Posterior Standard Deviation Research Articles

Effect of specialization on genetic parameters of studbook-entry inspection in Dutch Warmblood horses.

Recent studies on data from the Dutch Warmblood Studbook (KWPN) have shown that the ongoing specialization of horses for either dressage (DH) or show jumping (JH) has led to a decreasing genetic relationship between the two subpopulations. The aim of this study was to analyse the effect of the specialization process on the genetic parameters of traits measured in the studbook-entry inspection of KWPN during the last fifteen years. Data from 18,125 DH and 23,800 JH recorded from 1998 until 2013 were used to analyse 13 traits scored in both DH and JH. Analyses were performed in a Bayesian framework. Firstly, variance components were estimated based on the whole data set. Secondly, genetic correlations between traits measured in DH or JH were estimated using bivariate analyses. Thirdly, three time periods were defined and genetic correlations between subpopulations were estimated within each period. Heritability was moderate (0.17-0.39) for both DH and JH. Genetic correlations between traits measured in DH or JH were not different from one considering the posterior standard deviation of the estimation; however, in most of the traits, a clear trend in reduction of the genetic correlation for traits expressed in DH and JH and an increase in their posterior standard deviation for recent years was observed. These results suggest that specialization could lead to differences in traits measured in DH and JH in the recent years.

Understanding dynamic stability from pelvis accelerometer data and the relationship to balance and mobility in transtibial amputees

This study investigated whether pelvis acceleration-derived parameters can differentiate between dynamic stability states for transtibial amputees during level (LG) and uneven ground (UG) walking. Correlations between these parameters and clinical balance and mobility measures were also investigated. A convenience sample of eleven individuals with unilateral transtibial amputation walked on LG and simulated UG while pelvis acceleration data were collected at 100Hz. Descriptive statistics, Fast Fourier Transform, ratio of even to odd harmonics, and maximum Lyapunov exponent measures were derived from acceleration data. Of the 26 pelvis acceleration measures, seven had a significant difference (p≤0.05) between LG and UG walking conditions. Seven distinct, stability-relevant measures appeared in at least one of the six regression models that correlated accelerometer-derived measures to Berg Balance Scale (BBS), Community Balance and Mobility Scale (CBMS), and Prosthesis Evaluation Questionnaire (PEQ) scores, explaining up to 100% of the variability in these measures. Of these seven measures, medial–lateral acceleration range was the most frequent model variable, appearing in four models. Anterior–posterior acceleration standard deviation and stride time appeared in three models. Pelvis acceleration-derived parameters were able to differentiate between LG and UG walking for transtibial amputees. UG walking provided the most relevant data for balance and mobility assessment. These results could translate to point of patient contact assessments using a wearable system such as a smartbelt or accelerometer-equipped smartphone.

Longitudinal analysis of residual feed intake and BW in mink using random regression with heterogeneous residual variance

The aim of this study was to determine the genetic background of longitudinal residual feed intake (RFI) and BW gain in farmed mink using random regression methods considering heterogeneous residual variances. The individual BW was measured every 3 weeks from 63 to 210 days of age for 2139 male+female pairs of juvenile mink during the growing-furring period. Cumulative feed intake was calculated six times with 3-week intervals based on daily feed consumption between weighing’s from 105 to 210 days of age. Genetic parameters for RFI and BW gain in males and females were obtained using univariate random regression with Legendre polynomials containing an animal genetic effect and permanent environmental effect of litter along with heterogeneous residual variances. Heritability estimates for RFI increased with age from 0.18 (0.03, posterior standard deviation (PSD)) at 105 days of age to 0.49 (0.03, PSD) and 0.46 (0.03, PSD) at 210 days of age in male and female mink, respectively. The heritability estimates for BW gain increased with age and had moderate to high range for males (0.33 (0.02, PSD) to 0.84 (0.02, PSD)) and females (0.35 (0.03, PSD) to 0.85 (0.02, PSD)). RFI estimates during the growing period (105 to 126 days of age) showed high positive genetic correlations with the pelting RFI (210 days of age) in male (0.86 to 0.97) and female (0.92 to 0.98). However, phenotypic correlations were lower from 0.47 to 0.76 in males and 0.61 to 0.75 in females. Furthermore, BW records in the growing period (63 to 126 days of age) had moderate (male: 0.39, female: 0.53) to high (male: 0.87, female: 0.94) genetic correlations with pelting BW (210 days of age). The result of current study showed that RFI and BW in mink are highly heritable, especially at the late furring period, suggesting potential for large genetic gains for these traits. The genetic correlations suggested that substantial genetic gain can be obtained by only considering the RFI estimate and BW at pelting, however, lower genetic correlations than unity indicate that extra genetic gain can be obtained by including estimates of these traits during the growing period. This study suggests random regression methods are suitable for analysing feed efficiency and BW gain; and genetic selection for RFI in mink is promising.

