Design-based stereology: Planning, volumetry and sampling are crucial steps for a successful study

Quantitative data obtained by means of design-based stereology can add valuable information to studies performed on a diversity of organs, in particular when correlated to functional/physiological and biochemical data. Design-based stereology is based on a sound statistical background and can be used to generate accurate data which are in line with principles of good laboratory practice. In addition, by adjusting the study design an appropriate precision can be achieved to find relevant differences between groups. For the success of the stereological assessment detailed planning is necessary. In this review we focus on common pitfalls encountered during stereological assessment. An exemplary workflow is included, and based on authentic examples, we illustrate a number of sampling principles which can be implemented to obtain properly sampled tissue blocks for various purposes.

Efficiency of risk-based vs . random sampling for the monitoring of tetracycline residues in slaughtered calves in Switzerland

In all European Union countries, chemical residues are required to be routinely monitored in meat. Good farming and veterinary practice can prevent the contamination of meat with pharmaceutical substances, resulting in a low detection of drug residues through random sampling. An alternative approach is to target-monitor farms suspected of treating their animals with antimicrobials. The objective of this project was to assess, using a stochastic model, the efficiency of these two sampling strategies. The model integrated data on Swiss livestock as well as expert opinion and results from studies conducted in Switzerland. Risk-based sampling showed an increase in detection efficiency of up to 100% depending on the prevalence of contaminated herds. Sensitivity analysis of this model showed the importance of the accuracy of prior assumptions for conducting risk-based sampling. The resources gained by changing from random to risk-based sampling should be transferred to improving the quality of prior information.

A connection between two nonparametric tests for perfect ranking in balanced ranked set sampling

In this note, we show that an extension of a test for perfect ranking in a balanced ranked set sample given by Li and Balakrishnan (2008) to the multi-cycle case turns out to be equivalent to the test statistic proposed by Frey et al. (2007). This provides an alternative interpretation and motivation for their test statistic.

Atmospheric CO₂, δ(O₂/N₂) and δ13CO2 measurements at Jungfraujoch, Switzerland: results from a flask sampling intercomparison program

We present results from an intercomparison program of CO2, δ(O2/N2) and δ13CO2 measurements from atmospheric flask samples. Flask samples are collected on a bi-weekly basis at the High Altitude Research Station Jungfraujoch in Switzerland for three European laboratories: the University of Bern, Switzerland, the University of Groningen, the Netherlands and the Max Planck Institute for Biogeochemistry in Jena, Germany. Almost 4 years of measurements of CO2, δ(O2/N2) and δ13CO2 are compared in this paper to assess the measurement compatibility of the three laboratories. While the average difference for the CO2 measurements between the laboratories in Bern and Jena meets the required compatibility goal as defined by the World Meteorological Organization, the standard deviation of the average differences between all laboratories is not within the required goal. However, the obtained annual trend and seasonalities are the same within their estimated uncertainties. For δ(O2/N2) significant differences are observed between the three laboratories. The comparison for δ13CO2 yields the least compatible results and the required goals are not met between the three laboratories. Our study shows the importance of regular intercomparison exercises to identify potential biases between laboratories and the need to improve the quality of atmospheric measurements.

Design and analysis for three level hierarchical structures with count response

In numerous intervention studies and education field trials, random assignment to treatment occurs in clusters rather than at the level of observation. This departure of random assignment of units may be due to logistics, political feasibility, or ecological validity. Data within the same cluster or grouping are often correlated. Application of traditional regression techniques, which assume independence between observations, to clustered data produce consistent parameter estimates. However such estimators are often inefficient as compared to methods which incorporate the clustered nature of the data into the estimation procedure (Neuhaus 1993).1 Multilevel models, also known as random effects or random components models, can be used to account for the clustering of data by estimating higher level, or group, as well as lower level, or individual variation. Designing a study, in which the unit of observation is nested within higher level groupings, requires the determination of sample sizes at each level. This study investigates the design and analysis of various sampling strategies for a 3-level repeated measures design on the parameter estimates when the outcome variable of interest follows a Poisson distribution. ^ Results study suggest that second order PQL estimation produces the least biased estimates in the 3-level multilevel Poisson model followed by first order PQL and then second and first order MQL. The MQL estimates of both fixed and random parameters are generally satisfactory when the level 2 and level 3 variation is less than 0.10. However, as the higher level error variance increases, the MQL estimates become increasingly biased. If convergence of the estimation algorithm is not obtained by PQL procedure and higher level error variance is large, the estimates may be significantly biased. In this case bias correction techniques such as bootstrapping should be considered as an alternative procedure. For larger sample sizes, those structures with 20 or more units sampled at levels with normally distributed random errors produced more stable estimates with less sampling variance than structures with an increased number of level 1 units. For small sample sizes, sampling fewer units at the level with Poisson variation produces less sampling variation, however this criterion is no longer important when sample sizes are large. ^ 1Neuhaus J (1993). “Estimation efficiency and Tests of Covariate Effects with Clustered Binary Data”. Biometrics , 49, 989–996^

