To investigate the effect of transfer learning on computed tomography (CT) images for the benign and malignant classification on renal tumors and to attempt to improve the classification accuracy by building patient-level models. One hundred ninety-two cases of renal tumors were collected and identified by pathologic diagnosis within 15 days after enhanced CT examination (66% male, 70% malignant renal tumors, average age of 62.27 ± 12.26 years). The InceptionV3 model pretrained by the ImageNet dataset was cross-trained to perform this classification. Five image-level models were established for each of the Slice, region of interest (ROI), and rectangular box region (RBR) datasets. Then, two patient-level models were built based on the optimal image-level models. The network's performance was evaluated through analysis of the receiver operating characteristic (ROC) and five-fold cross-validation. In the image-level models, the test results of model trained on the Slice dataset [accuracy (ACC) = 0.69 and Matthews correlation coefficient (MCC) = 0.45] were the worst. The corresponding results on the ROI dataset (ACC = 0.97 and MCC = 0.93) were slightly better than those on the RBR dataset (ACC = 0.93 and MCC = 0.85) when freezing the weights before the mixed6 layer. Compared with the image-level models, both patient-level models could discriminate better (ACC increased by 2%-5%) on the RBR and Slice datasets. Deep learning can be used to classify benign and malignant renal tumors from CT images. Our patient-level models could benefit from 3D data to improve the accuracy.

Purpose: The neural mechanisms that give rise to the phantom sound of tinnitus have not been fully elucidated. Neuroimaging studies have revealed abnormalities in resting-state activity that could represent the neural signature of tinnitus, but there is considerable heterogeneity in the data. To address this issue, we conducted a meta-analysis of published neuroimaging studies aimed at identifying a common core of resting-state brain abnormalities in tinnitus patients. Methods: A systematic search was conducted for whole-brain resting-state neuroimaging studies with SPECT, PET and functional MRI that compared chronic tinnitus patients with healthy controls. The authors searched PubMed, Science Direct, Web of Knowledge and Embase databases for neuroimaging studies on tinnitus published up to September 2016. From each study, coordinates were extracted from clusters with significant differences between tinnitus subjects and controls. Meta-analysis was performed using the activation likelihood estimation method. Results: Data were included from nine resting-state neuroimaging studies that reported a total of 51 distinct foci. The meta-analysis identified consistent regions of increased resting-state brain activity in tinnitus patients relative to controls that included, bilaterally, the insula, middle temporal gyrus, inferior frontal gyrus, parahippocampal gyrus, cerebellum posterior lobe and right superior frontal gyrus. Moreover, decreased brain activity was only observed in the left cuneus and right thalamus. Conclusions: The current meta-analysis is, to our knowledge, the first to demonstrate a characteristic pattern of resting-state brain abnormalities that may serve as neuroimaging markers and contribute to the understanding of neuropathophysiological mechanisms for chronic tinnitus.

Background: Resting-state functional magnetic resonance imaging (fMRI) has confirmed disrupted visual network connectivity in migraine without aura (MwoA). The thalamus plays a pivotal role in a number of pain conditions, including migraine. However, the significance of altered thalamo-visual functional connectivity (FC) in migraine remains unknown. The goal of this study was to explore thalamo-visual FC integrity in patients with MwoA and investigate its clinical significance. Methods: Resting-state fMRI data were acquired from 33 patients with MwoA and 22 well-matched healthy controls. After identifying the visual network by independent component analysis, we compared neural activation in the visual network and thalamo-visual FC and assessed whether these changes were linked to clinical characteristics. We used voxel-based morphometry to determine whether functional differences were dependent on structural differences. Results: The visual network exhibited significant differences in regions (bilateral cunei, right lingual gyrus and left calcarine sulcus) by inter-group comparison. The patients with MwoA showed significantly increased FC between the left thalami and bilateral cunei and between the right thalamus and the contralateral calcarine sulcus and right cuneus. Furthermore, the neural activation of the left calcarine sulcus was positively correlated with visual analogue scale scores (r = 0.319, p = 0.043), and enhanced FC between the left thalamus and right cuneus in migraine patients was negatively correlated with Generalized Anxiety Disorder scores (r = − 0.617, p = 0.005). Conclusion: Our data suggest that migraine distress is exacerbated by aberrant feedback projections to the visual network, playing a crucial role in migraine physiological mechanisms. The current study provides further insights into the complex scenario of migraine mechanisms.

