Background and Purpose: Aneurysmal wall enhancement (AWE) on vessel wall magnetic resonance imaging (VW-MRI) has been described as a new imaging biomarker of unstable unruptured intracranial aneurysms (UIAs). Previous studies of symptomatic UIAs are limited due to small sample sizes and lack of AWE quantification. Our study aims to investigate whether qualitative and quantitative assessment of AWE can differentiate symptomatic and asymptomatic UIAs. Methods: Consecutive patients with UIAs were prospectively recruited for vessel wall magnetic resonance imaging at 3T from October 2014 to October 2019. UIAs were categorized as symptomatic if presenting with sentinel headache or oculomotor nerve palsy directly related to the aneurysm. Evaluation of wall enhancement included enhancement pattern (0=none, 1=focal, and 2=circumferential) and quantitative wall enhancement index (WEI). Univariate and multivariate analyses were used to identify the parameters associated with symptoms. Results: Two hundred sixty-seven patients with 341 UIAs (93 symptomatic and 248 asymptomatic) were included in this study. Symptomatic UIAs more frequently showed circumferential AWE than asymptomatic UIAs (66.7% versus 17.3%, P <0.001), as well as higher WEI (median [interquartile range], 1.3 [1.0–1.9] versus 0.3 [0.1–0.9], P <0.001). In multivariate analysis, both AWE pattern and WEI were independent factors associated with symptoms (odds ratio=2.03 across AWE patterns [95% CI, 1.21–3.39], P =0.01; odds ratio=3.32 for WEI [95% CI, 1.51–7.26], P =0.003). The combination of AWE pattern and WEI had an area under the curve of 0.91 to identify symptomatic UIAs, with a sensitivity of 95.7% and a specificity of 73.4%. Conclusions: In a large cohort of UIAs with vessel wall magnetic resonance imaging, both AWE pattern and WEI were independently associated with aneurysm-related symptoms. The qualitative and quantitative features of AWE can potentially be used to identify unstable intracranial aneurysms.

Background and Purpose— This study aims to investigate the association between the characteristics of atherosclerotic plaques of middle cerebral artery and recurrent ischemic stroke using magnetic resonance vessel wall imaging. Methods— One hundred and five patients with ischemic stroke attributed to middle cerebral artery plaque underwent high-resolution black-blood magnetic resonance vessel wall imaging. They were divided into group 1, with the first episode of acute stroke (imaging within 4 weeks of stroke, n=44); group 2, with recurrent acute stroke (n=29); and group 3, with chronic stroke (imaging after 3 months of stroke, n=32). Plaque characteristics including plaque area, plaque burden, contrast-enhancement ratio, eccentricity, and degree of stenosis were measured and compared across 3 groups. Association between plaque characteristics and recurrent strokes was investigated by multivariate analysis. Results— Plaque burden was significantly greater in recurrent stroke group than the other 2 groups (median: group 2, 82.7%, versus group 1, 76.3%, and group 3, 73.4%; P =0.001). Patients with acute stroke had higher enhancement ratio than patients with chronic stroke (median: group 1, 1.59, and group 2, 1.90, versus group 3, 1.33; P =0.014). Comparing to first-onset acute stroke patients, recurrent stroke patients were older, more likely with female sex and hypertension, and had higher plaque burden. After adjustment of clinical factors, plaque burden was the only independent imaging feature associated with recurrent stroke (odds ratio, 2.26, per 10% increase [95% CI, 1.03–4.96]; P =0.042). Conclusions— Higher plaque burden of middle cerebral artery identified on magnetic resonance vessel wall imaging is independently associated with recurrent ischemic stroke.

The renal vein variations are not unusual, particularly in the RRV. Anomalies of the LRV are more complex than those of the RRV. The renal vein anatomy can be well depicted by MDCT angiography. Our new classification of the renal vein variations will improve the recognition of the renal vein morphology preoperatively.

