Bioinformatics analysis of gene expression profile of central nervous system primitive neuroectodermal tumors / 西安交通大学学报(医学版)

【Objective】 To analyze the gene expression profile of central nervous system primitive neuroectodermal tumors （CNS-PNETs） by bioinformatics methods so as to explore the possible pathogenesis of CNS-PNETs at the molecular level. 【Methods】 The gene expression profile of CNS-PNETs was downloaded from the GEO database， GSE35493 and GSE74195. The differentially expressed genes （DEGs） were screened by the online analysis tool of GEO2R and Venn software， DEGs were analyzed by using the online analysis tools of David database， such as Gene Ontology （GO） and pathway enrichment （KEGG）. The protein interaction network analysis （PPI） of CNS-PNETs was made by using STRING online analysis tool， Cytoscape software and its plug-in cytohubba to find the key genes. 【Results】 We obtained 262 DEGs， including 49 upregulated genes and 213 downregulated genes. The analysis of GO function and KEGG signal pathway enrichment showed that DEG was involved in DNA transcription and mitosis， cell division， synaptic signal transmission and other biological processes， and associated with cell cycle， tumor-related pathway， p53 signal pathway， synapsis-related signal pathway， cAMP signal pathway and calcium ion signal pathway. Ten key genes， namely， CDK1， CDC20， MAD2L1， KIF11， ASPM， TOP2A， TTK， NDC80， NUSAP1 and DLGAP5 were screened out by STRING analysis. 【Conclusion】 Ten key genes including CDK1 may play an important role in the initiation and progression of CNS-PNETs， providing new clues for exploring the pathogenesis of CNS-PNETs.

The study on the segmentation of carotid vessel wall in multicontrast MR images based on U-Net neural network / 中华放射学杂志

Objective@#To investigate the value of automatic segmentation of carotid vessel wall in multicontrast MR images using U-Net neural network.@*Methods@#Patients were retrospectively collected from 2012 to 2015 in Carotid Atherosclerosis Risk Assessment (CARE II) study. All patients who recently suffered ischemic stroke and/or transient ischemic attack underwent identical, state-of-the-art multicontrast MRI technique. A total of 17 568 carotid vessel wall MR images from 658 subjects were included in this study after inclusion criteria and exclusion criteria. All MR images were analyzed using customized analysis platform (CASCADE). Randomly, 10 592 images were assigned into training dataset, 3 488 images were assigned into validating dataset and 3 488 images were assigned into test dataset according to a ratio of 6∶2∶2. Data augmentation was performed to avoid over fitting and improve the ability of model generalization. The fine-tuned U-Net model was utilized in the segmentation of carotid vessel wall in multicontrast MR images. The U-Net model was trained in the training dataset and validated in the validating dataset. To evaluate the accuracy of carotid vessel wall segmentation, the sensitivity, specificity and Dice coefficient were used in the testing dataset. In addition, the interclass correlation and the Bland-Altman analysis of max wall thickness and wall area were obtained to demonstrate the agreement of the U-Net segmentation and the manual segmentation.@*Results@#The sensitivity, specificity and Dice coefficient of the fine-tuned U-Net model achieved 0.878,0.986 and 0.858 in the test dataset, respectively. The interclass correlation (95% confidence interval) was 0.921 (0.915-0.925) for max wall thickness and 0.929 (0.924-0.933) for wall area. In the Bland-Altman analysis, the difference of max wall thickness was (0.037±0.316) mm and the difference of wall area was (1.182±4.953) mm2. The substantial agreement was observed between U-Net segmentation method and manual segmentation method.@*Conclusion@#Automatic segmentation of carotid vessel wall in multicontrast MR images can be achieved using fine-tuned U-Net neural network, which is trained and tested in the large scale dataset labeled by professional radiologists.

The study on the segmentation of carotid vessel wall in multicontrast MR images based on U?Net neural network / 中华放射学杂志

Objective To investigate the value of automatic segmentation of carotid vessel wall in multicontrast MR images using U?Net neural network. Methods Patients were retrospectively collected from 2012 to 2015 in Carotid Atherosclerosis Risk Assessment (CARE II) study. All patients who recently suffered ischemic stroke and/or transient ischemic attack underwent identical, state?of?the?art multicontrast MRI technique. A total of 17 568 carotid vessel wall MR images from 658 subjects were included in this study after inclusion criteria and exclusion criteria. All MR images were analyzed using customized analysis platform (CASCADE). Randomly, 10 592 images were assigned into training dataset, 3 488 images were assigned into validating dataset and 3 488 images were assigned into test dataset according to a ratio of 6∶2∶2. Data augmentation was performed to avoid over fitting and improve the ability of model generalization. The fine?tuned U?Net model was utilized in the segmentation of carotid vessel wall in multicontrast MR images. The U?Net model was trained in the training dataset and validated in the validating dataset. To evaluate the accuracy of carotid vessel wall segmentation, the sensitivity, specificity and Dice coefficient were used in the testing dataset. In addition, the interclass correlation and the Bland?Altman analysis of max wall thickness and wall area were obtained to demonstrate the agreement of the U?Net segmentation and the manual segmentation. Results The sensitivity, specificity and Dice coefficient of the fine?tuned U?Net model achieved 0.878,0.986 and 0.858 in the test dataset, respectively. The interclass correlation (95% confidence interval) was 0.921 (0.915-0.925) for max wall thickness and 0.929 (0.924-0.933) for wall area. In the Bland?Altman analysis, the difference of max wall thickness was (0.037±0.316) mm and the difference of wall area was (1.182±4.953) mm2. The substantial agreement was observed between U?Net segmentation method and manual segmentation method. Conclusion Automatic segmentation of carotid vessel wall in multicontrast MR images can be achieved using fine?tuned U?Net neural network, which is trained and tested in the large scale dataset labeled by professional radiologists.

Clinical application value and research progress of artificial pancreas closed-Loop control in diabctes mellitus / 中国综合临床

Ideal blood glucose control requires accurate insulin injections under the guidance of frequent glucose monitoring.Artificial pancreas (AP),the closed-loop control system can adjust the input amount of insulin automatically with the body's blood glucose levels.The AP allows diabetics to control blood glucose ideal,then get the benefit of prevention of complications and bring convenience and safety in clinical application.Accuracy is the key issue of the AP.To improve the accuracy of such a system need to improve the detection accuracy and reliability,increase speed and accuracy of the output control,and improve the accuracy of the system regulation model.