifacts. This article provides a region‐based approach to the imaging evaluation of head and neck infections, using both conventional and advanced imaging techniques.

Human immunodeficiency virus (HIV) infection causes substantial morbidity and mortality worldwide. Although antiretroviral therapy (ART) has changed the epidemiology of HIV in the last 20 years with increased survival and decreasing incidence of opportunistic infections (OI), CNS OI remain a major cause of morbidity. Improved survival has also increased neurological presentations due to co morbid conditions, treatment related side effects and inflammatory syndromes. Being familiar with the imaging findings, the impact of ART and interpretation of imaging in the context of clinical and laboratory findings is important for radiologists as well as clinicians in the management of HIV-infected patients.

Purpose This study tested the utility of diffusion-weighted imaging (DWI) and dynamic contrast-enhanced imaging (DCE-MRI) in differentiating paragangliomas and metastases in the jugular foramen in combination with conventional imaging. Methods Forty-nine consecutive patients with paragangliomas or metastases between January 2015 and April 2022 were included in this retrospective study. All patients had pretreatment DWI and DCE-MRI. Between paragangliomas and metastases, normalized apparent diffusion coefficient (nADCmean) and DCE-MRI parameters were compared along with conventional imaging features (enhancement pattern, presence of flow voids, cystic/necrotic change, and bone erosion). The diagnostic performance was tested using receiver operating characteristic (ROC) analysis. Results Thirty-five paragangliomas (5 male; median 49 years) and 14 metastases (9 male; median 61 years) were analyzed. The most common 3 primary cancers included 4 lung cancers, 3 breast cancers, and 3 melanomas. The presence of flow void was significantly different between paragangliomas and metastases (21/35 vs 2/14; P = 0.0047) in conventional imaging features, while fractional plasma volume (Vp) was significantly different between the two tumor types (median 0.46 vs 0.19; P < 0.001) in DWI and DCE-MRI parameters. The areas under the ROC curves (AUCs) of the presence of flow void and Vp were 0.72 and 0.93, respectively. The AUC of the combination of the presence of flow void and Vp was 0.95 and significantly improved compared to that of the presence of flow void ( P < 0.001). Conclusion Adding DCE-MRI to the head and neck protocol can aid in the precise differentiation between jugular foramen paragangliomas and metastases.

Hypophysitis is an inflammatory disease affecting the pituitary gland. Hypophysitis can be classified into multiple types depending on the mechanisms (primary or secondary), histology (lymphocytic, granulomatous, xanthomatous, plasmacytic/IgG4 related, necrotizing, or mixed), and anatomy (adenohypophysitis, infundibulo-neurohypophysitis, or panhypophysitis). An appropriate diagnosis is vital for managing these potentially life-threatening conditions. However, physiological morphological alterations, remnants, and neoplastic and non-neoplastic lesions may masquerade as hypophysitis, both clinically and radiologically. Neuroimaging, as well as imaging findings of other sites of the body, plays a pivotal role in diagnosis. In this article, we will review the types of hypophysitis and summarize clinical and imaging features of both hypophysitis and its mimickers.

The International Society for the Study of Vascular Anomalies (ISSVA) classification is a comprehensive histology-based scheme that was updated in 2018. It is important for cardiothoracic imagers to understand this classification to ensure that accurate terminology is used and that archaic terms are avoided when vascular lesions are described. Knowledge of the various malformations (including common conditions, such as venous malformation, arteriovenous fistula, and arteriovenous malformation) and vascular tumors allows for timely diagnosis and appropriate management. This review describes various vascular anomalies, in accordance with ISSVA classification and terminology; highlights key imaging features associated with each; and discusses the role of different imaging modalities. Keywords: Pulmonary, Soft Tissues/Skin, Vascular, Arteriovenous Malformation © RSNA, 2023

Stroke-like migraine attacks after radiation therapy (SMART) syndrome is a delayed complication of cranial irradiation, with subacute onset of stroke-like symptoms including seizures, visual disturbance, speech impairment, unilateral hemianopsia, facial droop, and aphasia, often associated with migraine-type headache. The diagnostic criteria were initially proposed in 2006. However, the diagnosis of SMART syndrome is challenging because clinical symptoms and imaging features of SMART syndrome are indeterminate and overlap with tumor recurrence and other neurologic diseases, which may result in inappropriate clinical management and unnecessary invasive diagnostic procedures. Recently, various imaging features and treatment recommendations for SMART syndrome have been reported. Radiologists and clinicians should be familiar with updates on clinical and imaging features of this delayed radiation complication because recognition of this entity can facilitate proper clinical work-up and management. This review provides current updates and a comprehensive overview of the clinical and imaging features of SMART syndrome.

