Parameter estimation in distributed models of blood-tissue Exchange: A monte carlo strategy to assess precision

Distributed parameter models of blood-tissue exchange are increasingly used to interpret multiple tracer dilution data in regional kinetic studies. To derive a measure of the precision with which the model parameters are estimated is there-fore of paramount importance. The standard approach to deriving precision of estimates does not take into account the fact that some of the model parameters are fixed. Thus, the precision of parameter estimates is not realistic and, in all likelihood, it is overestimated. The aim of this study is to describe a Monte Carlo method devised to obtain a theoretically sound measure of the precision of estimates, which takes into account both measurement error and the uncertainty associated with the fixed parameters. The fixed parameter values are taken from a probability distribution. By letting the fixed parameters vary according to their distribution, a large number of synthetic datasets is generated. Noise is then added. Estimating the parameters in each of these synthetic datasets allows the derivation of a Monte Carlo mean and standard deviation, which provides a realistic measure of precision. The methodology is illustrated for a simulated data case study dealing with the estimation of the capillary permeability-surface area product in a two tracer experiment.     

A 300 IR sublingual tablet is an effective,safe treatment for house dust mite–induced allergic rhinitis: An international,double-blind,placebo-controlled,randomized phase III clinical trial

1. Download : Download high-res image (382KB)
2. Download : Download full-size image

     

Sensitivity Analysis of Kernel Estimates: Implications in Nonlinear Physiological System Identification

Many techniques have been developed for the estimation of the Volterra/Wiener kernels of nonlinear systems, and have found extensive application in the study of various physiological systems. To date, however, we are not aware of methods for estimating the reliability of these kernels from single data records. In this study, we develop a formal analysis of variance for least-squares based nonlinear system identification algorithms. Expressions are developed for the variance of the estimated kernel coefficients and are used to place confidence bounds around both kernel estimates and output predictions. Specific bounds are developed for two such identification algorithms: Korenberg's fast orthogonal algorithm and the Laguerre expansion technique. Simulations, employing a model representative of the peripheral auditory system, are used to validate the theoretical derivations, and to explore their sensitivity to assumptions regarding the system and data. The simulations show excellent agreement between the variances of kernel coefficients and output predictions as estimated from the results of a single trial compared to the same quantities computed from an ensemble of 1000 Monte Carlo runs. These techniques were validated with white and nonwhite Gaussian inputs and with white Gaussian and nonwhite non-Gaussian measurement noise on the output, provided that the output noise source was independent of the test input.     

Removal of Cardiac Beat Artifact in Esophageal Pressure Measurement via a Modified Adaptive Noise Cancellation Scheme

Information on volume–pressure relationships of human lungs is usually based on indirect determination of intrapleural pressure (P_ip) obtained from the esophagus. Unfortunately, cardiac beat artifact frequently corrupts measurement of esophageal pressure (P_es). In this study, we presented a modified adaptive noise cancellation (MANC) scheme for removing the cardiac beat artifact in the P_es signal. The proposed methodology used an airflow signal as the reference signal with least-mean-square method as the adaptive algorithm. The results of six experiments on two Brown–Norway rats showed a significant reduction of the apparent cardiac pulsation with minimal distortion of the P_es signal. The MANC filter also showed evidence of peak suppression at integer multiples of heart rate in the fast Fourier transform of the P_es signal while leaving the remaining spectrum largely unperturbed. A t-test method and the ratio of standard deviation to mean (std/mean) statistics of airway resistance (R_aw) values were used to evaluate the performance of the MANC filter. In all six experiments, a reduction of std/mean of R_aw by 12.5%–68% was obtained, indicating the effectiveness of the proposed scheme. © 2001 Biomedical Engineering Society. PAC01: 8780-y, 8719Hh, 8719Uv     

Limitations on impulse conduction at the branch point of afferent axons in frog dorsal root ganglion

