As the largest organ of the body, the human skin protects all subcutaneous tissues. Despite its many attributes, the skin is vulnerable to pressure ulcers. The number of pressure ulcers and venous leg ulcers is on the rise, but healing rates have not improved over the past decade. The reason may be a tendency to focus on one or two fundamentals of wound healing, but not on all 3 fundamentals equally. The 3 fundamentals of wound healing are (1) pressure relief and nursing care, (2) dressings, and (3) nutrition. Nutrition is the area that is most often overlooked, which commonly causes the care plan to be out of balance. In the United States, few clinicians would consider malnutrition to be an issue in the homecare and long-term care setting, yet nutritional status and risk for pressure ulcer formation are well documented and strongly correlated. Our aging population will continue to survive previously catastrophic events, only to present with pressure ulcers or the potential for developing pressure ulcers. Clinicians caring for residents with pressure ulcers must strike a balance between pressure relief, dressings, and nutrition. Functional gastrointestinal-tract impairments must be diagnosed and addressed. Wounds must be treated aggressively with high-protein calorically-balanced diets because wounds heal from the inside out.
Introduction Artificial intelligence (AI) can be used for various tasks in medicine and specifically in cardiology. Medical data such as electrocardiogram recordings (ECGs) are widely used and universally accepted as diagnostic and prognostic tools. It has been shown that deep learning methods using ECGs yield excellent results detecting cardiac pathologies. A significant amount of reliable data is required for supervised learning algorithms such as deep learning models. However, only a small fraction of ECG data generated in daily practice is available in a fully digital and machine-readable format, such as XML. Frequently, used ECG devices produce PDF files or even paper-based print outs, which need to be digitised later for inclusion in clinical information systems. Such ECGs cannot be used without further effort for training or application of deep learning models. Therefore, aim of the present project was to develop a data-pipeline that generates machine-readable ECG data for AI use data irrespective of the initial ECG format. Methods We propose an end-to-end pipeline that can not only process data from modern digital ECG devices but is also capable of extracting all necessary information from PDF files (both scanned hard copies and digitally generated PDFs) (see Figure 1). By using different techniques including adaption of open source libraries for vectorisation of image data, and modern computer vision technologies, such as optical character recognition (OCR), our pipeline is able to flexibly process data from different recording devices and read both data in PDF format and data from native digital devices delivered in XML. The processed files from various sources are either saved as a common and easily accessible CSV file format, or are processed directly with deep learning models (see Figure 2). Results The developed data-pipeline was validated using data from a set of 113 12-lead ECGs for which data was available in multiple formats. Each format dataset was separately processed by our pipeline and then used for training and validation of a deep learning architecture for myocardial scar detection based on raw ECG signals. The quality of the extraction process by our pipeline was assessed by the respective deep learning models with their prediction capability depicted by receiver operator characteristic analyses (ROC). Comparing the benchmark model that was generated from XML data against a model that was purely trained on PDF data processed by the pipeline shows that both models produced comparable results, reaching area under the curve (AUC) values of 0:79±0:10 (XML) and 0:83±0:07 (PDF). Conclusion The data pipeline facilitates acceleration of ECG-based AI research and application of AI algorithms by providing access to ECG data irrespective of the format of the stored ECG. Future work will focus on independent validation as well as expanding this pipeline to include additional ECG types. Funding Acknowledgement Type of funding sources: Public Institution(s). Main funding source(s): Flexi Funds by Forschungscampus Mittelhessen
This is a commonly asked question, but unfortunately, there is not a simple answer. Currently, there is no agreed-on method to measure the depth of a patient's nutritional risk.
Background Cardiac troponins have become the gold standard for appropriate and rapid diagnosis of an acute myocardial infarction (AMI). However, elevated cardiac troponins levels can be observed in the absence of AMI in elderly patients and therefore a loss of specificity is seen in the diagnosis of AMI in elderly patients. Purpose This study aims to define old age in suspected AMI and to evaluate the effect of older age on the diagnostic specificity of a high sensitivity troponin I (hs-cTnI) and high sensitivity troponin T (hs-cTnT) assay in elderly women and men. Methods This study used data from patients with suspected AMI who were enrolled between 08–2011 and 10–2016 to a multicentre biomarker registry and an University Hospital and with available hs-cTnI and hs-cTnT upon admission (n=564 patients). First, we investigated the effect of age on the specificity of both hs-cTnI and hs-cTnT to detect AMI continuously. We then classified over seventy-year-olds as elderly (n=281; 49,8%) and compared the specificity of hs-cTnI and hs-cTnT between elderly and younger patients. Results In the group of older patients, 62.6% (n=176) were diagnosed with AMI compared with 50.9% (n=144) in younger age patients (p<0.01). Elderly patients had higher median levels of hs-cTnI and hs-cTnT than younger patients: 45.4 ng/L vs. 13.1 ng/L hs-cTnI (p<0.001) and 36.8 ng/L vs. 12.8 ng/L hs-cTnT (p<0.001) irrespective of the final diagnosis. Figure1 shows a shift in diagnostic specificity for the whole troponin concentration range in elderly patients compared to younger patients starting at the age of 60 years with clear impact in patients older than 70 years used as threshold for old age in further analyses. Applying the 99th percentile cut-off led to a specificity of 93.5% (95% CI: 88.1–97.0%) in younger patients vs. 88.6% (95% CI: 80.9–94.0%) in elderly patients for hs-cTnI and of 91.4% (95% CI: 85.4–95.5%) vs. 65.7% (95% CI: 55.8–74.7%) for hs-cTnT. To achieve the same specificity in elderly patients as in younger patients, optimized rule-in cut-offs were calculated with 2.6-fold of the 99th percentile for hs-cTnI (68.9 ng/L) and 3.1-fold of the 99th percentile for hs-TnT (43.9 ng/L). Conclusions The specificity of both, hs-cTnI and hs-cTnT is significantly reduced in elderly patients due to higher hs-cTnT and hs-cTnI levels in elderly patients at admission irrespective of the final diagnosis. A relevant loss in diagnostic specificity is observed at 70 years of age starting already at 60 years. Use of assay-specific adjusted cut-offs for hs-cTnI and hs-cTnT for elderly patients regains diagnostic specificity that leads to a more accurate decision-making concerning “rule-in” for older patients with suspected AMI. Funding Acknowledgement Type of funding source: None
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.