Comparative quantification of focal and diffuse visual field loss by the SPARK Precision threshold algorithm and SITA.

PURPOSE: The aims of this paper were to examine focal and diffuse visual field loss in terms of threshold agreement between the widely used SITA Standard Humphrey Field Analyser (HFA) threshold algorithm with the SPARK Precision algorithm (Oculus Twinfield 2). METHODS: A total of 39 treated glaucoma patients (34 primary open angle and 5 primary angle closure glaucoma) and 31 cataract patients without glaucoma were tested in succession with the Oculus Twinfield 2 (Oculus Optikgeräte GmbH, Wetzlar, Germany) using the SPARK Precision algorithm and with the HFA 3 (Carl Zeiss Meditec, Dublin, CA) using the 30-2 SITA Standard algorithm. RESULTS: SPARK Precision required around half the testing time of SITA Standard. There was a good correlation between the MS of the two threshold algorithms but MD and PSD were significantly less severe with SPARK Precision in both glaucoma (focal field loss) and cataract (diffuse field loss) groups (p < 0.001). There was poor agreement for all global indices (MS, MD and PSD) between the two algorithms and there was a significant proportional bias of MD in the glaucoma group and PSD in both glaucoma and cataract groups. The pointwise sensitivity analysis yielded higher threshold estimates in SPARK Precision than in SITA Standard in the nasal field. Classification of glaucoma severity using AGIS was significantly lower with SPARK Precision compared to SITA Standard (p < 0.001). CONCLUSION: SITA renders deeper defects than SPARK. Compared to the SITA Standard threshold algorithm, SPARK Precision cannot quantify early glaucomatous field loss. This may be due to the mathematical linear interpolation of threshold sensitivity or deeper scotomas due to the plateau effect caused by the reduced dynamic range of the Twinfield 2 perimeter. Although not of clinical significance in early glaucoma, the plateau effect may hinder the long-term follow-up of patients during disease progression.

Removal of dissolved copper(II) and zinc(II) by aerobic granular sludge.

This study investigated the adsorption kinetics of dissolved copper(II) and zinc(II) by aerobic granular sludge. Two series of batch experiments were conducted at different initial copper(II), zinc(II) concentrations (Co) and initial granule concentrations (Xo). Results showed that the biosorption kinetics of individual copper(II) and zinc(II) by aerobic granules were closely related to Co and Xo. The maximum biosorption capacity of individual copper(II) and zinc(II) by aerobic granules was 246.1 mg g(-1) and 180 mg g(-1), respectively. In order to theoretically interpret the results obtained, two kinetic models previously developed for biosorption were employed and compared in this study. It was found that the model proposed by Liu et al. (2003) could fit the experimental data very well, but the second-order model failed to fit the data in some cases. It appears that aerobic granules would be potential biosorbent with high efficiency for the removal of dissolved copper(II) and zinc(II) from wastewater.

Parallel implementation of backpropagation neural networks on a heterogeneous array of transputers.

This paper analyzes parallel implementation of the backpropagation training algorithm on a heterogeneous transputer network (i.e., transputers of different speed and memory) connected in a pipelined ring topology. Training-set parallelism is employed as the parallelizing paradigm for the backpropagation algorithm. It is shown through analysis that finding the optimal allocation of the training patterns amongst the processors to minimize the time for a training epoch is a mixed integer programming problem. Using mixed integer programming optimal pattern allocations for heterogeneous processor networks having a mixture of T805-20 (20 MHz) and T805-25 (25 MHz) transputers are theoretically found for two benchmark problems. The time for an epoch corresponding to the optimal pattern allocations is then obtained experimentally for the benchmark problems from the T805-20, TS805-25 heterogeneous networks. A Monte Carlo simulation study is carried out to statistically verify the optimality of the epoch time obtained from the mixed integer programming based allocations. In this study pattern allocations are randomly generated and the corresponding time for an epoch is experimentally obtained from the heterogeneous network. The mean and standard deviation for the epoch times from the random allocations are then compared with the optimal epoch time. The results show the optimal epoch time to be always lower than the mean epoch times by more than three standard deviations (3sigma) for all the sample sizes used in the study thus giving validity to the theoretical analysis.

A pilot study on intensive lipid lowering--a new approach to cardiac rehabilitation.