Evaluation of the accuracy of multilevel analysis parameter estimates for intra-cluster correlation in correlated binary data

Many studies in biostatistics deal with binary data. Some of these studies involve correlated observations, which can complicate the analysis of the resulting data. Studies of this kind typically arise when a high degree of commonality exists between test subjects. If there exists a natural hierarchy in the data, multilevel analysis is an appropriate tool for the analysis. Two examples are the measurements on identical twins, or the study of symmetrical organs or appendages such as in the case of ophthalmic studies. Although this type of matching appears ideal for the purposes of comparison, analysis of the resulting data while ignoring the effect of intra-cluster correlation has been shown to produce biased results.^ This paper will explore the use of multilevel modeling of simulated binary data with predetermined levels of correlation. Data will be generated using the Beta-Binomial method with varying degrees of correlation between the lower level observations. The data will be analyzed using the multilevel software package MlwiN (Woodhouse, et al, 1995). Comparisons between the specified intra-cluster correlation of these data and the estimated correlations, using multilevel analysis, will be used to examine the accuracy of this technique in analyzing this type of data. ^

TDOA for Narrow-band Signal with Low Sampling Rate and Imperfect Synchronization

Time-based localization techniques such as multilateration are favoured for positioning to wide-band signals. Applying the same techniques with narrow-band signals such as GSM is not so trivial. The process is challenged by the needs of synchronization accuracy and timestamp resolution both in the nanoseconds range. We propose approaches to deal with both challenges. On the one hand, we introduce a method to eliminate the negative effect of synchronization offset on time measurements. On the other hand, we propose timestamps with nanoseconds accuracy by using timing information from the signal processing chain. For a set of experiments, ranging from sub-urban to indoor environments, we show that our proposed approaches are able to improve the localization accuracy of TDOA approaches by several factors. We are even able to demonstrate errors as small as 10 meters for outdoor settings with narrow-band signals.

Asynchronous ECG time sampling: Saving bits with Golomb-Rice encoding

We present a technique for online compression of ECG signals using the Golomb-Rice encoding algorithm. This is facilitated by a novel time encoding asynchronous analog-to-digital converter targeted for low-power, implantable, long-term bio-medical sensing applications. In contrast to capturing the actual signal (voltage) values the asynchronous time encoder captures and encodes the time information at which predefined changes occur in the signal thereby minimizing the sensor's energy use and the number of bits we store to represent the information by not capturing unnecessary samples. The time encoder transforms the ECG signal data to pure time information that has a geometric distribution such that the Golomb-Rice encoding algorithm can be used to further compress the data. An overall online compression rate of about 6 times is achievable without the usual computations associated with most compression methods.

Sampling strategies for capillary isoelectric focusing with electroosmotic zone mobilization assessed by high-resolution dynamic computer simulation

The impact of initial sample distribution on separation and focusing of analytes in a pH 3–11 gradient formed by 101 biprotic carrier ampholytes under concomitant electroosmotic displacement was studied by dynamic high-resolution computer simulation. Data obtained with application of the analytes mixed with the carrier ampholytes (as is customarily done), as a short zone within the initial carrier ampholyte zone, sandwiched between zones of carrier ampholytes, or introduced before or after the initial carrier ampholyte zone were compared. With sampling as a short zone within or adjacent to the carrier ampholytes, separation and focusing of analytes is shown to proceed as a cationic, anionic, or mixed process and separation of the analytes is predicted to be much faster than the separation of the carrier components. Thus, after the initial separation, analytes continue to separate and eventually reach their focusing locations. This is different to the double-peak approach to equilibrium that takes place when analytes and carrier ampholytes are applied as a homogenous mixture. Simulation data reveal that sample application between two zones of carrier ampholytes results in the formation of a pH gradient disturbance as the concentration of the carrier ampholytes within the fluid element initially occupied by the sample will be lower compared to the other parts of the gradient. As a consequence thereof, the properties of this region are sample matrix dependent, the pH gradient is flatter, and the region is likely to represent a conductance gap (hot spot). Simulation data suggest that sample placed at the anodic side or at the anodic end of the initial carrier ampholyte zone are the favorable configurations for capillary isoelectric focusing with electroosmotic zone mobilization.