Combining High-Resolution Optical and InSAR Features for Height Estimation of Buildings With Flat Roofs

In this paper, we contribute to answer the question: How accurately can we estimate heights of buildings with flat roofs given one high-resolution single-pass interferometric synthetic aperture radar (InSAR) image pair and one aerial orthophoto? What makes this problem challenging are the different sensor geometries and the sound stochastic combination of all available elevation cues. We revisit already existing methods and develop novel approaches to determine building heights. A rigorous stochastic approach based on least squares adjustment with functionally dependent parameters is introduced to combine all height measurements per building to one robust height estimate. Observation accuracies of the stochastic model are either taken from the literature or estimated empirically. A major benefit of adjustment is that it delivers a posterior standard deviation per height, which can be interpreted as a precision indicator and is of high relevance for practical applications. Estimated heights of an urban scene are compared to ground truth acquired with airborne laser scanning, allowing us to assess height accuracies that can be achieved under nearly optimal conditions. We conduct statistical tests that validate our model and show that a weighted combination of optical and synthetic aperture radar (SAR) data with least squares adjustment delivers reliable height estimates with meter accuracy for flat-roofed buildings. Additionally, we empirically estimate a confidence interval of the estimated heights that directly tells the user the security margin to be included, for example, in case of building evacuations for an anticipated flooding event, under the condition that the data and model have the same specifications as in this paper.

Genetic correlations between claw health and feet and leg conformation in Norwegian Red cows

The aim of this study was to estimate genetic correlations between claw disorders and feet and leg conformation traits in Norwegian Red cows. A total of 188,928 cows with claw health status recorded at claw trimming from 2004 to September 2013 and 210,789 first-lactation cows with feet and leg conformation scores from 2001 to September 2013 were included in the analyses. Traits describing claw health were corkscrew claw, infectious claw disorders (dermatitis, heel horn erosion, and interdigital phlegmon), and laminitis-related claw disorders (sole ulcer, white line disorder, and hemorrhage of sole and white line). The feet and leg conformation traits were rear leg rear view (new and old definition), rear leg side view, foot angle, and hoof quality. Feet and leg conformation traits were scored linearly from 1 to 9, with optimum scores depending on the trait. Claw disorders were defined as binary (0/1) traits for each lactation. Threshold sire models were used to model claw disorders, whereas the feet and leg conformation traits were described by linear sire models. Three multivariate analyses were performed, each including the 5 feet and leg conformation traits and 1 of the 3 claw disorders at a time. Posterior means of heritability of liability of claw disorders ranged from 0.10 to 0.20 and heritabilities of feet and leg conformation traits ranged from 0.04 to 0.11. Posterior standard deviation of heritability was ≤0.01 for all traits. Genetic correlations between claw disorders and feet and leg conformation traits were all low or moderate, except between corkscrew claw and hoof quality (−0.86), which are supposed to measure the same trait. The genetic correlations between rear leg rear view (new) and infectious claw disorders (−0.20) and laminitis-related claw disorders (0.26), and between hoof quality and laminitis-related claw disorders (−0.33) were moderate. Eight of the 15 genetic correlations between claw disorders and feet and leg conformation traits had 0 included in the 95% highest posterior density interval. These results imply that selection for feet and leg conformation is not an efficient approach to genetically improve claw health in Norwegian Red cattle.