Human immunodeficiency virus (HIV) infection causes substantial morbidity and mortality worldwide. Although antiretroviral therapy (ART) has changed the epidemiology of HIV in the last 20 years with increased survival and decreasing incidence of opportunistic infections (OI), CNS OI remain a major cause of morbidity. Improved survival has also increased neurological presentations due to co morbid conditions, treatment related side effects and inflammatory syndromes. Being familiar with the imaging findings, the impact of ART and interpretation of imaging in the context of clinical and laboratory findings is important for radiologists as well as clinicians in the management of HIV-infected patients.

Background and PurposeWernicke's encephalopathy (WE) is a severe acute disorder related to thiamine deficiency. This study was aimed at revealing the relationship between clinical and imaging findings and WE recovery. MethodsWe retrospectively reviewed 34 cases of WE diagnosed between 2003 and 2020 (median age: 57 years, 14 females) at two academic institutions. WE cases were divided into two groups with symptomatic recovery within 4 weeks (group 1) or later (group 2). The lesion sites were divided into typical and atypical sites (total sites defined as when either typical or atypical sites were involved). Clinical and MRI features were compared between them as appropriate. ResultsWE patients were divided into group 1 (19 cases, median age: 57 years, 10 females) and group 2 (15 cases, median age: 57 years, four females). Regarding clinical features, only cerebellar ataxia was more often observed in group 1 than in group 2. Regarding MRI features, signal abnormality on T2-weighted image (WI)/fluid-attenuated inversion recovery (FLAIR) was more often observed in atypical sites between groups 1 and 2 (1/19 vs. 7/15; p = .01). There were significant differences between groups 1 and 2 regarding the presence of both vasogenic edema and cytotoxic edema in total sites (4/11 vs. 11/15, p = .005; 1/19 vs. 6/15, p = .03), with a significant difference in the presence of vasogenic edema in typical sites (4/19 vs. 10/15, p = .01). ConclusionThe early recovered group showed a lower incidence of T2WI/FLAIR abnormality in atypical sites and diffusion signal abnormality in total or typical sites with a lower incidence of cerebellar ataxia.

Purpose: Differentiation of paragangliomas and meningiomas can be a challenge. This study aimed to assess the utility of dynamic susceptibility contrast perfusion MRI (DSC-MRI) to distinguish paragangliomas from meningiomas.Methods: This retrospective study included 40 patients with paragangliomas and meningiomas in the cer-ebellopontine angle and jugular foramen region between March 2015 and February 2022 in a single institution. Pretreatment DSC-MRI and conventional MRI were performed in all cases. Normalized relative cerebral blood volume (nrCBV), relative cerebral blood flow (nrCBF), relative mean transit time (nrMTT), and time to peak (nTTP) as well as conventional MRI features were compared between the 2 tumor types and between menin-gioma subtypes as appropriate. Receiver operating characteristic curve and multivariate logistic regression analysis were performed.Results: Twenty-eight meningiomas including 8 WHO grade II meningiomas (12 males, 16 females; median age 55 years) and 12 paragangliomas (5 males, 7 females; median age 35 years) were included in this study. Par-agangliomas had a higher rate of cystic/necrotic changes (10/12 vs 10/28; P = 0.014), a higher rate of internal flow voids (9/12 vs 8/28; P = 0.013), higher nrCBV (median 9.78 vs 6.64; P = 0.04), and shorter nTTP (median 0.78 vs 1.06; P < 0.001) than meningiomas. There was no difference in conventional imaging features and DSC-MRI parameters between meningioma subtypes. nTTP was identified as the most significant parameter for the 2 tumor types in the multivariate logistic regression analysis (P = 0.009).Conclusions: In this small retrospective study, DSC-MRI perfusion differences were observed between para-gangliomas and meningiomas, but not between grade I and II meningiomas.

