Objective To observe the efficacy of self-guided attention network for detecting responsible lesions related to cerebral palsy(CP)in children with periventricular white matter injury(PVWMI).Methods Totally 383 children with PVWMI were retrospectively enrolled and divided into CP group(n=243)and non-CP group(n=140),while 214 children without obvious brain abnormality on brain MRI were taken as control group.ROI of 4 key anatomical structures related to CP,i.e.centrum semiovale,posterior limb of internal capsule,cerebral peduncle and thalamus were delineated on T1WI,while responsible lesions related to CP within the key anatomical structures were labeled on T2WI,and the images were then registrated and used as input of the networks.ResNet34 network was adopted combined with attention and self-guided networks to train the network for detecting responsible lesions related to CP in children with PVWMI,and their efficacies were evaluated.The optimal network was screened,and its efficacy for segmenting the key anatomical structures was evaluated.Results Self-guided attention network was the optimal network,its area under the curve(AUC)for detecting lesions was 0.794-0.914,and the Dice similarity coefficient for segmenting the key anatomical structures was 0.702-0.764.Conclusion Self-guided attention network could be used for preliminarily detecting responsible lesions related to CP in children with PVWMI.

Objective To observe the efficacy of self-guided attention network for detecting responsible lesions related to cerebral palsy(CP)in children with periventricular white matter injury(PVWMI).Methods Totally 383 children with PVWMI were retrospectively enrolled and divided into CP group(n=243)and non-CP group(n=140),while 214 children without obvious brain abnormality on brain MRI were taken as control group.ROI of 4 key anatomical structures related to CP,i.e.centrum semiovale,posterior limb of internal capsule,cerebral peduncle and thalamus were delineated on T1WI,while responsible lesions related to CP within the key anatomical structures were labeled on T2WI,and the images were then registrated and used as input of the networks.ResNet34 network was adopted combined with attention and self-guided networks to train the network for detecting responsible lesions related to CP in children with PVWMI,and their efficacies were evaluated.The optimal network was screened,and its efficacy for segmenting the key anatomical structures was evaluated.Results Self-guided attention network was the optimal network,its area under the curve(AUC)for detecting lesions was 0.794-0.914,and the Dice similarity coefficient for segmenting the key anatomical structures was 0.702-0.764.Conclusion Self-guided attention network could be used for preliminarily detecting responsible lesions related to CP in children with PVWMI.

Objective To identify the potential differently expressed genes and microRNAs(mi-RNAs)in sepsis survivors and non-survivors through bioinformatics-based research based on gene expression omnibus(GEO).Methods Two gene expression profile microarray datasets of human blood samples(GSE48080 and GSE54514)were downloaded from the GEO database.The differential expression genes(DEGs)between sepsis survivors and non-survivors at two time points(diagnosis of sepsis,course of sep-sis)were screened with the GEO2R online tool.The gene ontology(GO)and Kyoto encyclopedia of genes and genomes(KEGG)pathway enrichment analysis were used to study the pathophysiological processes and potential signaling pathways involved in sepsis related death DEGs.STRING online tool was used to construct the DEGs protein-protein interaction(PPI).Cytoscape with CytoHubba was used to investigate the potential hub genes.NetworkAnalyst was used to construct targeted miRNAs of the hub genes.Real-time quantitative PCR(RT-qPCR)was established to evaluate the expression of potential hub genes in our sepsis survivors and non-survivors.Results During the course of sepsis,there was heterogeneity in gene expre-ssion between sepsis survivors and non-survivors.Fifteen DEGs were found to be remarkably differentially expressed between sepsis survivors and non-survivors during the course of sepsis.Four KEGG pathways,in-cluding staphylococcus aureus infection,NOD-like receptor signaling pathway,sulfur metabolism and col-lecting duct acid secretion,were significantly enriched.In combination with the results of the PPI network and CytoHubba,ten hub genes(SLC4A1,EPB42,LTF,LCN2,DEFA4,HBM,HBG1,GMPR,CAMP,OLFM4)were selected as potential biomarkers for sepsis-associated mortality.With NetworkAnalyst analysis,ten miRNAs were predicted as potential key miRNAs.RT-qPCR confirmed that the expressions of five of these genes(SLC4A1,EPB42,LCN2,DEFA4,OLFM4)were in accordance with the microarray results.Conclusion Bioinformatics analysis based on GEO database showed DEGs between sepsis suvivors and non-survivors in the course of sepsis,which contributed to identification of potential biomarkers and risk factors for sepsis-associated mortality.