9,717 research outputs found
Blood levels of homocysteine, folate, vitamin B-6 and B-12 in women using oral contraceptives compared to non-users
No full textBackground and objectives: To compare the levels of total homocysteine (tHcy), folate, vitamin B-6 and B-12, in women not using oral contraceptives (OC) vs. those using OC. Materials and methods: 219 healthy women were enrolled in the study; 159 of them had not been using OC for at least 12 months prior to their enrolment, while 60 were on regular OC treatment. Results: The median levels of vitamin B-6 and B-12 were significantly lower in OC users than in non-users (24.2 vs. 32.9 nmol/l, p = 0.029; 278 vs. 429 ng/ml, p < 0.001). There were no statistically significant differences in the levels of tHcy (fasting and post-methionine loading) and folate. Conclusions: In our cross-sectional study, OC use was associated with low vitamin B-6 and B-12 levels. Since low vitamin B-6 levels are independently associated with heightened risks for arterial and venous thromboembolism (TE), they could partly account for the increased TE risk of OC users
Determination of total homocysteine in plasma: comparison of the Abbott Imx immunoassay with high performance liquid chromatography
No full textBackground and Objectives. The aim of this study was to compare the performance of a commercially available IMx immunoassay with that of a reversed-phase high performance liquid chromatography (HPLC) method for measuring plasma total homocysteine (tHcy). Methods. The levels of tHcy before and after oral methionine loading (ML) were measured in 135 healthy subjects and 39 patients scheduled for routine tHcy determination. The IMx method uses fluorescence polarization immunoassay (FPIA) technology. The HPLC-method includes derivatization with ABD-F and post-column fluorescence detection. Results. The imprecision was very low with both methods for both normal (11 μmol/L) and high (29 μmol/L) tHcy levels. The within and between-run coefficients of variation were < 5%. Both methods were able to discriminate between similar concentrations of tHcy both at normal and moderately high levels. There was a good correlation between measurements obtained with the two methods (r = 0.985, p = 0.001). The mean levels of tHcy measured with the IMx assay tended to be slightly higher than those with the HPLC both in the fasting state (mean difference = 0.8 μmol/L) and after ML (5.3 μmol/L). However only the difference in post-ML levels was statistically significant (p < 0.001). The percentage of patients with hyperhomocysteinemia identified with the two methods was similar. Interpretation and Conclusions. The IMx method compares well with an established HPLC method for measurement of fasting tHcy plasma levels
Metadata Representations for Queryable ML Model Zoos
No full textMachine learning (ML) practitioners and organizations are building model zoos of pre-trained models, containing metadata describing properties of the ML models and datasets that are useful for reporting, auditing, reproducibility, and interpretability purposes. The metatada is currently not standardised; its expressivity is limited; and there is no interoperable way to store and query it. Consequently, model search, reuse, comparison, and composition are hindered. In this paper, we advocate for standardized ML model metadata representation and management, proposing a toolkit supported to help practitioners manage and query that metadata.Web Information SystemsHuman-Centred Artificial Intelligenc
A Manifesto of Nodalism
No full textThis paper proposes the notion of Nodalism as a means describing contemporary culture and of understanding my own creative practice in electronic music composition. It draws on theories and ideas from Kirby, Bauman, Bourriaud, Deleuze, Guatarri, and Gochenour, to demonstrate how networks of ideas or connectionist neural models of cognitive behaviour can be used to contextualize, understand and become a creative tool for the creation of contemporary electronic music
Molecular bases of defective signal transduction in the platelet P2Y12 receptor of a patient with congenital bleeding
No full textWe have identified structural attributes required for signal transduction through a seven-transmembrane-domain receptor. Platelets from a patient (AC) with a congenital bleeding disorder had normal shape change but reduced and reversible aggregation in response to 4 μM ADP, similar to normal platelets with blocked P2Y12 receptor. The response to 20 μM ADP, albeit still decreased, was more pronounced and was reduced by a P2Y12 antagonist, indicating some residual receptor function. ADP failed to lower the adenylyl cyclase activity stimulated by prostaglandin E1 in the patient's platelets, even though the number and affinity of 2-methylthioadenosine 5′-[33P]diphosphate-binding sites was normal. Analysis of the patient's P2Y12 gene revealed a G-to-A transition in one allele, changing the codon for Arg-256 in the sixth transmembrane domain to Gln, and a C-to-T transition in the other allele, changing the codon for Arg-265 in the third extracellular loop to Trp. Neither mutation interfered with receptor surface expression but both altered function, since ADP inhibited the forskolin-induced increase of cAMP markedly less in cells transfected with either mutant P2Y12 as compared with wild-type receptor. These studies delineate a region of P2Y12 required for normal function after ADP binding
Optimizing ML Inference Queries Under Constraints
No full textThe proliferation of pre-trained ML models in public Web-based model zoos facilitates the engineering of ML pipelines to address complex inference queries over datasets and streams of unstructured content. Constructing optimal plan for a query is hard, especially when constraints (e.g. accuracy or execution time) must be taken into consideration, and the complexity of the inference query increases. To address this issue, we propose a method for optimizing ML inference queries that selects the most suitable ML models to use, as well as the order in which those models are executed. We formally define the constraint-based ML inference query optimization problem, formulate it as a Mixed Integer Programming (MIP) problem, and develop an optimizer that maximizes accuracy given constraints. This optimizer is capable of navigating a large search space to identify optimal query plans on various model zoos.Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.Web Information SystemsHuman-Centred Artificial Intelligenc
Tamoxifen reduces plasma homocysteine levels in healthy women