Background A tap test is established as an evaluation method to indicate shunt surgery for hydrocephalus, especially idiopathic normal pressure hydrocephalus (iNPH). The timing of gait assessment after the test is still controversial, while some studies reported the effectiveness of the gait evaluation up to 2nd day after tap tests. Our study explored whether the gait evaluation on the 7th day after a tap test is necessary. Methods We retrospectively evaluated 129 consecutive cases with possible iNPH who performed gait assessment on all 1st, 3rd, and 7th days after tap tests between May 2020 and February 2022. We reviewed the following items of the patients: age, sex, modified Rankin scale, iNPH grading scale, Mini-Mental State Evaluation (MMSE), and neurological imaging. The number of probable iNPH patients who improved their gait each day after the test was analyzed. We also assessed the number of definite iNPH patients and revealed the background characteristics of the patients who showed gait improvement on the 7th day after the tests. Results Of the 129 patients who met our inclusion criteria, 57 were judged as probable iNPH on the 1st day, 28 were new on the 3rd, and 23 were new on the 7th. The number of probable iNPH patients up to the 7th day after tests was significantly more extensive than that of those up to the 3rd (108 [83.7%] vs. 85 [65.9%]; 95% confidence interval [CI], p < 0.0001). The number of definite iNPH patients was also significantly more prominent when the evaluation after the tests was performed on all of the 1st, the 3rd, and the 7th days than just on the 1st (72 vs. 42; 95% CI, p = 0.00016) or both of the 1st and the 3rd (72 vs. 61; 95% CI, p = 0.00074). No statistically significant difference existed in the patients’ backgrounds except for the pre-tap test MMSE. Conclusion Gait evaluation on the 7th day after tap tests, in addition to the first few days, may reduce the number of iNPH patients who miss the opportunity of getting beneficial treatment.

Injuries to the plantar plate complex of the first toe and the lesser toes, although closely related anatomically, differ significantly in their pathogenesis. Lesser metatarsophalangeal joint plantar plate degeneration and tear typically presents as an attritional pattern of capsuloligamentous deficiency in middle-aged patients, whereas first metatarsophalangeal joint capsuloligamentous injury is typically acute and occurs in younger adult athletes engaged in sports involving repetitive running and jumping. Consequently, considerations regarding surgical decision-making also differ. Knowledge of the anatomy, pathophysiological basis, common patterns, grading and classification of these injuries, and indications for surgery, will aid imaging interpretation in the preoperative setting. The acuity and extent of injury, tissue quality and functional requirements of the patient (such as timely return to sport) influence clinical decision-making with respect to conservative versus surgical management and are informed by the use of optimized multimodal imaging. This section will provide an overview of these injuries in 2 parts, separated into those affecting the first toe and the lesser toes, and will highlight imaging findings relevant to surgical decision-making.

Although machine learning (ML) has shown promise in numerous domains, there are concerns about generalizability to out-of-sample data. This is currently addressed by centrally sharing ample, and importantly diverse, data from multiple sites. However, such centralization is challenging to scale (or even not feasible) due to various limitations. Federated ML (FL) provides an alternative to train accurate and generalizable ML models, by only sharing numerical model updates. Here we present findings from the largest FL study to-date, involving data from 71 healthcare institutions across 6 continents, to generate an automatic tumor boundary detector for the rare disease of glioblastoma, utilizing the largest dataset of such patients ever used in the literature (25,256 MRI scans from 6,314 patients). We demonstrate a 33% improvement over a publicly trained model to delineate the surgically targetable tumor, and 23% improvement over the tumor's entire extent. We anticipate our study to: 1) enable more studies in healthcare informed by large and diverse data, ensuring meaningful results for rare diseases and underrepresented populations, 2) facilitate further quantitative analyses for glioblastoma via performance optimization of our consensus model for eventual public release, and 3) demonstrate the effectiveness of FL at such scale and task complexity as a paradigm shift for multi-site collaborations, alleviating the need for data sharing.

