Objective A vast amount of medical data is still stored in unstructured text documents. We present an automated method of information extraction from German unstructured clinical routine data from the cardiology domain enabling their usage in state-of-the-art data-driven deep learning projects. Methods We evaluated pre-trained language models to extract a set of 12 cardiovascular concepts in German discharge letters. We compared three bidirectional encoder representations from transformers pre-trained on different corpora and fine-tuned them on the task of cardiovascular concept extraction using 204 discharge letters manually annotated by cardiologists at the University Hospital Heidelberg. We compared our results with traditional machine learning methods based on a long short-term memory network and a conditional random field. Results Our best performing model, based on publicly available German pre-trained bidirectional encoder representations from the transformer model, achieved a token-wise micro-average F1-score of 86% and outperformed the baseline by at least 6%. Moreover, this approach achieved the best trade-off between precision (positive predictive value) and recall (sensitivity). Conclusion Our results show the applicability of state-of-the-art deep learning methods using pre-trained language models for the task of cardiovascular concept extraction using limited training data. This minimizes annotation efforts, which are currently the bottleneck of any application of data-driven deep learning projects in the clinical domain for German and many other European languages.
The downregulation of β-adrenergic receptors (β-AR) and decreased cAMP-dependent protein kinase activity in failing hearts results in decreased phosphorylation and inactivation of phosphatase-inhibitor-1 (I-1), a distal amplifier element of β-adrenergic signaling, leading to increased protein phosphatase 1 activity and dephosphorylation of key phosphoproteins, including phospholamban. Downregulated and hypophosphorylated I-1 likely contributes to β-AR desensitization; therefore its modulation is a promising approach in heart failure treatment. Aim of our study was to assess the effects of adeno-associated virus serotype 9 (AAV9) - mediated cardiac-specific expression of constitutively active inhibitor-1 (I-1c) and to investigate whether I-1c is able to attenuate the development of heart failure in mice subjected to transverse aortic constriction (TAC). 6-8 week old C57BL/6 N wild-type mice were subjected to banding of the transverse aorta (TAC). Two days later 2.8 × 10 AAV-9 vector particles harbouring I-1c cDNA under transcriptional control of a human troponin T-promoter (AAV9/I-1c) were intravenously injected into the tail vein of these mice (n=12). AAV9 containing a Renilla luciferase reporter (AAV9/hRluc) was used as a control vector (n=12). Echocardiographic analyses were performed weekly to evaluate cardiac morphology and function. 4 weeks after TAC pressure- volume measurements were performed and animals were sacrificed for histological and molecular analyses. Both groups exhibited progressive contractile dysfunction and myocardial remodeling. Surprisingly, echocardiographic assessment and histological analyses showed significantly increased left ventricular hypertrophy in AAV9/I-1c treated mice compared to AAV9/hRluc treated controls as well as reduced contractility. Pressure-volume loops revealed significantly impaired contractility after AAV9/I-1c treatment. At the molecular level, hearts of AAV9/I-1c treated TAC mice showed a hyperphosphorylation of the SR Ca-ATPase inhibitor phospholamban. In contrast, expression of AAV9/I-1c in unchallenged animals resulted in selective enhancement of phospholamban phosphorylation and augmented cardiac contractility. Our data suggest that AAV9-mediated cardiac-specific overexpression of I-1c, previously associated with enhanced calcium cycling, improves cardiac contractile function in unchallenged animals but failed to protect against cardiac remodeling induced by hemodynamic stress questioning the use of I-1c as a potential strategy to treat heart failure in conditions with increased afterload.
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
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.