1,720,997 research outputs found
Un approccio guidato dai modelli per la verifica e la validazione precoci di sistemi embedded
No full text
From UML/MARTE Specifications to ESL HW/SW Co-Design: Early Functional Verification and Timing Validation
Get PDFInsulin sensitivity and beta-cell function: Relationships to glycemic control in newly diagnosed Type 2 diabetic patients
No full textnsulin sensitivity and beta-cell function: Relationships to glycemic control in newly diagnosed Type 2 diabeti
CC4CS: An off-the-shelf unifying statement-level performance metric for HW/SW technologies
No full textHow to Sustainably Monitor ML-Enabled Systems? Accuracy and Energy Efficiency Tradeoffs in Concept Drift Detection
Get PDFML-enabled systems that are deployed in a production environment typically suffer from decaying model prediction quality through concept drift, i.e., a gradual change in the statistical characteristics of a certain real-world domain. To combat this, a simple solution is to periodically retrain ML models, which unfortunately can consume a lot of energy. One recommended tactic to improve energy efficiency is therefore to systematically monitor the level of concept drift and only retrain when it becomes unavoidable. Different methods are available to do this, but we know very little about their concrete impact on the tradeoff between accuracy and energy efficiency, as these methods also consume energy themselves. To address this, we therefore conducted a controlled exper-iment to study the accuracy vs. energy efficiency tradeoff of seven common methods for concept drift detection. We used five synthetic datasets, each in a version with abrupt and one with gradual drift, and trained six different ML models as base classifiers. Based on a full factorial design, we tested 420 combinations (7 drift detectors × 5 datasets × 2 types of drift × 6 base classifiers) and compared energy consumption and drift detection accuracy. Our results indicate that there are three types of detectors: a) detectors that sacrifice energy efficiency for detection accuracy (KSWIN), b) balanced detectors that consume low to medium en-ergy with good accuracy (HDDM_ W, ADWIN), and c) detectors that consume very little energy but are unusable in practice due to very poor accuracy (HDDM_A, PageHinkley, DDM, EDDM). By providing rich evidence for this energy efficiency tactic, our findings support ML practitioners in choosing the best suited method of concept drift detection for their ML-enabled systems.</p
Elevated serum uric acid concentrations independently predict cardiovascular mortality in type 2 diabetic patients.
No full textOBJECTIVE— There is limited information on whether increased serum uric acid levels are independently associated with cardiovascular mortality in type 2 diabetes. We assessed thepredictive role of serum uric acid levels on all-cause and cardiovascular mortality in a large cohort of type 2 diabetic individuals.RESEARCH DESIGN AND METHODS— The cohort included 2,726 type 2 diabetic outpatients, who were followed for a mean period of 4.7 years. The independent association of serum uric acid levels with all-cause and cardiovascular mortality was assessed by Cox proportional hazards models and adjusted for conventional risk factors and several potential confounders.RESULTS— During follow-up, 329 (12.1%) patients died, 44.1% (n = 145) of whom from cardiovascular causes. In univariate analysis, higher serum uric acid levels were significantly associated with increased risk of all-cause (hazard ratio 19 [95% CI 1.12–1.27], P < 0.001) and cardiovascular (1.25 [1.16 –1.34], P < 0.001) mortality. After adjustment for age, sex, BMI, smoking, hypertension, dyslipidemia, diabetes duration, A1C, medication use (allopurinol or hypoglycemic, antihypertensive, lipid-lowering, and antiplatelet drugs), estimated glomerular filtration rate, and albuminuria, the association of serum uric acid with cardiovascular mortalityremained statistically significant (1.27 [1.01–1.61], P = 0.046), whereas the association of serum uric acid with all-cause mortality did not.CONCLUSIONS— Higher serum uric acid levels are associated with increased risk of cardiovascular mortality in type 2 diabetic patients, independent of several potential confounders, including renal function measures
The aspartate aminotransferase-to-alanine aminotransferase ratio predicts all-cause and cardiovascular mortality in patients with type 2 diabetes
No full textAn increased aspartate aminotransferase-to-alanine aminotransferase ratio (AAR) has been widely used as a marker of advanced hepatic fibrosis. Increased AAR was also shown to be significantly associated with the risk of developing cardiovascular (CV) disease. The aim of this study was to assess the relationship between the AAR and mortality risk in a well-characterized cohort of patients with type 2 diabetes.A cohort of 2529 type 2 diabetic outpatients was followed-up for 6 years to collect cause-specific mortality. Cox regression analyses were modeled to estimate the independent association between AAR and the risk of all-cause and CV mortality.Over the 6-year follow-up period, 12.1% of patients died, 47.5% of whom from CV causes. An increased AAR, but not its individual components, was significantly associated with an increased risk of all-cause (adjusted-hazard risk 1.83, confidence interval [CI] 95% 1.14-2.93, P = 0.012) and CV (adjusted-hazard risk 2.60, CI 95% 1.38-4.90, P < 0.003) mortality after adjustment for multiple clinical risk factors and potential confounding variables.The AAR was independently associated with an increased risk of both all-cause and CV mortality in patients with type 2 diabetes. These findings suggest that an increased AAR may reflect more systemic derangements that are not simply limited to liver damage. Further studies are needed to elucidate the pathophysiological implications of an increased AAR
Aortic and mitral annular calcification are predictive of all-cause and cardiovascular mortality in patients with type 2 diabetes.