The deleterious effects of raised plasma lipids on the progression of coronary artery disease has been well established. Recent research data suggests that intensive lipid lowering represents a new approach to Cardiac Rehabilitation. There was a total of twenty-four participants with twenty-one (87.5%) males and three (12.5%) females. There were seventeen Chinese (70.8%), six Indians (25%) and one Malay (4.2%). Their ages ranged from group forty to forty-nine, five participants (20%); fifty to fifty-nine, twelve participants (50%); sixty to sixty-nine, six participants (25%) and seventy to seventy-nine, one participant (5%). In this study, twelve individuals were treated with Cholestyramine, seven by Gemfibrozil, four by Clofibrate and three by Bezafibrate. The duration of antilipid treatment ranged from two to eighteen months (mean eight months). Side effects were reported in twelve patients (50%) on Cholestyramine who complained of constipation while one (4%) on Bezafibrate had abnormal liver function tests (raised SGOT and SGPT). All patients had been on an exercise and diet control program for one year which they were enrolled in an intensive lipid-lowering protocol. The pre and post treatment lipid profiles with lipid-lowering agents showed 18% reduction for Triglycerides, 15% reduction for Total Cholesterol, 5% increase for HDL-Cholesterol, 19% decrease in LDL Cholesterol, 19% decrease in Total Cholesterol: HDL ratios. The initial results suggest that an intensive lipid-lowering strategy in a Cardiac Rehabilitation Program appears to have a beneficial effect on lipid profiles and represents an expanded approach to lipid management involving exercise, nutrition and pharmacological intervention.

Regulation of folylpoly-gamma-glutamate synthesis in mammalian cells. In vivo and in vitro synthesis of pteroylpoly-gamma-glutamates by Chinese hamster ovary cells.

Cultured Chinese hamster ovary cells accumulated labeled folate and 5-methyltetrahydrofolate and metabolized these compounds to polyglutamates of chain length up to 10, with hexa-, hepta-, and octaglutamates predominating. Intracellular vitamin was concentrated up to 50-fold, although the intracellular pteroylmonoglutamate concentration did not exceed that in the medium. Folylpolyglutamate levels in mutant cells (AUXB1), which are defective in the enzyme folylpolyglutamate synthetase, were less than 1.5% of wild type levels, while pteroylmonoglutamate levels were normal. Tetrahydropteroylpolyglutamates were the most effective polyglutamate substrates for the Chinese hamster ovary cell folylpolyglutamate synthetase. The final distribution of folylpolyglutamates in this cells appeared to be a reflection of the ability of different glutamate chain length folates to act as substrates for the synthetase. The proportions of individual folylpolyglutamates were unaffected by culturing the cells in medium containing or lacking products of one carbon metabolism such as glycine, adenosine, or thymidine. However, these proportions were markedly influenced by the level of methionine in the culture medium. The possible significance of changes in the glutamate chain length of folates under different nutritional conditions in relationship to the regulation of one carbon metabolism is discussed.

Folate derivatives in human cells: studies on normal and 5,10-methylenetetrahydrofolate reductase-deficient fibroblasts.

The distribution of folylpolyglutamates in normal and methylenetetrahydrofolate reductase-deficient human fibroblasts cultured in medium containing folic acid or 5-methyltetrahydrofolic acid has been determined. Human fibroblasts concentrated these folates to higher levels than in the medium, an effect that was more pronounced with methyltetrahydrofolate as the folate source. Over 95% of the intracellular vitamin derivatives were polyglutamates of chain length 2 to 10. The major derivatives were hexaglutamates in cells cultured with folic acid and heptaglutamates in cells cultured with methyltetrahydrofolic acid. No significant differences were detected in the polyglutamate distribution between normal and methylenetetrahydrofolate reductase-deficient fibroblast. Excess medium methionine reduced cell growth rates and intracellular vitamin levels and changed the predominant polyglutamate in cells cultured with methyltetrahydrofolate from hepta- to hexaglutamate. No significant differences were seen between the overall folate polyglutamate distributions of different one-carbon folate pools of normal fibroblasts, although slight changes in the proportions of individual polyglutamate forms were detected in the different pools.

Folylpoly-gamma-glutamate synthesis by bacteria and mammalian cells.

The purification and properties of folylpolyglutamate synthetase from Corynebacterium sp, and some properties of partially purified enzyme from Lactobacillus casei, Streptococcus faecalis, Neurospora crassa, pig liver, and Chinese hamster ovary cells, are described. The Corynebacterium enzyme catalyzes a MgATP-dependent addition of glutamate to a variety of reduced pteroate and pteroylmono-, di-, and triglutamate substrates, with the concomitant production of MgADP and phosphate. Although glutamate moieties are added in a sequential fashion, the kinetic mechanism, which is Ordered Ter Ter, precludes the sequential addition of glutamate moieties to enzyme-bound folate. It is suggested that catalysis precedes via the formation of a pteroyl-gamma-glutamyl phosphate intermediate. The in vivo distribution of folylpolyglutamates in bacteria and mammalian cells, which differ from source to source, appear to be a reflection of the ability of folylpolyglutamates to act as substrates for folylpolyglutamate synthetases from different sources. Only one enzyme appears to be involved in the conversion of pteroylmonoglutamates to polyglutamate forms in both bacteria and mammalian cells. Bacterial folylpolyglutamate synthetases use a variety of pteroylmonoglutamates as their preferred monoglutamate substrate, but use 5,10-methylenetetrahydro-pteroylpolyglutamates as their preferred, and sometimes only, polyglutamate substrate. Mono- and polyglutamyl forms of tetrahydrofolate are the preferred substrates of mammalian folylpolyglutamate synthetases.