Head and neck (HN) rhabdomyosarcoma (RMS) is an aggressive malignancy, which is rarely encountered and is commonly misdiagnosed as another type of tumor. The aim of the present study was to investigate the computed tomography (CT) and magnetic resonance imaging (MRI) features of HNRMS and analyze the correlations between the imaging observations and the pathological subtypes. A total of 10 HNRMS patients (three males and seven females; median age, 16 years) were reviewed retrospectively by only CT (n=1), only MRI (n=2), as well as CT and MRI (n=7). In addition, the clinical data, imaging observations and pathological results were recorded and analyzed. The origins of the 10 HNRMSs (eight embryonal and two alveolar subtypes) included the ethmoid sinus (n=4), maxillary sinus (n=1), orbit (n=3), nasopharynx (n=1) and frontotemporal subcutaneous area (n=1). On the CT and MRI images, the soft-tissue masses exhibited ill-defined borders (n=9), bony destruction (n=10), multi-cavity growth (n=7) and cervical lymph node metastasis (n=2), whereas calcification and hemorrhaging were not identified. On CT, eight of the HNRMSs appeared slightly hypodense (2/8) or isodense (6/8) with homogeneous enhancement (4/4). On T1-weighted images (WI), nine tumors exhibited isointensity (9/9) and on T2WI, six tumors demonstrated homogeneous hyperintensity with homogeneous enhancement on contrast-enhanced (CE)-T1WI. In addition, three embryonal RMSs, which originated from the ethmoid sinus, exhibited heterogeneous hyperintensity on T2WI and nodule-shaped enhancement patterns on CE-T1WI. The results of the present study indicated that MRI may accurately demonstrate the location and extent of HNRMS and that the imaging features of HNRMS may be similar to those of other tumors. However, a tumor exhibiting heterogeneous hyperintensity on T2WI and a nodule-shaped enhancement pattern on CE-T1WI in the ethmoid sinus may present specific MRI features, which clearly indicates the botryoid subtype of embryonal RMS.

• TSR is a recognized independent prognostic factor in many solid tumours. • D and f values measured by IVIM MRI are independently correlated with TSR while D* is not. • IVIM offers the potential to predict TSR.

No abstract

Background and objectiveClinical characteristics of obesity are heterogenous, but current classification for diagnosis is simply based on BMI or metabolic healthiness. The purpose of this study was to use machine learning to explore a more precise classification of obesity subgroups towards informing individualized therapy.Subjects and MethodsIn a multi-center study (n=2495), we used unsupervised machine learning to cluster patients with obesity from Shanghai Tenth People’s hospital (n=882, main cohort) based on three clinical variables (AUCs of glucose and of insulin during OGTT, and uric acid). Verification of the clustering was performed in three independent cohorts from external hospitals in China (n = 130, 137, and 289, respectively). Statistics of a healthy normal-weight cohort (n=1057) were measured as controls.ResultsMachine learning revealed four stable metabolic different obese clusters on each cohort. Metabolic healthy obesity (MHO, 44% patients) was characterized by a relatively healthy-metabolic status with lowest incidents of comorbidities. Hypermetabolic obesity-hyperuricemia (HMO-U, 33% patients) was characterized by extremely high uric acid and a large increased incidence of hyperuricemia (adjusted odds ratio [AOR] 73.67 to MHO, 95%CI 35.46-153.06). Hypermetabolic obesity-hyperinsulinemia (HMO-I, 8% patients) was distinguished by overcompensated insulin secretion and a large increased incidence of polycystic ovary syndrome (AOR 14.44 to MHO, 95%CI 1.75-118.99). Hypometabolic obesity (LMO, 15% patients) was characterized by extremely high glucose, decompensated insulin secretion, and the worst glucolipid metabolism (diabetes: AOR 105.85 to MHO, 95%CI 42.00-266.74; metabolic syndrome: AOR 13.50 to MHO, 95%CI 7.34-24.83). The assignment of patients in the verification cohorts to the main model showed a mean accuracy of 0.941 in all clusters.ConclusionMachine learning automatically identified four subtypes of obesity in terms of clinical characteristics on four independent patient cohorts. This proof-of-concept study provided evidence that precise diagnosis of obesity is feasible to potentially guide therapeutic planning and decisions for different subtypes of obesity.Clinical Trial Registrationwww.ClinicalTrials.gov, NCT04282837.

No abstract