Explaining share price disparity with parameter uncertainty: Evidence from Chinese A- and H-shares

The price disparity between the A- and H-share markets for dual-listed firms in China is one of the most intriguing puzzles in the Mainland and Hong Kong financial markets. In this paper, we revisit this price disparity puzzle using the channel of parameter uncertainty. In the presence of information asymmetry and market segmentation, investors have different views on a firm’s asset volatility, and hence different valuations of the same reference firm. We estimate a structural model for equity pricing using a Bayesian approach, in which the uncertainty of investor model parameters is represented by the posterior standard deviation of the firm’s asset volatility. Our regression analysis shows that in addition to other market-based and macro factors, parameter uncertainty explains variations in price disparity.

Determining the preferred modality for real-time biofeedback during balance training

Vestibular rehabilitation therapy has been shown to improve balance and gait stability in individuals with vestibular deficits. However, patient compliance with prescribed home exercise programs is variable. Real-time feedback of exercise performance can potentially improve exercise execution, exercise motivation, and rehabilitation outcomes. The goal of this study is to directly compare the effects of visual and vibrotactile feedback on postural performance to inform the selection of a feedback modality for inclusion in a home-based balance rehabilitation device. Eight subjects (46.6±10.6years) with peripheral vestibular deficits and eight age-matched control subjects (45.3±11.1years) participated in the study. Subjects performed eyes-open tandem Romberg stance trials with (vibrotactile, discrete visual, continuous visual, and multimodal) and without (baseline) feedback. Main outcome measures included medial–lateral (M/L) and anterior–posterior mean and standard deviation of body tilt, percent time spent within a no-feedback zone, and mean score on a comparative ranking survey. Both groups improved performance for each feedback modality compared to baseline, with no significant differences in performance observed among vibrotactile, discrete visual, or multimodal feedback for either group. Subjects with vestibular deficits performed best with continuous visual feedback and ranked it highest. Although the control subjects performed best with continuous visual feedback in terms of mean M/L tilt, they ranked it lowest. Despite the observed improvements, continuous visual feedback involves tracking a moving target, which was noted to induce dizziness in some subjects with vestibular deficits and cannot be used during exercises in which head position is actively changed or during eyes-closed conditions.

Relationship between body condition score and health traits in first-lactation Canadian Holsteins

The objective of this research was to estimate daily genetic correlations between longitudinal body condition score (BCS) and health traits by using a random regression animal model in first-lactation Holsteins. The use of indicator traits may increase the rate of genetic progress for functional traits relative to direct selection for functional traits. Indicator traits of interest are those that are easier to record, can be measured early in life, and are strongly genetically correlated with the functional trait of interest. Several BCS records were available per cow, and only 1 record per health trait (1=affected; 0=not affected) was permitted per cow over the lactation. Two bivariate analyses were performed, the first between BCS and mastitis and the second between BCS and metabolic disease (displaced abomasum, milk fever, and ketosis). For the first analysis, 217 complete herds were analyzed, which included 28,394 BCS records for 10,715 cows and 6,816 mastitis records for 6,816 cows. For the second analysis, 350 complete herds were analyzed, which included 42,167 BCS records for 16,534 cows and 13,455 metabolic disease records for 13,455 cows. Estimation of variance components by a Bayesian approach via Gibbs sampling was performed using 400,000 samples after a burn-in of 150,000 samples. The average daily heritability (posterior standard deviation) of BCS was 0.260 (0.026) and the heritabilities of mastitis and metabolic disease were 0.020 (0.007) and 0.041 (0.012), respectively. Heritability estimates were similar to literature values. The average daily genetic correlation between BCS and mastitis was −0.730 (0.110). Cows with a low BCS during the lactation are more susceptible to mastitis, and mastitic cows are likely to have low BCS. Daily estimates of genetic correlations between BCS and mastitis were moderate to strong throughout the lactation, becoming stronger as the lactation progressed. The average daily genetic correlation between BCS and metabolic disease was −0.438 (0.125), and was consistent throughout the lactation. A lower BCS during the lactation is genetically associated with the occurrence of mastitis and metabolic disease.

