Project description:Purpose: To explore the possible pathogenesis of venous thromboembolism from the perspective of circRNAs. Methods: The thrombus model of inferior vena cava in SD rats was established by stenosis method. Peripheral blood samples of DVT rats were collected at 1h, 6h,12h and 3d, which were sequenced with Illumina circRNAs and mRNA. The differential circRNAs and mRNA were analyzed by STEM analysis software. Clusters that can reflect the evolution of thrombus were selected for in-depth analysis, and circRNAs and mRNA were screened according to the pathway enrichment analysis of the selected clusters. Real time fluorescent quantitative PCR (qPCR) was used to verify and determine the candidate circRNAs and mRNA. The function of target circRNA was verified in DVT rats. The interacting miRNAs of circRNAs and mRNA were predicted by online database respectively. Candidate miRNAs were identified by integrating the related information of miRNAs and verified by qPCR. Results:The SD rat model of deep venous thrombosis was successfully established by the stenosis method. In model group, there were 736, 3691, 3406 and 2639 circRNAs up-regulated and 944, 327, 318 and 397 circRNAs down-regulated at 1h, 6h, 12h and 3d, respectively. Compared with Sham group, 169, 885, 298 and 500 mRNA were up-regulated and 231, 291,75 and 73 mRNA were down-regulated at 1h, 6h, 12h and 3d in model group, respectively. STEM analysis found that cluster 45 was similar to thrombus development. The pathway related to platelet function was selected in cluster 45, and 10 circrRNAs and 4 interacting mRNAs were selected for qPCR verification by combining the gene in platelet pathway and annotation gene of cluster 45 and mRNA-circRNA interaction network. According to qPCR results and whether ID is found in rat circRNA database (CIRCpedia-V2), CBT15_circR_3235 (ID in circpedia-v2: rno_CIRCpedia_1953) is determined as the target circRNA. In vivo，knockdown of rno_CIRCpedia_1953 attenuates the formation of DVT in rats. The potential target miRNAs of rno_CIRCpedia_1953 and Tuba4a were predicted by using online miRDB and miRWalk databases respectively, and rno-miR-3543, rno-miR-320-5p and rno-miR-26b-3p are preliminarily identified as potential target miRNAs that rno_CIRCpedia_1953 may combine to play the ceRNA mechanism. Conclusions: Animal model of DVT in rats was successfully established by stenosis method. High-throughput sequencing analysis of circRNAs showed that circRNAs participated in the evolution of DVT. According to STEM functional analysis of mRNA sequencing, a target circRNA rno_CIRCpedia _1953 was screened, verified and identified. rno_CIRCpedia_1953 may participate in the development of DVT by acting as miRNA molecular sponge.

Project description:BackgroundAn increase in the incidence of central venous catheter (CVC)-associated deep venous thrombosis (CADVT) has been reported in pediatric patients over the past decade. At the same time, current screening guidelines for venous thromboembolism risk have low sensitivity for CADVT in hospitalized children. This study utilized a multimodal deep learning model to predict CADVT before it occurs.MethodsChildren who were admitted to intensive care units (ICUs) between December 2015 and December 2018 and with CVC placement at least 3 days were included. The variables analyzed included demographic characteristics, clinical conditions, laboratory test results, vital signs and medications. A multimodal deep learning (MMDL) model that can handle temporal data using long short-term memory (LSTM) and gated recurrent units (GRUs) was proposed for this prediction task. Four benchmark machine learning models, logistic regression (LR), random forest (RF), gradient boosting decision tree (GBDT) and a published cutting edge MMDL, were used to compare and evaluate the models with a fivefold cross-validation approach. Accuracy, recall, area under the ROC curve (AUC), and average precision (AP) were used to evaluate the discrimination of each model at three time points (24 h, 48 h and 72 h) before CADVT occurred. Brier score and Spiegelhalter's z test were used measure the calibration of these prediction models.ResultsA total of 1830 patients were included in this study, and approximately 15% developed CADVT. In the CADVT prediction task, the model proposed in this paper significantly outperforms both traditional machine learning models and existing multimodal deep learning models at all 3 time points. It achieved 77% accuracy and 90% recall at 24 h before CADVT was discovered. It can be used to accurately predict the occurrence of CADVT 72 h in advance with an accuracy of greater than 75%, a recall of more than 87%, and an AUC value of 0.82.ConclusionIn this study, a machine learning method was successfully established to predict CADVT in advance. These findings demonstrate that artificial intelligence (AI) could provide measures for thromboprophylaxis in a pediatric intensive care setting.

Dataset Information

Resolution of Deep Venous Thrombosis: Proposed Immune Paradigms.

Publications

Resolution of Deep Venous Thrombosis: Proposed Immune Paradigms.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets