A5057092506- Face and Expression Recognition
- Sparse and Compressive Sensing Techniques
- Machine Learning and Algorithms
- Bacterial Identification and Susceptibility Testing
- Statistical Methods and Inference
- Bayesian Methods and Mixture Models
- Metaheuristic Optimization Algorithms Research
- Neural Networks and Applications
- Antibiotic Resistance in Bacteria
- Machine Learning and ELM
- Clostridium difficile and Clostridium perfringens research
- Evolutionary Algorithms and Applications
- Domain Adaptation and Few-Shot Learning
- Gut microbiota and health
- Algorithms and Data Compression
- Artificial Intelligence in Healthcare
- Blind Source Separation Techniques
- Machine Learning and Data Classification
- Pharmaceutical and Antibiotic Environmental Impacts
- Stochastic Gradient Optimization Techniques
- Diabetes Management and Research
- Image Enhancement Techniques
- Robotic Path Planning Algorithms
- Actinomycetales infections and treatment
- Mathematical Approximation and Integration
University of Waikato
2014-2024
University of Utah
2015-2019
ARUP Laboratories (United States)
2015-2019
University of California, San Francisco
2019
University of Birmingham
2008-2013
The Edgbaston Hospital
2010
University Pathologists
2008
Abstract Aims To ascertain the frequency and identify predictors of self‐reported hypoglycaemia in Type 1 insulin‐treated 2 diabetes. Methods A random sample 267 people with diabetes were recruited from a population‐based register Tayside, Scotland. Each subject prospectively recorded number mild severe hypoglycaemic episodes experienced over 1‐month period. Ordinal logistic regression was performed to potential hypoglycaemia. Results Five hundred seventy‐two events reported by 155 patients....
Estimations of distribution algorithms (EDAs) are a major branch evolutionary (EA) with some unique advantages in principle. They able to take advantage correlation structure drive the search more efficiently, and they provide insights about space. However, model building high dimensions is extremely challenging, as result existing EDAs may become less attractive large-scale problems because associated large computational requirements. Large-scale continuous global optimisation key many...
Abstract We used metagenomic next-generation sequencing to longitudinally assess the gut microbiota and antimicrobial resistomes of international travelers clarify global exchange resistant organisms. Travel resulted in an increase resistance genes a greater proportion Escherichia species within microbial communities without impacting diversity.
We consider random projections in conjunction with classification, specifically the analysis of Fisher's Linear Discriminant (FLD) classifier randomly projected data spaces.
Matrix-assisted laser desorption ionization-time of flight mass spectrometry for identification Nocardia species remains challenging. By identifying 83.1% (64 77) and 80% (8 10) to the complex levels, respectively, 94.3% (82 87) genus level, we show that an approach using routine sample preparation, up-to-date commercial database minimally augmented with custom spectra, testing at early stage growth is promising.
Neonatal sepsis causes significant mortality and morbidity worldwide. Diagnosis is usually confirmed via blood culture results. Blood confirmation can take days suffer from contamination false negatives. Empiric therapy with antibiotics common. This study aims to retrospectively describe compare treatments of culture-confirmed unconfirmed, but suspected, within the University Utah Hospital system. Electronic health records were obtained 1,248 neonates January 1, 2006, December 31, 2017....
Stool culture for Clostridium difficile, while necessary strain typing and antimicrobial surveillance, cannot determine toxin production. We prospectively tested in triplicate 91 C difficile cultured isolates production by 2 enzyme immunoassays (EIAs) (Meridian Premier Toxins A&B, Meridian Bioscience, Cincinnati, OH; TechLab Tox A/B II, TechLab, Blacksburg, VA) cytotoxin neutralization bioassay (CTN). By CTN, 88% (80/91) were toxigenic. Reproducibility was 93% (85/91) 80% (73/91) EIA, 79%...
Summary— A series of 27 patients with measurable metastases from primary carcinoma the bladder received cis‐platinum 50 mg/m 2 /3‐weekly. Twenty‐six per cent (32% if receiving only one injection are excluded) responded for a median 4 months.
Estimation of distribution algorithms (EDA) are a major branch evolutionary (EA) with some unique advantages in principle. They able to take advantage correlation structure drive the search more efficiently, and they provide insights about space. However, model building high dimensions is extremely challenging as result existing EDAs lose their strengths large scale problems.
Digital video incurs many distortions during processing, compression, storage, and transmission, which can reduce perceived quality. Developing adaptive transmission methods that provide increased bandwidth reduced storage space while preserving visual quality requires metrics accurately describe how people perceive distortion. A severe problem for developing new is the limited data on early human system simultaneously processes spatial temporal information. The exacerbated by fact few...
Nurses must have sufficient education and training in microbiology to perform many roles within clinical nursing practice (e.g., administering antibiotics, collecting specimens, preparing specimens for transport delivery, educating patients families, communicating results the healthcare team, developing care plans based on of studies patient immunological status). It is unclear whether current courses required students United States focus topics that are most relevant practice. To gauge...
We consider the problem of classification in nonadaptive dimensionality reduction. Specifically, we bound increase error Fisher's Linear Discriminant classifier resulting from randomly projecting high dimensional data into a lower space and both learning performing projected space. Our is reasonably tight, unlike existing bounds on data, it becomes tighter as quantity training increases without requiring any sparsity structure data.
We consider the problem of high dimensional blackbox optimisation via Estimation Distribution Algorithms (EDA) and use heavy-tailed search distributions in this setting. Some authors have suggested that employing a heavy tailed distribution, such as Cauchy, may make EDA better explore space. However, other found Cauchy are less effective than Gaussian problems. In paper, we set out to resolve controversy. To achieve run extensive experiments on battery high-dimensional test functions,...
Abstract We engaged metagenomic next generation sequencing to longitudinally assess the gut microbiota and antimicrobial resistomes of international travelers understand global exchange resistant organisms. Travel resulted in an increase resistance genes a greater proportion Escherichia species within microbial communities without impacting diversity.