Objectives To characterize and interpret the CT imaging signs of the 2019 novel coronavirus (COVID-19) pneumonia in China. Materials and methods The CT images of 130 patients diagnosed as COVID-19 pneumonia from several hospitals in China were collected and their imaging features were analyzed and interpreted in detail. Results Among the 130 patients, we can see (1) distribution: 14 cases with unilateral lung disease and 116 cases with bilateral disease, the distribution was mainly lobular core (99 cases) and subpleural (102 cases); (2) number: 9 cases with single lesion, 113 cases with multiple lesions, and 8 cases with diffuse distribution; (3) density: 70 cases of pure ground glass opacity (GGO), and 60 cases of GGO with consolidation; (4) accompanying signs: vascular thickening (100 cases), “parallel pleura sign” (98 cases), “paving stone sign” (100 cases), “halo sign” (18 cases), “reversed halo sign” (6 cases), pleural effusion (2 cases), and pneumonocele (2 cases). After follow-up CT examination on 35 patients, 21 cases turned better and 14 became worse. There were signs of consolidation with marginal contraction, bronchiectasis, subpleural line, or fibrous streak. Conclusion GGO and consolidation are the most common CT signs of COVID-19 pneumonia, mainly with lobular distribution and subpleural distribution. The main manifestations were tissue organization and fibrosis at late stage. The most valuable features are the parallel pleura sign and the paving stone sign. Key Points • The CT signs of the COVID-19 pneumonia are mainly distributed in the lobular core, subpleural and diffused bilaterally. • The CT signs include the “parallel pleura sign,” “paving stone sign,” “halo sign,” and “reversed halo sign.” • During the follow-up, the distribution of lobular core, the fusion of lesions, and the organization changes at late stage will appear.

Purpose: Recently, high-resolution magnetic resonance imaging (HR-MRI) has been used to depict the wall characteristics of the intracranial arteries. The aim of this study was to explain the relationship between the remodeling patterns and acute ischemic stroke in patients with atherosclerotic middle cerebral artery (MCA) stenosis using HR-MRI. Material and methods: From August 2015 to May 2016, we prospectively screened 33 consecutive patients with unilateral MCA stenosis using time-to-flight (TOF) MR angiography, including 15 patients with symptomatic MCA stenosis and 18 patients with asymptomatic MCA stenosis. Among them, 14 patients were diagnosed as positive remodeling (PR) and 19 as negative remodeling or non-remodeling. The cross-sectional images of the stenotic MCA wall on HR-MRI including T1WI, T2WI, and PDWI were compared between the symptomatic group and the asymptomatic group as well as the PR group and the non-PR group, based on the vessel area, lumen area, wall area, plaque area, degree of stenosis, remodeling index, and NIHSS score. Results: The symptomatic group had larger wall area (P=0.040), plaque area (P<0.001), degree of stenosis (P=0.038), remodeling index (P<0.001), and NIHSS score (P=0.003) as well as smaller lumen area (P=0.001) than the asymptomatic group. In addition, more PR patients were observed in symptomatic group. The PR group had larger plaque area (P=0.014) and NIHSS score (P=0.037) than the non-PR group. Demographic and clinical characteristics between the symptomatic group and the asymptomatic group, the PR group and the non-PR group showed no statistical difference. Conclusions: The current study suggests that the HR-MRI has emerged as a promising tool to detect the characteristics of intracranial arteries wall and reveal the relationship between remodeling patterns and ischemic stroke. The PR is an unsafe remodeling way and is prone to cause acute ischemic stroke.