A rapid urease test for presumptive identification of Cryptococcus neoformans

Extra-contractual referrals (ECRs) can be a cause of considerable anxiety to purchasing authorities, mainly because of their potential to generate unexpected expenditure. But ECRs can also be used as a tool for monitoring the demand for, and quality of, clinical services. ECRs were studied in the Darlington Health Authority district using a variety of methods including inter-disciplinary meetings, a series of interviews with local GPs, and a questionnaire to general practitioners on 230 consecutive ECRs. The methods and results of the questionnaire study are presented. The commonest reasons for making ECRs included the mistaken belief that a contract existed with the ECR provider, patient dissatisfaction with the local provider, and referral to benefit from shorter waiting lists. ECRs for bone-mass densitometry, orthopaedics, and ear nose and throat services were over-represented. Questionnaire results were validated by comparison with an interview study of all GPs in the district. We conclude that trends in ECRs can be monitored as a convenient "early warning system' to alert purchasing authorities to changes in demand or perceived problems with local provider units. ECR data must be interpreted in the context of further local background information from sources such as GPs and public health physicians. In the case of Darlington, scrutiny of ECRs has led to changes in services and contracts.     

Pronunciation prediction with Default&Refine

The Default&Refine algorithm is a new rule-based learning algorithm that was developed as an accurate and efficient pronunciation prediction mechanism for speech processing systems. The algorithm exhibits a number of attractive properties including rapid generalisation from small training sets, good asymptotic accuracy, robustness to noise in the training data, and the production of compact rule sets. We describe the Default&Refine algorithm in detail and demonstrate its performance on two benchmarked pronunciation databases (the English OALD and Flemish FONILEX pronunciation dictionaries) as well as a newly-developed Afrikaans pronunciation dictionary. We find that the algorithm learns more efficiently (achieves higher accuracy on smaller data sets) than any of the alternative pronunciation prediction algorithms considered. In addition, we demonstrate the ability of the algorithm to generate an arbitrarily small rule set in such a way that the trade-off between rule set size and accuracy is well controlled. A conceptual comparison with alternative algorithms (including Dynamically Expanding Context, Transformation-Based Learning and Pronunciation by Analogy) clarifies the competitive performance obtained with Default&Refine.     

Two sepedi-english code-switched speech corpora

We report on the development of two reference corpora for the analysis of Sepedi-English code-switched speech in the context of automatic speech recognition. For the first corpus, possible English events were obtained from an existing corpus of transcribed Sepedi-English speech. The second corpus is based on the analysis of radio broadcasts: actual instances of code switching were transcribed and reproduced by a number of native Sepedi speakers. We describe the process to develop and verify both corpora and perform an initial analysis of the newly produced data sets. We find that, in naturally occurring speech, the frequency of code switching is unexpectedly high for this language pair, and that the continuum of code switching (from unmodified embedded words to loanwords absorbed into the matrix language) makes this a particularly challenging task for speech recognition systems.

     

Collecting and evaluating speech recognition corpora for 11 South African languages

We describe the Lwazi corpus for automatic speech recognition (ASR), a new telephone speech corpus which contains data from the eleven official languages of South Africa. Because of practical constraints, the amount of speech per language is relatively small compared to major corpora in world languages, and we report on our investigation of the stability of the ASR models derived from the corpus. We also report on phoneme distance measures across languages, and describe initial phone recognisers that were developed using this data. We find that a surprisingly small number of speakers (fewer than 50) and around 10 to 20 h of speech per language are sufficient for the purposes of acceptable phone-based recognition.     

Biological Kinetic Data Evaluation of an Activated Sludge System Coupled with an Ultrafiltration Membrane

A membrane bioreactor (MBR) system treating wastewater containing high molecular weight compounds was operated at solids retention times (SRTs) ranging from 30 to 2 days. Chemical oxygen demand removal efficiencies exceeded 99% and effective nitrification was obtained at SRTs between 30 and 5 days. A significant shift in the biological population structure was observed at the 2 days SRT as the content of gram-negative microorganisms increased and nitrifying bacteria were washed out. At this low SRT, limitations in the biological reaction kinetics resulted in incomplete degradation of the feed protein increasing the presence of soluble organic matter in the effluent. Furthermore, the diluted mixed liquor prevented the formation of a filtration cake on the membrane surface, further deteriorating effluent quality. Biological kinetic data parameters were analyzed using three different representations for biomass: volatile suspended solids, lipid phosphates, and total enzymatic activity. All three indicators exhibited similar trends resulting in very comparable estimates for endogenous decay coefficients, thus demonstrating the reliability of volatile suspended solids as a measure for biological activity in activated sludge. Lower than typical endogenous decay rates in the MBR suggested favorable environmental conditions for respiration and a lower potential for self oxidation and predation. The true yield coefficient was in the range of conventional activated sludge systems, refuting previous suggestions of lower yields in MBRs.     

Hydrogen peroxide is involved in lymphocyte activation mechanisms to induce angiogenesis

Ligands of the benzodiazepine binding site allosterically modulate gamma-aminobutyric acidA receptors. Their binding pocket is made up of amino acid residues located on both alpha and gamma subunits. We transiently expressed wild-type alpha1beta2gamma2 and mutant GABAA receptors in human embryonic kidney 293 cells and determined their binding properties. Receptors containing the mutant alphaY209A showed approximately 40-fold decrease in affinity for [3H]Ro 15-1788 and diazepam, whereas zolpidem displayed no measurable affinity. Receptors containing the mutant alphaY209F showed a small-to-moderate decrease in affinity for [3H]Ro 15-1788, diazepam, zolpidem, methyl-6,7-dimethoxy-4-ethyl-beta-carboline-3-carboxylate, and Cl 218872, amounting to 2-8-fold. Receptors containing the mutant alphaY209Q appeared in the surface membrane of transfected cells, bound [3H]muscimol with wild-type affinity, but failed to bind [3H]Ro 15-1788 or [3H]flunitrazepam with detectable affinity. If these mutant receptors were expressed in Xenopus laevis oocytes, the apparent affinity for GABA was only slightly decreased, whereas the ability of the currents to be stimulated by low concentrations of flunitrazepam was abolished. Receptors containing a point mutant of another amino acid residue, alphaT206A, surprisingly showed an increase in affinity of 5- and 16-fold, for the negative allosteric modulator methyl-6,7-dimethoxy-4-ethyl-beta-carboline-3-carboxylate and the partial positive allosteric modulator Cl 218872, respectively, whereas there was only a small decrease in affinity for Ro 15-1788, diazepam, and zolpidem, amounting to 2-, 4-, and 5-fold. Both alpha206 and alpha209 are thus both important in determining the binding affinities for ligands of the benzodiazepine binding site. The residues are spaced at an interval of three amino acids and may be part of an alpha helix.