The influence of sampling design on tree-ring-based quantification of forest growth

Tree-rings offer one of the few possibilities to empirically quantify and reconstruct forest growth dynamics over years to millennia. Contemporaneously with the growing scientific community employing tree-ring parameters, recent research has suggested that commonly applied sampling designs (i.e. how and which trees are selected for dendrochronological sampling) may introduce considerable biases in quantifications of forest responses to environmental change. To date, a systematic assessment of the consequences of sampling design on dendroecological and-climatological conclusions has not yet been performed. Here, we investigate potential biases by sampling a large population of trees and replicating diverse sampling designs. This is achieved by retroactively subsetting the population and specifically testing for biases emerging for climate reconstruction, growth response to climate variability, long-term growth trends, and quantification of forest productivity. We find that commonly applied sampling designs can impart systematic biases of varying magnitude to any type of tree-ring-based investigations, independent of the total number of samples considered. Quantifications of forest growth and productivity are particularly susceptible to biases, whereas growth responses to short-term climate variability are less affected by the choice of sampling design. The world's most frequently applied sampling design, focusing on dominant trees only, can bias absolute growth rates by up to 459% and trends in excess of 200%. Our findings challenge paradigms, where a subset of samples is typically considered to be representative for the entire population. The only two sampling strategies meeting the requirements for all types of investigations are the (i) sampling of all individuals within a fixed area; and (ii) fully randomized selection of trees. This result advertises the consistent implementation of a widely applicable sampling design to simultaneously reduce uncertainties in tree-ring-based quantifications of forest growth and increase the comparability of datasets beyond individual studies, investigators, laboratories, and geographical boundaries.

Therapeutic drug monitoring of once daily aminoglycoside dosing: comparison of two methods and investigation of the optimal blood sampling strategy.

PURPOSE Therapeutic drug monitoring of patients receiving once daily aminoglycoside therapy can be performed using pharmacokinetic (PK) formulas or Bayesian calculations. While these methods produced comparable results, their performance has never been checked against full PK profiles. We performed a PK study in order to compare both methods and to determine the best time-points to estimate AUC0-24 and peak concentrations (C max). METHODS We obtained full PK profiles in 14 patients receiving a once daily aminoglycoside therapy. PK parameters were calculated with PKSolver using non-compartmental methods. The calculated PK parameters were then compared with parameters estimated using an algorithm based on two serum concentrations (two-point method) or the software TCIWorks (Bayesian method). RESULTS For tobramycin and gentamicin, AUC0-24 and C max could be reliably estimated using a first serum concentration obtained at 1 h and a second one between 8 and 10 h after start of the infusion. The two-point and the Bayesian method produced similar results. For amikacin, AUC0-24 could reliably be estimated by both methods. C max was underestimated by 10-20% by the two-point method and by up to 30% with a large variation by the Bayesian method. CONCLUSIONS The ideal time-points for therapeutic drug monitoring of once daily administered aminoglycosides are 1 h after start of a 30-min infusion for the first time-point and 8-10 h after start of the infusion for the second time-point. Duration of the infusion and accurate registration of the time-points of blood drawing are essential for obtaining precise predictions.

A Parametric Test of Perfect Ranking in Balanced Ranked Set Sampling

Many techniques based on data which are drawn by Ranked Set Sampling (RSS) scheme assume that the ranking of observations is perfect. Therefore it is essential to develop some methods for testing this assumption. In this article, we propose a parametric location-scale free test for assessing the assumption of perfect ranking. The results of a simulation study in two special cases of normal and exponential distributions indicate that the proposed test performs well in comparison with its leading competitors.

Sparse computation for large-scale binary classification

Well-known data mining algorithms rely on inputs in the form of pairwise similarities between objects. For large datasets it is computationally impossible to perform all pairwise comparisons. We therefore propose a novel approach that uses approximate Principal Component Analysis to efficiently identify groups of similar objects. The effectiveness of the approach is demonstrated in the context of binary classification using the supervised normalized cut as a classifier. For large datasets from the UCI repository, the approach significantly improves run times with minimal loss in accuracy.