Long short-term memory and Logistic regression for mortality risk prediction of intensive care unit patients with stroke / 北京大学学报(医学版)

OBJECTIVE@#To select variables related to mortality risk of stroke patients in intensive care unit (ICU) through long short-term memory (LSTM) with attention mechanisms and Logistic regression with L1 norm, and to construct mortality risk prediction model based on conventional Logistic regression with important variables selected from the two models and to evaluate the model performance.@*METHODS@#Medical Information Mart for Intensive Care (MIMIC)-Ⅳ database was retrospectively analyzed and the patients who were primarily diagnosed with stroke were selected as study population. The outcome was defined as whether the patient died in hospital after admission. Candidate predictors included demogra-phic information, complications, laboratory tests and vital signs in the initial 48 h after ICU admission. The data were randomly divided into a training set and a test set for ten times at a ratio of 8 ∶2. In training sets, LSTM with attention mechanisms and Logistic regression with L1 norm were constructed to select important variables. In the test sets, the mean importance of variables of ten times was used as a reference to pick out the top 10 variables in each of the two models, and then these variables were included in conventional Logistic regression to build the final prediction model. Model evaluation was based on the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and accuracy. And the model performance was compared with the forward Logistic regression model which hadn't conducted variable selection previously.@*RESULTS@#A total of 2 755 patients with 2 979 ICU admission records were included in the analysis, of which 526 recorded deaths. The AUC of Logistic regression model with L1 norm was statistically better than that of LSTM with attention mechanisms (0.819±0.031 vs. 0.760±0.018, P < 0.001). Age, blood glucose, and blood urea nitrogen were at the top ten important variables in both of the two models. AUC, sensitivity, specificity, and accuracy of Logistic regression models were 0.85, 85.98%, 71.74% and 74.26%, respectively. And the final prediction model was superior to forward Logistic regression model.@*CONCLUSION@#The variables selected by Logistic regression with L1 norm and LSTM with attention mechanisms had good prediction performance, which showed important implications on the mortality prediction of stroke patients in ICU.

MicroRNA-340 suppresses cell proliferation and induces cell apoptosis in hepatocellular carcinoma / 第二军医大学学报

Objective To investigate the expression of microRNA-340 (miR-340) in hepatocellular carcinoma (HCC) and its effect on cell biological behavior. Methods We collected 40 frozen HCC tissues and adjacent non-tumor tissues from patients undergoing hepatectomy of HCC at The First Affiliated Hospital of Chongqing Medical University from Mar. 2015 to Sep. 2016. The expression of miR-340 in all tissues was detected by qPCR and the relationship between miR-340 expression and clinicopathological parameters was analyzed. Simultaneously, the expression of miR-340 in normal hepatocyte (HL-7702) and four hepatoma cells lines (Hep3B, Bel-7402, HepG2, SMMC-7721) was detected by qPCR after incubation for 48 h. The eukaryotic expression vector with miR-340 or control reagent was transfected into SMMC-7721 cells using EndoFection™-Max to increase or inhibit the expression of miR-340, and then the cells were cultured for 24 h, 48 h and 72 h. The proliferation of SMMC-7721 cells was detected by CCK-8 assay, and the apoptosis was detected by flow cytometry. The target gene of miR-340 was predicted by bioinformatics software, and the effect of miR-340 on target gene was further verified by qPCR and Western blotting. Results The expression of miR-340 in HCC tissues was significantly lower than that in the adjacent non-tumor tissues (P〈0. 01), and was correlated with hepatitis B surface antigen, HBV DNA, tumor size and TNM stage (all P〈0. 01). Besides, the expression of miR-340 in HL-7702 cells was significantly higher than that in Hep3B, Bel-7402, HepG2 and SMMC-7721 cells (P〈0. 05, P〈0. 01). CCK-8 assay results showed that overexpression of miR-340 inhibited proliferation of SMMC-7721 cells, while inhibition of miR-340 promoted cell proliferation (P〈0. 05, P〈0. 01). Overexpression of miR-340 significantly promoted SMMC-7721 cells apoptosis, while suppression of miR-340 significantly inhibited cells apoptosis (all P〈0. 01). S-phase kinase-associated protein 2 (SKP2) was a target gene of miR-340 as indicated by bioinformatics software. Further, qPCR and Western blotting results showed that overexpression of miR-340 inhibited the mRNAand protein expression of SKP2, while inhibition of miR-340 increased the mRNA and protein expression of SKP2. Conclusion The abnormal expression of miR-340 may be associated with the HBV infection, and miR-340 may be an indicator to evaluate the progression and prognosis of HCC. MiR-340 can inhibk proliferation and promote apoptosis of SMMC-7721 cells, which may be effected by inhibiting the SKP2 expression.

Regulation of SIRT1 by SREBP-1 in Adipogenic Differentiation of Mesenchymal Stem Cells / 中国实验血液学杂志

<p><b>OBJECTIVE</b>To study the regulation of SIRT1 by transcription factor SREBP-1 in adipogeneic differentiation of bone marrow mesenchymal stem cells (BMMSC).</p><p><b>METHODS</b>Oil red O staining was used to identify the adipogenic differentiation of BMMSC; the mRNA transcription levels of AP2, LPL, SREBF-1, SIRT1 gene were detected by RT-PCR; the expession level of SREBP-1 was determined by Western-blot. The chromatin immunoprecipitation (ChIP) assay was used to investigate the binding of SREBP-1 to SIRT1 promoter.</p><p><b>RESULTS</b>BMMSC exposed to adipogenesis inducing medium become mature adipocytes at day 14; the mRNA transcription levels of AP2, LPL, SREBF-1, SIRT1 genes were up-regulated in adipocyte differentiation of BMMSC; the protein level of SREBP-1 was higher obviously; SIRT1 gene sequences was succesfully amplified from the genomic DNA immunoprecipitated by SREBP-1 antibody.</p><p><b>CONCLUSION</b>SREBP-1 can bind to the promoter region of the SIRT1 gene in adipogenesis of BMMSC, and may be involved in the transcriptional regulation of the SIRT1 gene.</p>