Genetic parameters for nematode resistance in periparturient ewes and post-weaning lambs in Uruguayan Merino sheep

Gastrointestinal (GI) parasites are one of the main sanitary and economic constraints for sheep production worldwide. In Uruguay, resistance to GI parasites has been included in genetic evaluations of Merino and Corriedale breeds since 1994, using Faecal worm Egg Count measured in post-weaning lambs (lambFEC) as a selection criterion. Although adult categories are more resistant to GI parasites, a temporary loss of acquired immunity is present in the periparturient period (=periparturient rise). The purpose of the present study is to estimate the genetic parameters of resistance to GI parasites in periparturient ewes and post-weaning Merino lambs. A total 2110 faecal samples of 748 periparturient ewes (eweFEC), the progeny of 107 sires, were collected in 2009 and 2010. 9458 lambFEC records from 7506 lambs born between 2001 and 2009 were analysed as well. (Co)variance components and systematic effects were estimated using a multi-trait animal model, with a Bayesian analysis using the Gibbs sampler algorithm. Direct and correlated responses (DR and CR) of eweFEC to selection using eweFEC and lambFEC respectively, were also estimated. Posterior medians (posterior standard deviation) for heritability (h2) and repeatability were 0.25 (0.03) and 0.34 (0.02) for lambFEC, and 0.08 (0.03) and 0.18 (0.03) for eweFEC, respectively. Posterior median for genetic correlation between both traits was 0.81 (0.11). CR obtained by selecting for lambFEC was two times more efficient than DR from selection by eweFEC. In conclusion, eweFEC has a lower h2 than lambFEC while their genetic correlation is moderate to high. Therefore, indirect selection by using lambFEC will be more effective than direct selection on eweFEC, and will produce ewes that eliminate less worm eggs, resulting in less pasture contamination.

Genetic correlation of longevity with growth, post-mortem, docility and some morphological traits in the Pirenaica beef cattle breed

Survival or longevity is an economically relevant trait in cattle. However, it is not currently included in cattle selection criteria because of the delayed recording of phenotypic data and the high computational demand of survival techniques under proportional hazard models. The identification of longevity-correlated traits that can be early registered in lifetime would therefore be very useful for beef cattle selection processes. The aim of this study was to estimate the genetic correlation of survival (SURV) with: growth – birth weight (BW), weight at 120 days (W120), weight at 210 days (W210); carcass – cold carcass weight (CCW), conformation (CON), fatness (FAT) and meat colour (COL); teat morphology – teat thickness (TT), teat length (TL) and udder depth (UD); leg morphology – forward (FL) and backward legs (BL); milk production (MILK) and docility (DOC). In the statistical analysis, SURV was measured in discrete-time intervals and modelled via a sequential threshold model. A series of independent bivariate Bayesian analyses between cow survival and each recorded trait were carried out. The posterior mean estimates (and posterior standard deviation) for the heritability of SURV was 0.05 (0.01); and for the relevant genetic correlations with SURV were 0.07 (0.04), 0.12 (0.05), 0.10 (0.05), 0.15 (0.05), −0.18 (0.06), 0.33 (0.06) and 0.27 (0.15) for BW, W120, W210, CCW, CON, FAT and COL, respectively.

Determinants and Dynamics of Price Disparity in Onshore and Offshore Renminbi Forward Exchange Rate Markets

Price disparities between the renminbi onshore deliverable forward and offshore non-deliverable forward exchange rates is an intriguing puzzle in financial economics. This paper investigates the determinants of these price disparities focusing on the possibility of parameter uncertainty. In the presence of information asymmetry and market segmentation among onshore and offshore investors, it is possible that they formulate different views on the Mainland economy which translate into a different assessment of the outlook for Mainland interest rates. Through a no arbitrage condition that relates the forward rate to the spot rate and interest rate differential, a different assessment of the path of interest rates can lead to a different valuation of forward prices. We estimate a term structure model for the implied renminbi interest rate using a Bayesian approach, in which investors’ model parameter uncertainty is represented by the posterior standard deviation of the volatility of the interest rate. We show that parameter uncertainty can help to explain price disparities, in addition to market-wide aggregate uncertainty and illicit capital flows in the Mainland’s balance of payment.

