Search

Invasive angiography is the reference standard for coronary artery disease (CAD) diagnosis but is expensive and associated with certain risks. Machine learning (ML) using clinical and noninvasive imaging parameters can be used for CAD diagnosis to avoid the side effects and cost of angiography. However, ML methods require labeled samples for efficient training. The labeled data scarcity and high labeling costs can be mitigated by active learning. This is achieved through selective query of challenging samples for labeling. To the best of our knowledge, active learning has not been used for CAD diagnosis yet. An Active Learning with Ensemble of Classifiers (ALEC) method is proposed for CAD diagnosis, consisting of four classifiers. Three of these classifiers determine whether a patient's three main coronary arteries are stenotic or not. The fourth classifier predicts whether the patient has CAD or not. ALEC is first trained using labeled samples. For each unlabeled sample, if the outputs of the classifiers are consistent, the sample along with its predicted label is added to the pool of labeled samples. Inconsistent samples are manually labeled by medical experts before being added to the pool. The training is performed once more using the samples labeled so far. The interleaved phases of labeling and training are repeated until all samples are labeled. Compared with 19 other active learning algorithms, ALEC combined with a support vector machine classifier attained superior performance with 97.01% accuracy. Our method is justified mathematically as well. We also comprehensively analyze the CAD dataset used in this paper. As part of dataset analysis, features pairwise correlation is computed. The top 15 features contributing to CAD and stenosis of the three main coronary arteries are determined. The relationship between stenosis of the main arteries is presented using conditional probabilities. The effect of considering the number of stenotic arteries on sample discrimination is investigated. The discrimination power over dataset samples is visualized, assuming each of the three main coronary arteries as a sample label and considering the two remaining arteries as sample features., Competing Interests: Declaration of competing interest None Declared., (Copyright © 2023 The Authors. Published by Elsevier Ltd.. All rights reserved.)

Background and Objectives: The clinical problems and needs of our oldest old (≥85 years) are often substantially different from those of younger patients, and are arguably different from younger elderly patients (age 65-84). With the increasing number of frail oldest olds residing in Nursing Homes (NH), we aim to identify differences in prognostic indicators and outcomes in this age group compared to younger NH residents., Methods: We retrospectively identified all consecutive admissions from NHs to an Acute Medical Assessment Unit between January 2005 and December 2007. Admission prognostic indicators and outcomes at follow-up were compared between younger (<85) and older (≥85) age groups. Using multiple regression methods controlling for potential confounders, we compared in-hospital mortality and long-term survival after discharge between the groups., Results: Three hundred and sixteen patients (mean age 84.3, SD 8.34 years) were included (68 % females). Admission characteristics were mostly similar between age groups. In-hospital mortality rates were not significantly different between groups, even after adjusting for possible confounders. Oldest old patients had a significantly greater hazard of dying after discharge (HR 1.37; 1.03-1.83) compared to the younger group after removing explanatory variables with more than 5 % missing data., Conclusion: Whilst the admission characteristics are similar between younger and older patients from NHs, there is evidence to suggest worse long-term survival prospects for oldest old patients.

Background: Emergency admissions from nursing homes (NHs) are associated with high mortality. Understanding the predictors of early mortality in these patients may guide clinicians in choosing appropriate site and level of care., Methods: We identified all consecutive admissions from NHs (all ages) to an Acute Medical Assessment Unit between January 2005 and December 2007. Analysis was performed at the level of the admission. The predictors of in-patient mortality at 7 days were examined using a generalized estimating equations analysis., Results: A total of 314 patients [32% male, mean age: 84.2 years (SD: 8.3 years)] were admitted during the study period constituting 410 emergency episodes. Twenty-three percent of admissions resulted in hospital mortality with 73% of deaths occurring within 1 week (50% within the first 3 days). For 7-day mortality outcome, patients with a modified early warning score (MEWS) of 4-5 on admission had 12 times the odds of death [95% confidence interval (CI) 1.40-103.56], whereas those with a score of ≥6 had 21 times the odds of death (95% CI 2.71-170.57) compared with those with a score of ≤1. An estimated glomerular filtration rate (eGFR) of 30-60 and <30 ml/min/m(2) was associated with nearly a 3-fold increase in the odds of death at 1 week (95% CI 1.10-7.97) and a 5-fold increase in the odds of death within 1 week (95% CI 1.75-14.96), respectively, compared with eGFR > 60 ml/min/m(2). C-reactive protein (CRP) >100 mg/l on admission was also associated with a 2.5 times higher odds of death (95% CI 1.23-4.95). Taking eight or more different medication items per day was associated with only a third of the odds of death (95% CI 0.09-0.98) compared with patients taking only three or fewer per day., Conclusion: In acutely ill NH residents, MEWS is an important predictor of early hospital mortality and can be used in both the community and the hospital settings to identify patients whose death maybe predictable or unavoidable, thus allowing a more holistic approach to management with discussion with patient and relatives for planning of immediate care. In addition, CRP and eGFR levels on admission have also been shown to predict early hospital mortality in these patients and can be used in conjunction with MEWS in the same way to allow decision making on the appropriate level of care at the point of hospital admission.

Background and Aims: Hospital admissions from Nursing Homes (NHs) are associated with high mortality. Identifying people with a poor prognosis admitted from NHs is essential to inform appropriate clinical decision making., Methods: We identified all consecutive admissions from NHs (all ages) to an Acute Medical Assessment Unit (AMU) of a large District General Hospital in UK with a catchment population of ~360,000 between January 2005-December 2007 and reviewed their outcome to end of March 2009 (median follow-up=133 days). The relation between admission vital signs (systolic blood pressure, pulse rate, respiratory rate, body temperature and Glasgow Coma Scale-GCS) and outcomes including in-patient mortality, hospital length of stay and mortality at followup were examined using logistic and Cox regression models., Results: The cohort consisted of 316 patients (32% male), mean age at admission was 83.8 years (SD 8.36 yrs; range=49-99 yrs). Sixty-seven (21%) had at least two admissions during the study period; the maximum number of readmissions was five. We found strong evidence that lower systolic blood pressure and higher respiratory rate at the time of admission were associated with increased probability of in-patient death and reduced survival time but not with length of stay. Older age and lower admission GCS were additionally associated with overall poor prognosis., Conclusion: Simple and readily available hospital admission parameters predict not only the in-patient mortality but also longer term outcome for NH residents who require acute hospital admission. Further studies are required to examine whether opportunities exist to intervene and improve outcome in this patient population.