RESUMO
Featuring a burning sensation in the tongue or other oral sites in the absence of observable lesions or laboratory findings, burning mouth syndrome (BMS) is a chronic intraoral pain disorder, which is one of the most common medically unexplained oral symptoms/syndromes. Previous studies have suggested that brain changes are involved in BMS; however, the small number of participants in these studies limited the conclusions that could be drawn. The present study aimed to further elucidate the brain anatomical and functional changes in BMS with a relatively large sample. Fifty-three patients (26 BMS patients and 27 gender- and age-matched controls) were recruited. Demographic information was collected via interviews. Visual analogue scale (VAS), anxiety, and depression scale were administered. Participants underwent an MRI scan (including one high-resolution structural scan, one diffusion tensor image, and one session of resting state scan) on the same day. The results showed that BMS patients had higher depression and anxiety levels than controls. BMS patients showed lower gray matter volume (GMV) in the bilateral ventromedial prefrontal cortex (VMPFC) and increased functional connectivity between this region and the bilateral amygdala. Region of interest (ROI) analysis suggested that the functional connectivity between the bilateral VMPFC and amygdala correlated with the years of BMS illness in patients. The brain measures could predict the years of symptoms in the BMS group. These results suggest A potential neuromarker for the diagnosis and treatment of BMS.
RESUMO
OBJECTIVE: To explore the feasibility and efficacy of artificial neural network for differentiating high-grade glioma and low-grade glioma using image information.â© Methods: A total of 130 glioma patients with confirmed pathological diagnosis were selected retrospectively from 2012 to 2017. Forty one imaging features were extracted from each subjects based on 2-dimension magnetic resonance T1 weighted imaging with contrast-enhancement. An artificial neural network model was created and optimized according to the performance of feature selection. The training dataset was randomly selected half of the whole dataset, and the other half dataset was used to verify the performance of the neural network for glioma grading. The training-verification process was repeated for 100 times and the performance was averaged.â© Results: A total of 5 imaging features were selected as the ultimate input features for the neural network. The mean accuracy of the neural network for glioma grading was 90.32%, with a mean sensitivity at 87.86% and a mean specificity at 92.49%. The area under the curve of receiver operating characteristic curve was 0.9486.â© Conclusion: As a technique of artificial intelligence, neural network can reach a relatively high accuracy for the grading of glioma and provide a non-invasive and promising computer-aided diagnostic process for the pre-operative grading of glioma.