No full textTreatment with tamoxifen is associated with reduced incidence of myocardial infarction. As plasma homocysteine is an independent risk factor for cardiovascular disease, we studied the effects of tamoxifen on plasma homocysteine in 66 healthy women participating in the Italian prevention trial of breast cancer who were randomized in a double-blind manner to tamoxifen 20 mg day-1 or placebo for 5 years. They were aged between 35 and 70 years, had undergone previous hysterectomy for non-malignant conditions and had no contraindications to the use of tamoxifen. Plasma levels of total homocysteine (tHcy) were measured at randomization and after 2 and 6 months. The mean ± s.d. plasma levels of tHcy were 7.59 ± 1.71 μmol l-1, 7.25 ± 1.61 and 7.09 ± 1.33 in the tamoxifen group and 8.07 ± 2.06, 7.93 ± 1.77 and 8.12 ± 2.04 in the placebo group at 0, 2 and 6 months (P = 0.008 for the between-group difference over time). The higher the baseline tHcy level, the greater was the lowering effect of tamoxifen. No statistically significant effect of age, body mass index or smoking habit on baseline tHcy levels and its variation over time was found. In conclusion, tamoxifen (20 mg day-1 for 6 months) decreased plasma tHcy levels in healthy women. This effect may contribute to its protective effect on myocardial infarction
Low vitamin B(6) plasma levels, a risk factor for thrombosis, in inflammatory bowel disease: role of inflammation and correlation with acute phase reactants
No full textOBJECTIVES: Individuals with inflammatory bowel disease (IBD) are at increased risk for thrombosis and vitamin deficiencies. Low plasma levels of vitamin B6 are an independent risk factor for thrombosis and may cause hyperhomocysteinemia, another recognized risk factor for thrombosis. The aim of this study was to evaluate vitamin B6 plasma levels in IBD patients. METHODS: We studied 61 IBD patients: 32 with Crohn's disease and 29 with ulcerative colitis. For each patient, three sex- and age-matched healthy control subjects were studied. RESULTS: Median vitamin B6 levels were significantly lower in IBD patients (22.0 pmol/L, range 3.6 -231.0) than in controls (31.1 pmol/L, 3.7-363.4; p < 0.01). In all, 13.1% IBD patients and 5.5% controls had plasma vitamin B6 levels lower than the 5th percentile of distribution in normal controls (p < 0.05). Low vitamin B6 levels were significantly more frequent in patients with active disease than in patients with quiescent disease (seven of 26, 26.9%, vs one of 35, 2.9%; p < 0.001). Moreover, patients with active disease had significantly lower median vitamin B6 levels (13.4 pmol/L, range 3.6-124.0) than patients in a quiescent phase (27.0 pmol/L, 7.8-231.0; p < 0.001). Low vitamin B6 levels were significantly correlated with serum concentrations of C-reactive protein (r = -0.36, 95% CI = -0.59 to -0.09, p < 0.01) and α1-acid-glycoprotein (r = -0.37, 95% CI = -0.59 to -0.10, p < 0.01). Hyperhomocysteinemia was more frequent in patients with low vitamin B6 levels (three of eight, 37.5%) than in patients with normal levels (nine of 53, 17.0%; p = 0.18). There was no statistically significant correlation between vitamin B6 and homocysteine plasma levels (r = -0.13, 95% CI = -0.37 to +0.14, p = 0.33). CONCLUSIONS: Low vitamin B6 plasma levels, an independent risk factor for thrombosis, are frequent in patients with IBD, especially those with active disease
Building a generalisable ML pipeline at ING
No full textAdvances in data science have caused an increase in the use of Artificial Intelligence (AI), specifically Machine Learning (ML), throughout various fields. Not only in research but in the industry as well, has ML been receiving increasing amounts of interest. Many companies rely on ML models to increase the efficiency of existing processes or offer new services and products. The industry, however, is facing several additional challenges compared to the academic context. One of those challenges is applying the Development Operations (DevOps) model to an ML application, also referred to as MLOps. This thesis sets out to find the specific challenges that practitioners encounter while operationalising ML models. To do so, we perform a single-case case study on an ML pipeline built by the Trade & Communication Surveillance team at the ING bank. This case study consists of conducting a set of interviews and performing a manual code inspection of the pipeline. The team faces challenges ranging from having insufficient time for operationalising each ML project individually to operating in the highlyregulated fintech context. Their pipeline is able to deploy a single ML model but it does not generalise well to other projects. We present the first version of an application that mitigates these challenges. The application is able to deploy ML models to the development environment at ING and can be operated by data scientists to reduce the effort of operationalising an ML model. Computer Science | Software Technolog
'Project smells' - Experiences in Analysing the Software Quality of ML Projects with mllint
No full textMachine Learning (ML) projects incur novel challenges in their development and productionisation over traditional software applications, though established principles and best practices in ensuring the project's software quality still apply. While using static analysis to catch code smells has been shown to improve software quality attributes, it is only a small piece of the software quality puzzle, especially in the case of ML projects given their additional challenges and lower degree of Software Engineering (SE) experience in the data scientists that develop them. We introduce the novel concept of project smells which consider deficits in project management as a more holistic perspective on software quality in ML projects. An open-source static analysis tool mllint was also implemented to help detect and mitigate these. Our research evaluates this novel concept of project smells in the industrial context of ING, a global bank and large software- and data-intensive organisation. We also investigate the perceived importance of these project smells for proof-of-concept versus production-ready ML projects, as well as the perceived obstructions and benefits to using static analysis tools such as mllint. Our findings indicate a need for context-aware static analysis tools, that fit the needs of the project at its current stage of development, while requiring minimal configuration effort from the user. Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.Software EngineeringSoftware Technolog
- …