Explaining Share Price Disparity with Parameter Uncertainty: Evidence from Chinese A- and H-Shares

The price disparity between the A- and H-share markets for dual-listed firms in China is one of the most intriguing puzzles in the mainland and Hong Kong financial markets. In this paper, we revisit this price disparity puzzle using the channel of parameter uncertainty. In the presence of information asymmetry and market segmentation, investors have different views on a firm's asset volatility, and hence different valuations of the same reference firm. We estimate a structural model for equity pricing using a Bayesian approach, in which the uncertainty of investor model parameters is represented by the posterior standard deviation of the firm's asset volatility. Our regression analysis shows that in addition to other market-based and macro factors, parameter uncertainty explains variations in price disparity.

Rethinking Analytical Strategies for Surveillance of Medical Devices

Randomized trials that sometimes serve as the basis for device approval are small, short term, and generalizable to an increasingly smaller percentage of patients. Some of the most common and challenging devices are those used in hip replacement. Artificial hips are implanted in thousands to alleviate pain caused by noninflammatory joint disease and to restore patient mobility. During 2004 in the United States, although 68% of hospital stays for partial or total hip replacements were for those aged 65 years and older, younger patients will account for 52% by 2030. Using hierarchical modeling, we propose a framework for combining information from premarket and postmarket settings. Our key assumption is that device performance characteristics and outcomes obtained from 1 cohort are related to device characteristics and outcomes of the same or similar devices observed in other cohorts. We illustrate methods by jointly modeling Harris Hip Scores (HHSs) and revision-success data from 1851 subjects who participated in 3 pivotal randomized or observational studies of artificial hips. Subjects participating in randomized studies had better 2-year HHS than those in observational studies (posterior mean increase in HHS = 4.1, posterior standard deviation = 0.6). Patients implanted with ceramic-on-polyethylene hip used in 1 study had higher 2-year HHS than those implanted with a different ceramic-on-polyethylene hip in another study (mean difference = 4.2, standard deviation = 0.6). Our approach is feasible and will advance regulatory science using a transparent and dynamic new paradigm for knowledge management throughout the total product life cycle.

A Genetic Analysis of Mortality in Pigs

An analysis of mortality is undertaken in two breeds of pigs: Danish Landrace and Yorkshire. Zero-inflated and standard versions of hierarchical Poisson, binomial, and negative binomial Bayesian models were fitted using Markov chain Monte Carlo (MCMC). The objectives of the study were to investigate whether there is support for genetic variation for mortality and to study the quality of fit and predictive properties of the various models. In both breeds, the model that provided the best fit to the data was the standard binomial hierarchical model. The model that performed best in terms of the ability to predict the distribution of stillbirths was the hierarchical zero-inflated negative binomial model. The best fit of the binomial hierarchical model and of the zero-inflated hierarchical negative binomial model was obtained when genetic variation was included as a parameter. For the hierarchical binomial model, the estimate of the posterior mean of the additive genetic variance (posterior standard deviation in brackets) at the level of the logit of the probability of a stillbirth was 0.173(0.039) in Landrace and 0.202(0.048) in Yorkshire. The implications of these results from a breeding perspective are briefly discussed.

