Numeracy plays a key role in natural language understanding. However, existing NLP approaches, either traditional word2vec approach or contextualized transformer-based language models, fail to learn numeracy. As the result, the performance of these models is limited when they are applied to number-intensive applications in clinical and financial domains. In this work, we propose a simple number embedding approach based on knowledge graph. We construct a knowledge graph consisting of number entities and magnitude relations. Knowledge graph embedding method is then applied to obtain number vectors. Our approach is easy to implement, and experiment results on various numeracy-related NLP tasks demonstrate the effectiveness and efficiency of our method.
Purpose Guidelines have not recommended routine transthoracic echocardiography (TTE) for elderly patients prior to noncardiac surgery. We aimed to evaluate the significance of preoperative TTE to predict perioperative cardiac complications (PCCs) for elderly patients with coronary artery disease (CAD) undergoing noncardiac surgery. Patients and methods We retrospectively reviewed 2204 patients over 65 years of age with CAD who underwent TTE before intermediate- or high-risk noncardiac surgery in a teaching hospital in China between September 2013 and August 2019. The revised cardiac risk index (RCRI) was assessed. PCCs comprised acute coronary syndrome, heart failure, new-onset severe arrhythmia, nonfatal cardiac arrest, and cardiac death. Logistic regression was used to build the prediction model for PCCs. Discrimination was evaluated using receiver operating characteristic curves, and a nomogram of the predictive model was constructed. Results PCCs occurred in 189 (8.6%) patients. Multivariable analysis showed that eight clinical risk factors (age, history of myocardial infarction, insulin therapy for diabetes, New York Heart Association classification, preoperative serum creatinine, preoperative electrocardiogram ST-T abnormality and pathological Q wave, and American Society of Anesthesiologists classification) and five TTE parameters (left atrial anteroposterior dimension, left ventricular ejection fraction, left ventricular diastolic dysfunction, pulmonary hypertension, and regional ventricular wall motion abnormality) were associated with PCCs. The receiver operating characteristic curve for the clinical plus TTE model provided better discrimination for PCCs compared with the RCRI model (area under the curve: 0.731 vs 0.564; P < 0.001) and the clinical model (area under the curve: 0.731 vs 0.697, P = 0.001), respectively. The clinical plus TTE model was presented as a prognostic nomogram. Conclusion Preoperative TTE may help predict PCCs in elderly patients with CAD undergoing noncardiac surgery, and the prognostic nomogram from this study appeared to be useful for the assessment of perioperative cardiac risk.
BackgroundThe risk of perioperative cardiac complications (PCCs) in patients living in high-altitude areas may increase with more adverse clinical outcomes due to the special geographical environment, which has not yet been studied. We aimed to determine the incidence and analyze risk factors for PCCs in adult patients undergoing major noncardiac surgery in the Tibet Autonomous Region.MethodsThis prospective cohort study enrolled resident patients from high-altitude areas receiving major noncardiac surgery in Tibet Autonomous Region People's Hospital in China. Perioperative clinical data were collected, and the patients were followed up until 30 days after surgery. The primary outcome was PCCs during the operation and within 30 days after the surgery. Logistic regression was used to build the prediction models for PCCs. A receiver operating characteristic (ROC) curve was used to evaluate the discrimination. A prognostic nomogram was constructed to generate a numerical probability of PCCs for patients undergoing noncardiac surgery in high-altitude areas.ResultsAmong the 196 patients living in high-altitude areas involved in this study, 33 (16.8%) suffered PCCs perioperatively and within 30 days after surgery. Eight clinical factors were identified in the prediction model, including older age (P = 0.028), extremely high altitude above 4,000 m (P = 0.442), preoperative metabolic equivalent (MET) < 4 (P = 0.153), history of angina within 6 months (P = 0.037), history of great vascular disease (P = 0.073), increased preoperative high sensitivity C-reactive protein (hs-CRP) (P = 0.072), intraoperative hypoxemia (P = 0.025) and operation time >3 h (P = 0.043). The area under the curve (AUC) was 0.766 (95% confidence interval: 0.785–0.697). The score calculated from the prognostic nomogram predicted the risk of PCCs in high-altitude areas.ConclusionThe incidence of PCCs in resident patients living in high-altitude areas who underwent noncardiac surgery was high, and the risk factors included older age, high altitude above 4,000 m, preoperative MET < 4, history of angina within 6 months, history of great vascular disease, increased preoperative hs-CRP, intraoperative hypoxemia, and operation time >3 h. The prognostic nomogram of this study could help to assess the PCCs for patients in high-attitude areas undergoing noncardiac surgery.Clinical Trial RegistrationClinicalTrials.gov ID: NCT04819698.
Coronary artery disease is one of the common cardiovascular diseases and threatens our health. Percutaneous coronary intervention (PCI) is the main treatment for coronary artery disease. Bare-metal stents (BMS) and drug-eluting stents (DES) are popularly applied in clinical practice. With the development of clinic and technology, DES is widely used in coronary artery disease following antiplatelet therapy. Clinical guidelines require dual antiplatelet therapy (DAPT) after PCI in USA and Europe. It is effective to reduce the incident of stent thrombosis, myocardial infarction and major adverse cardiac. However, the risk of major adverse cardiac events and bleeding was increased by DAPT. Therefore, single antiplatelet therapy was induced in various clinical trials. We searched previous research from PubMed, Embase Web of Science
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.