BACKGROUND AND PURPOSE:Free flap reconstruction in patients with head and neck cancer carries a risk of postoperative complications, and radiologic predictive factors have been limited. The aim of this study was to assess the factors that predict free flap reconstruction failure using CT and MR perfusion.MATERIALS AND METHODS:& nbsp;This single-center prospective study included 24 patients (mean age, 62.7 [SD, 9.0]?years; 16 men) who had free flap reconstruction from January 2016 to May 2018. CT perfusion and dynamic contrast-enhanced MR imaging with conventional CT and MR imaging were performed between 2 and 4?days after the free flap surgery, and the wound assessments within 14?days after the surgery were conducted by the surgical team. The parameters of CT perfusion and dynamic contrast-enhanced MR imaging with conventional imaging findings and patient demographics were compared between the patients with successful free flap reconstruction and those with wound failure as appropriate. P < .05 was considered significant.RESULTS:& nbsp;There were 19 patients with successful free flap reconstruction and no wound complications (mean age, 63.9 [SD, 9.5]?years; 14 men), while 5 patients had wound failure (mean age, 58.0 [SD, 5.7]?years; 2 men). Blood flow, blood volume, MTT, and time maximum intensity projection (P?=?.007, .007, .015, and .004, respectively) in CT perfusion, and fractional plasma volume, volume transfer constant, peak enhancement, and time to maximum enhancement (P=.006, .039, .004, and .04, respectively) in dynamic contrast-enhanced MR imaging were significantly different between the 2 groups.CONCLUSIONS:& nbsp;CT perfusion and dynamic contrast-enhanced MR imaging are both promising imaging techniques to predict wound complications after head and neck free flap reconstruction.

Background and Purpose Astroblastoma is a rare type of glial tumor, histologically classified into two types with different prognoses: high and low grade. We aimed to investigate the CT and MRI findings of astroblastomas by collecting studies with analyzable neuroimaging data and extracting the imaging features useful for tumor grading. Methods We searched for reports of pathologically proven astroblastomas with analyzable neuroimaging data using PubMed, Scopus, and Embase. Sixty-five studies with 71 patients with astroblastomas met the criteria for a systematic review. We added eight patients from our hospital, resulting in a final study cohort of 79 patients. The proportion of high-grade tumors was compared in groups based on the morphology (typical and atypical) using Fisher's exact test. Results High- and low-grade tumors were 35/71 (49.3%) and 36/71 (50.7%), respectively. There was a significant difference in the proportion of high-grade tumors based on the tumor morphology (typical morphology: high-grade = 33/58 [56.9%] vs. atypical morphology, 2/13 [15.4%], p = .012). The reviews of neuroimaging findings were performed using the images included in each article. The articles had missing data due to the heterogeneity of the collected studies. Conclusions Detailed neuroimaging features were clarified, including tumor location, margin status, morphology, CT attenuation, MRI signal intensity, and contrast enhancement pattern. The classification of tumor morphology may help predict the tumor's histological grade, contributing to clinical care and future oncologic research.

BACKGROUND AND PURPOSE: Differentiation of skull base tumors, including chondrosarcomas, chordomas, and metastases, on conventional imaging remains a challenge. We aimed to test the utility of DWI and dynamic contrast-enhanced MR imaging for skull base tumors. MATERIALS AND METHODS: Fifty-nine patients with chondrosarcomas, chordomas, or metastases between January 2015 and October 2021 were included in this retrospective study. Pretreatment normalized mean ADC and dynamic contrast-enhanced MR imaging parameters were calculated. The Kruskal-Wallis H test for all tumor types and the Mann-Whitney U test for each pair of tumors were used. RESULTS: Fifteen chondrosarcomas (9 men; median age, 62 years), 14 chordomas (6 men; median age, 47 years), and 30 metastases (11 men; median age, 61 years) were included in this study. Fractional plasma volume helped distinguish all 3 tumor types (P = .003, <.001, and <.001, respectively), whereas the normalized mean ADC was useful in distinguishing chondrosarcomas from chordomas and metastases (P < .001 and P < .001, respectively); fractional volume of extracellular space, in distinguishing chondrosarcomas from metastases (P = .02); and forward volume transfer constant, in distinguishing metastases from chondrosarcomas/chondroma (P = .002 and .002, respectively) using the Kruskal-Wallis H test. The diagnostic performances of fractional plasma volume for each pair of tumors showed areas under curve of 0.86–0.99 (95% CI, 0.70–1.0); the forward volume transfer constant differentiated metastases from chondrosarcomas/chordomas with areas under curve of 0.82 and 0.82 (95% CI, 0.67–0.98), respectively; and the normalized mean ADC distinguished chondrosarcomas from chordomas/metastases with areas under curve of 0.96 and 0.95 (95% CI, 0.88–1.0), respectively. CONCLUSIONS: DWI and dynamic contrast-enhanced MR imaging sequences can be beneficial for differentiating the 3 common skull base tumors.