Genetic parameters for anovulation and pregnancy loss in dairy cattle

The objectives were to estimate heritabilities and genetic variances for anovulation at ∼50 d in milk and pregnancy loss occurring between first and second pregnancy diagnoses after artificial insemination. Data were originally collected for trials on reproductive management. Anovulation data consisted of 5,818 records from 13 studies in 8 herds with an overall prevalence of 23.3%. A Bayesian approach using Markov Chain Monte Carlo methods was used in mixed threshold models for both traits. The statistical model for anovulation included fixed effects [parity, herd-study-treatment, and body condition score (BCS)], covariates (inbreeding and milk yield), and random effects (sire and residual). A second statistical model included all terms in the first model except BCS. In addition, 2 bivariate, mixed sire models were used to analyze anovulation with BCS and anovulation with milk yield. The posterior mean heritability estimate for anovulation was 0.171 [posterior standard deviation (PSD)=0.052]. Correlations of anovulation with milk yield were as follows: genetic=0.168, PSD=0.187; residual=−0.046, PSD=0.022; and phenotypic=−0.036. Bivariate analysis of BCS with anovulation showed a genetic correlation (−0.301, PSD=0.177) and phenotypic correlation (−0.192, PSD=0.019). Pregnancy-loss data consisted of 3,775 records from 14 studies in 8 herds with an overall prevalence of 14.4%. Analysis of pregnancy loss used a sire–maternal grandsire threshold model with embryo survival as the subject of analysis. Independent variables consisted of fixed effects (parity and herd-study), covariates (embryo and maternal inbreeding), and random effects (sire of embryo, maternal grandsire of embryo, and residual). In addition, separate sire models were analyzed using embryo as the subject and cow as the subject of analysis. The sire–maternal grandsire model yielded a heritability for direct effect of 0.489 (PSD=0.221) and for maternal effects of 0.166 (PSD=0.113). In this study, the breeding value variance for embryo effects was 3 times the breeding value variance for maternal effects, indicating that, at the level of breeding values, the embryo's ability to survive has a greater effect on pregnancy loss than does the cow's ability to maintain the pregnancy. These results suggest that genetic improvement of reproductive performance could be enhanced by selection for fundamental measures such as abnormally long periods of postpartum anovulation and pregnancy loss. Larger studies of these traits are needed to obtain parameter estimates with greater precision.

Genetic determination of the onset of heat stress on daily milk production in the US Holstein cattle

Existence of individual variation in the onset of heat stress for daily milk yield of dairy cows was assessed. Data included 353,376 test-day records of 38,383 first-parity Holsteins from a random sample of US herds. Three hierarchical models were investigated. Model 1 inferred the value of a temperature-humidity index (THI) at which mean yield began to decline as well as the extent of that decline. Model 2 assumed individual variation in yield decline beyond a common THI threshold. Model 3 additionally assumed individual variation for the onset of heat stress. Deviance information criteria indicated the superiority of model 3 over model 2. For model 2, genetic correlation between milk yield in the absence of heat stress and the THI threshold for heat stress was −0.4 (0.11) [marginal posterior mean (marginal posterior standard deviation)]. For model 3, genetic correlations were −0.53 (0.05) between milk yield and THI threshold and −0.62 (0.08) between milk yield and yield decay beyond the THI threshold. Total standard deviation (sum of additive genetic and permanent environmental standard deviations) for the THI threshold was 3.95 (0.06), and more than half of that variation had an additive genetic origin [56% (5%)]. Because of the high genetic correlation [0.95 (0.03)] between yield decay and THI threshold with model 3, using only one of them as a selection criterion for heat tolerance would modify the other in the desired direction.

Prediction in Multilevel Generalized Linear Models

SummaryWe discuss prediction of random effects and of expected responses in multilevel generalized linear models. Prediction of random effects is useful for instance in small area estimation and disease mapping, effectiveness studies and model diagnostics. Prediction of expected responses is useful for planning, model interpretation and diagnostics. For prediction of random effects, we concentrate on empirical Bayes prediction and discuss three different kinds of standard errors; the posterior standard deviation and the marginal prediction error standard deviation (comparative standard errors) and the marginal sampling standard deviation (diagnostic standard error). Analytical expressions are available only for linear models and are provided in an appendix. For other multilevel generalized linear models we present approximations and suggest using parametric bootstrapping to obtain standard errors. We also discuss prediction of expectations of responses or probabilities for a new unit in a hypothetical cluster, or in a new (randomly sampled) cluster or in an existing cluster. The methods are implemented in gllamm and illustrated by applying them to survey data on reading proficiency of children nested in schools. Simulations are used to assess the performance of various predictions and associated standard errors for logistic random-intercept models under a range of conditions.

