Objective:Construction of deep learning classification models based on functional magnetic resonance imaging (fMRI) data assists the clinicians to achieve better diagnosis of bipolar disorder (BD), which can improve the recognition rate of BD by identifying the critical imaging features.Methods:A total of 146 patients who met the diagnosis criteria of BD according to DSM-Ⅳ and 234 healthy control (HC) were recruited for fMRI scans. Regional homogeneity (ReHo) and amplitude of low frequency fluctuation (ALFF) were used to analyze fMRI data. Based on ReHo and ALFF, the classification models were constructed by deeping neural network (DNN) and dual-channel convolution neural network (DCNN) respectively, and the best classification model was developed by comparing the accuracy and area under curve (AUC) of the two models. Based on each brain region divided by anatomical automatic labeling (AAL), the support vector machine (SVM) classification model was constructed using imaging index with a better performance, and the critical imaging features were identified by comparing the accuracy of each brain region.Results:The performances of the DCNN classification model (accuracy = 75.3%, and 72.6%, respectively, based on ReHo and ALFF) were significantly better than the DNN classification model (accuracy = 67.1%, and 65.1%, respectively). Meanwhile, the accuracy of classification model constructed using ReHo was higher than ALFF. Based on the SVM classification model, critical brain regions were identified above the accuracy of 65.0%, including the occipital lobe (middle occipital gyrus, superior occipital gyrus and lingual gyrus), hippocampus, and thalamus.Conclusion:The computational model based on DCNN using ReHo can help the clinicians to achieve better diagnosis of BD. Furthermore, occipital lobe, hippocampus and thalamus may be the critical imaging features for the auxiliary recognition of BD.

Objective:Construction of deep learning classification models based on functional magnetic resonance imaging (fMRI) data assists the clinicians to achieve better diagnosis of bipolar disorder (BD), which can improve the recognition rate of BD by identifying the critical imaging features.Methods:A total of 146 patients who met the diagnosis criteria of BD according to DSM-Ⅳ and 234 healthy control (HC) were recruited for fMRI scans. Regional homogeneity (ReHo) and amplitude of low frequency fluctuation (ALFF) were used to analyze fMRI data. Based on ReHo and ALFF, the classification models were constructed by deeping neural network (DNN) and dual-channel convolution neural network (DCNN) respectively, and the best classification model was developed by comparing the accuracy and area under curve (AUC) of the two models. Based on each brain region divided by anatomical automatic labeling (AAL), the support vector machine (SVM) classification model was constructed using imaging index with a better performance, and the critical imaging features were identified by comparing the accuracy of each brain region.Results:The performances of the DCNN classification model (accuracy = 75.3%, and 72.6%, respectively, based on ReHo and ALFF) were significantly better than the DNN classification model (accuracy = 67.1%, and 65.1%, respectively). Meanwhile, the accuracy of classification model constructed using ReHo was higher than ALFF. Based on the SVM classification model, critical brain regions were identified above the accuracy of 65.0%, including the occipital lobe (middle occipital gyrus, superior occipital gyrus and lingual gyrus), hippocampus, and thalamus.Conclusion:The computational model based on DCNN using ReHo can help the clinicians to achieve better diagnosis of BD. Furthermore, occipital lobe, hippocampus and thalamus may be the critical imaging features for the auxiliary recognition of BD.

OBJECTIVE: To investigate the differentially expressed microRNAs(miRNAs) in human embryonic lung fibroblast MRC-5 cells stimulated by transforming growth factor-β1(TGF-β1) using microarray chip, and screen for key genes and signaling pathways of fibroblast trans-differentiation. METHODS: The miRNA expression gene chip dataset GSE43992 on TGF-β1 stimulated MRC-5 cells were downloaded from high-throughput Gene Expression Omnibus(GEO) database of National Center for Biotechnology Information of the United States. The R language Limma package was used to screen the differentially expressed miRNAs. Corresponding target genes were predicted by miRWalk database performed by Gene Ontology(GO) functional enrichment analysis and Kyoto Encyclopedia of Genes and Genomes(KEGG) signaling pathway enrichment analysis. The protein-protein interaction(PPI) network was constructed by the search tool for the Retrieval of Interacting Genes database. RESULTS: A total of five differentially expressed miRNAs were identified, including four up-regulated miRNAs and one down-regulated miRNA; and 42 corresponding differentially expressed target genes were predicted. GO analysis indicated that the target genes were significantly enriched in collagen catabolic process, extracellular matrix organization, membrane organization, collagen fibril organization, and cellular response to amino acid stimulus. The results of KEGG pathway analysis showed that the signaling pathways corresponding to miRNAs and target genes were mainly concentrated in 18 signaling pathways, that were mainly related to the age-ethnic signaling pathways and protein digestion and absorption miRNAs in tumors and diabetic complications. The core genes transfected into the myofibroblasts by the three fibroblasts screened by the PPI network were threonine kinase 1, estrogen receptor 1 and β-catenin. CONCLUSION: Five differentially expressed miRNAs, 42 target genes, 18 signaling pathways, and 3 core genes related to TGF-β1-induced MRC-5 cell trans-differentiation were screened. It can provide new reference for the treatment and research of many diseases including pneumoconiosis and pulmonary fibrosis.