Predictive modeling with electronic health record (EHR) data is anticipated to drive personalized medicine and improve healthcare quality. Constructing predictive statistical models typically requires extraction of curated predictor variables from normalized EHR data, a labor-intensive process that discards the vast majority of information in each patient’s record. We propose a representation of patients’ entire raw EHR records based on the Fast Healthcare Interoperability Resources (FHIR) format. We demonstrate that deep learning methods using this representation are capable of accurately predicting multiple medical events from multiple centers without site-specific data harmonization. We validated our approach using de-identified EHR data from two US academic medical centers with 216,221 adult patients hospitalized for at least 24 h. In the sequential format we propose, this volume of EHR data unrolled into a total of 46,864,534,945 data points, including clinical notes. Deep learning models achieved high accuracy for tasks such as predicting: in-hospital mortality (area under the receiver operator curve [AUROC] across sites 0.93–0.94), 30-day unplanned readmission (AUROC 0.75–0.76), prolonged length of stay (AUROC 0.85–0.86), and all of a patient’s final discharge diagnoses (frequency-weighted AUROC 0.90). These models outperformed traditional, clinically-used predictive models in all cases. We believe that this approach can be used to create accurate and scalable predictions for a variety of clinical scenarios. In a case study of a particular prediction, we demonstrate that neural networks can be used to identify relevant information from the patient’s chart.
Long non-coding RNA plasmacytoma variant translocation 1 (PVT1) is up-regulated in various human cancers, and our results indicated that PVT1 was up-regulated in clear cell renal cell carcinoma tissues. The Cancer Genome Atlas cohort analysis revealed that in clear cell renal cell carcinoma, higher PVT1 expression correlated with advanced TNM stage, histological grade, and poor survival. PVT1 knockdown promoted apoptosis, inhibited renal cancer cell proliferation, decreased Mcl-1, and increased cleaved caspase-3 and cleaved PARP. PVT1 increased Mcl-1 mRNA levels in renal cancer cells by promoting mRNA stability without influencing its transcription. in vitro, the enhanced apoptosis arising from PVT1 suppression was attenuated by overexpressing Mcl-1. In addition, in vivo experiments showed that PVT1 knockdown repressed xenograft tumor growth, while Mcl-1 overexpression partially rescued xenograft tumor growth. These results indicate the PVT1/Mcl-1 pathway inhibits renal cancer cell apoptosis in vitro and in vivo. PVT1 may thus serve as a novel biomarker, and the PVT1/Mcl-1 pathway may be a useful therapeutic target for clear cell renal cell carcinoma.
Follicular helper CD4+ T (TFH) cells are critical for optimal B-cell-mediated humoral immunity by initiating, fueling, and sustaining germinal center reactions. The differentiation of TFH cells relies on multiple intrinsic and extrinsic factors; however, the details by which these factors are integrated to coordinate TFH differentiation are largely unknown. In this study, using a mouse model of acute lymphocytic choriomeningitis virus (LCMV) viral infection, we demonstrate that mTOR complex 2 (mTORC2) kinase integrates TCR signaling and ICOS-mediated co-stimulation to promote late differentiation and functional maturation of virus-specific TFH cells. Specifically, mTORC2 functions to maintain TFH lineage specifications, including phenotypes, migratory characteristics, and functional properties. Thus, our results highlight the importance of mTORC2 in guarding TFH phenotypic and functional maturation.
BackgroundSonodynamic therapy (SDT) is an emerging tumor-inhibiting method that has gained attention in cancer therapy in the last several years. Although autophagy has been observed in SDT-treated cancer cells, its role and mechanism of action remain unclear. This study aimed to investigate the effects of low-frequency ultrasound on autophagy and drug-resistance of paclitaxel (PTX)-resistant PC-3 cells via the endoplasmic reticulum stress (ERs)-mediated PI3K/AT/mTOR signaling pathway.MethodsCCK-8 assay was conducted to select the appropriate exposure time for PTX-resistant PC-3 cells under low-frequency ultrasound. PTX-resistant PC-3 cells were divided into a control group, PTX group, ultrasound group, ultrasound + PTX group, ultrasound + PTX + autophagy-related gene 5 (Atg5) siRNA group, and ultrasound + 4-PBA (an ERs inhibitor) group. Autophagy was observed by transmission electron microscopy (TEM) and fluorescence microscopy. Cell proliferation was evaluated using CCK-8 assay; apoptosis was detected by flow cytometry. Expression of multiple drug-resistance genes was detected by qRT-PCR. Western blotting was used to detect the expression of ERS-related proteins, autophagy-related proteins, apoptosis-related proteins, and PI3K/AKT/mTOR pathway-related proteins.ResultsTen-second exposure was selected as optimal for all experiments. Compared to the PTX group, the level of autophagy, inhibition rate, apoptosis rate, and expression of ERS-related proteins (GRP78) increased, whereas the expression of multiple drug-resistance genes (MRP3, MRP7, and P-glycoprotein), PI3K/AKT/mTOR pathway-related proteins (PI3K, p-AKT, mTORC1), and apoptosis-related proteins (Bcl-2, NF-κB) decreased in PTX-resistant PC-3 cells after low-frequency ultrasound and PTX treatment for 24 h. These trends were more obvious after treatment with Atg5 siRNA, excluding the autophagy level. Post 4-PBA-treatment, the expression of GRP78 and LC3II proteins decreased, whereas that of PI3K, p-AKT, and mTORC1 increased.ConclusionResults indicated that ultrasound induces autophagy by ERs-mediated PI3K/AKT/mTOR signaling pathway in PTX-resistant PC-3 cells; this autophagy acts as a cytoprotector during low-frequency ultrasound-mediated reversal of drug resistance.
In most of the cases, scientists depend on previous literature which is relevant to their research fields for developing new ideas. However, it is not wise, nor possible, to track all existed publications because the volume of literature collection grows extremely fast. Therefore, researchers generally follow, or cite merely a small proportion of publications which they are interested in. For such a large collection, it is rather interesting to forecast which kind of literature is more likely to attract scientists' response. In this paper, we use the citations as a measurement for the popularity among researchers and study the interesting problem of Citation Count Prediction (CCP) to examine the characteristics for popularity. Estimation of possible popularity is of great significance and is quite challenging. We have utilized several features of fundamental characteristics for those papers that are highly cited and have predicted the popularity degree of each literature in the future. We have implemented a system which takes a series of features of a particular publication as input and produces as output the estimated citation counts of that article after a given time period. We consider several regression models to formulate the learning process and evaluate their performance based on the coefficient of determination (R 2 ). Experimental results on a real-large data set show that the best predictive model achieves a mean average predictive performance of 0.740 measured in R 2 , which significantly outperforms several alternative algorithms.
No abstract
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
334 Leonard St
Brooklyn, NY 11211
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.