Bayesian Analysis of Severe Acute Respiratory Syndrome: The 2003 Hong Kong Epidemic

This paper analyzes data arising from a Severe Acute Respiratory Syndrome (SARS) epidemic in Hong Kong in 2003 involving 1755 cases. A discrete time stochastic model that uses a back‐projection approach is proposed. Markov Chain Monte Carlo (MCMC) methods are developed for estimation of model parameters. The algorithm is further extended to integrate numerically over unobserved variables of the model. Applying the method to SARS data from Hong Kong, a value of 3.88 with a posterior standard deviation of 0.09 was estimated for the basic reproduction number. An estimate of the transmission parameter at the beginning of the epidemic was also obtained as 0.149 with a posterior standard deviation of 0.003. A reduction in the transmission parameter during the course of the epidemic forced the effective reproduction number to cross the threshold value of one, seven days after control interventions were introduced. At the end of the epidemic, the effective reproduction number was as low as 0.001 suggesting that the epidemic was brought under control by the intervention measures introduced. (© 2008 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)

Strain Dyssynchrony Index Correlates With Improvement in Left Ventricular Volume After Cardiac Resynchronization Therapy Better Than Tissue Velocity Dyssynchrony Indexes

Various dyssynchrony indexes derived from tissue velocity and strain imaging have been proposed to predict the effectiveness of cardiac resynchronization therapy (CRT). We sought to compare the effect of CRT on dyssynchrony indexes derived by tissue velocity and strain and to determine which baseline intraventricular dyssynchrony parameters correlate with improvement in left ventricular volume after CRT. Echocardiography with tissue Doppler imaging was performed in 45 patients with systolic heart failure at baseline, 1 day after CRT, and a median of 6 months after CRT. We calculated septal-lateral delay and anteroseptal-posterior delay and standard deviation of time to peak systolic velocity in the 12 basal and mid-left ventricular segments (Tv-SD). The standard deviation for time to peak strain in the 12 basal and mid-left ventricular segments (Tepsilon-SD) was calculated as a strain-derived dyssynchrony index. None of the tissue velocity-derived dyssynchrony indexes improved after CRT (septal-lateral delay, P=0.39; anteroseptal-posterior delay, P=0.46; Tv-SD, P=0.30), whereas Tepsilon-SD decreased significantly after CRT (P<0.001). Improvement in Tepsilon-SD 1 day after CRT correlated with the reduction in end-systolic volume at follow-up (r=0.66; P<0.001). Baseline Tepsilon-SD demonstrated significant correlation with the reduction of end-systolic volume at follow-up (r=0.57; P<0.001); however, baseline tissue velocity-derived dyssynchrony indexes failed to predict the effect of CRT. The strain-derived dyssynchrony index is a better measurement than the tissue velocity dyssynchrony index for monitoring changes in mechanical dyssynchrony after CRT and for predicting reduction in left ventricular volume after CRT.

Variability and uncertainty of biokinetic model parameters: the discrete empirical Bayes approximation

In the Bayesian approach to internal dosimetry, uncertainty and variability of biokinetic model parameters need to be taken into account. The discrete empirical Bayes approximation replaces integration over biokinetic model parameters by discrete summation in the evaluation of Bayesian posterior averages using Bayes theorem. The discrete choices of parameters are taken as best-fit point determinations of model parameters for a study subpopulation with extensive data. A simple heuristic model is constructed to numerically and theoretically study this approximation. The heuristic example is the measurement of heights of a group of people, say from a photograph where measurement uncertainty is significant. A comparison is made of posterior mean and standard deviation of height after a measurement, (i) using the exact prior describing the distribution of true height in the population and (ii) using the approximate discrete empirical Bayes prior obtained from measurements of some study subpopulation.