No full textOBJECTIVE: To examine the association of aortic valve sclerosis (AVS) and mitral annulus calcification (MAC) with all-cause and cardiovascular mortality in type 2 diabetic individuals.RESEARCH DESIGN AND METHODS: We retrospectively analyzed the data from 902 type 2 diabetic outpatients, who had undergone a transthoracic echocardiography for clinical reasons during the years 1992-2007. AVS and MAC were diagnosed by echocardiography, and a heart valve calcium (HVC) score was calculated by summing up the AVS and MAC variables. The study outcomes were all-cause and cardiovascular mortality.RESULTS: At baseline, 477 (52.9%) patients had no heart valves affected (HVC-0), 304 (33.7%) had one valve affected (HVC-1), and 121 (13.4%) had both valves affected (HVC-2). During a mean follow-up of 9 years, 137 (15.2%) patients died, 78 of them from cardiovascular causes. Compared with patients with HVC-0, those with HVC-2 had the highest risk of all-cause and cardiovascular mortality, whereas those with HVC-1 had an intermediate risk (P < 0.0001 by the log-rank test). After adjustment for sex, age, BMI, systolic blood pressure, diabetes duration, A1C, LDL cholesterol, estimated glomerular filtration rate, smoking, history of myocardial infarction, and use of antihypertensive and lipid-lowering drugs, the hazard ratio of all-cause mortality was 2.3 (95% CI 1.1-4.9; P < 0.01) for patients with HVC-1 and 9.3 (3.9-17.4; P < 0.001) for those with HVC-2. Similar results were found for cardiovascular mortality.CONCLUSIONS: Our findings indicate that AVS and MAC, singly or in combination, are independently associated with all-cause and cardiovascular mortality in type 2 diabetic patients
Triglyceride-high-density lipoprotein cholesterol is associated with microvascular complications in type 2 diabetes mellitus.
No full textThe purpose of this study was to evaluate whether a high triglyceride/high-density lipoprotein cholesterol (TG/HDL-C) ratio is associated with an increased incidence of retinopathy and chronic kidney disease (CKD) in type 2 diabetes mellitus. Individuals with type 2 diabetes mellitus (n = 979) with an estimated glomerular filtration rate greater than 60 mL/min and without retinopathy and cardiovascular disease at baseline were followed up for the incidence of diabetic retinopathy (diagnosed by retinography) and CKD (diagnosed by estimated glomerular filtration rate ≤60 mL/min/1.73 m(2)). On follow-up (mean, 4.9 years), 217 (22.2\% of total) subjects experienced CKD and/or diabetic-specific retinal lesions (microvascular complication). Of these, 111 subjects developed isolated retinopathy, 85 developed CKD alone, and 21 developed both complications. The TG/HDL-C ratio was positively associated with an increased risk of incident retinopathy and/or CKD (composite microvascular end point) independently of age, sex, body mass index, diabetes duration, hemoglobin A(1c), hypertension, smoking history, low-density lipoprotein cholesterol, albuminuria, and current use of hypoglycemic, antihypertensive, lipid-lowering, or antiplatelet drugs (multivariable-adjusted odds ratio, 2.15; 95\% confidence intervals, 1.10-4.25; P = .04). These findings suggested that the TG/HDL-C ratio was associated with an increased incidence of microvascular complications in individuals with type 2 diabetes mellitus without prior cardiovascular disease, independently of several potential confounders
- …