Summary Impulse conduction at the branch point of afferent axons in dorsal root ganglion (DRG) has been studied using intracellular recording from frog DRG neurons in vitro. The least conduction interval (LCI, the minimum inter-response interval) was determined for pairs of impulses to successfully propagate through the branch point into the dorsal root. At 21°–23°C, average branch point LCI was significantly longer than for afferent fibers in the peripheral nerve. This result suggested that the branch point would limit the maximum frequency of action potentials that could conduct into the dorsal root. This was found to be the case. The maximum frequency of impulses in short trains ( 40 ms) which could conduct into the dorsal root without failure (363 Hz) was accurately predicted by branch point LCI and was far less than the maximum frequency predicted from the LCI of axons in the peripheral nerve (610 Hz). Branch point LCI was correlated (r = -0.78) with the natural log of peripheral axon conduction velocity (CV). However, the relationship of LCI and CV was different for different types of neurons and the shape of the somatic action potential was found to be a reliable predictor of branch point LCI. Neurons with long-duration somatic action potentials with a shoulder on the falling phase tended to have low CV and invariably had long LCI's. Neurons with brief, smooth action potentials had short LCI's regardless of CV. These cells, which appear to be the most differentiated type, have found a way to minimize branch point LCI which is virtually independent of their axonal CV. For the latter neurons, branch point LCI was correlated (r = 0.42) with the reciprocal of the hyperpolarization level, at the cell body, required to block conduction through the branch point, suggesting that the proximity of the cell body to the branch point might play a role in determining the LCI of some neurons. Over a range of 12 °C to around 35°C, branch point LCI was inversely related and maximum firing frequency directly related to temperature. At high temperatures (30°–40°C) conduction failure occurred at sites having particularly long LCI's. It is concluded that a) these axon branch points act as lowpass filters and set the maximum frequency of conducted impulses that can access the central nervous system; b) certain varieties of DRG neurons exhibit more branch point filtering action than others; and c) warming, within limits, reduces branch point filtering action.     

S型曲线的一种新的拟合方法——拟分布函数法

目的 提出一类新的S型曲线--拟分布曲线,给出拟分布曲线的拟合方法。方法 用最小二乘法估计参数,算法采用Brent二分法合并Runge-Kutta数值积分方法。结果 应用NoSA软件,获得各种拟分布曲线参数估计值与拟合优度。结论 根据实例演算,表明拟正态分布曲线在尾部的拟合效果明显优于logistic曲线,且参数能得到较好的解释。     

偏最小二乘回归的原理及应用

[目的]探讨偏最小二乘回归的理论及其应用。[方法]应用医学实例计算,对偏最小二乘回归与主成分回归及一般最小二乘回归进行比较。[结果]偏最小二乘回归对数据的拟合度优于主成分回归和一般最小二乘回归法。[结论]偏最小二乘回归适用于处理有多重共线性的资料,当解释变量个数多、样本量少时尤为有效,是稳健的数据“软”建模的统计方法。     

A meta-analysis of the effects of placement on academic and social skill outcome measures of students with disabilities

This study involved an investigation of differences between outcome measures of students with disabilities placed in more integrated settings with those of students placed in less integrated settings. A meta-analysis was conducted using the findings from 24 studies published in peer-reviewed journals from 1980 through 2013. Results from the analyses suggest that there were significant differences (p < 0.0001) between placement settings with the majority of students with disabilities in more integrated settings outperforming those in less integrated settings on both academic and social outcome measures. Overall these findings, combined with those from two prior meta-analytic studies, provide evidence spanning over 80 years suggesting separate settings are not as beneficial as are more integrated settings. Implications related to practice and policy, as well as avenues for future study, are discussed.     

Exposure to nicotine and ethanol in adolescent mice: effects on depressive-like behavior during exposure and withdrawal

Depression and use of addictive substances are two of the most frequent public health problems of adolescents. However, little is known about the association between depression and drug use. Considering that ethanol and nicotine are the most widely used and abused drugs by adolescents, here, we evaluated the depressive-like behavior of C57BL/6 male and female mice exposed to nicotine (NIC) and/or ethanol (ETOH) from the 30th to the 45th (PN30-45) postnatal day. Four groups were analyzed: 1) concomitant NIC (50 μg/ml in 2% saccharin to drink) and ETOH (25%, 2 g/kg i.p. injected every other day) exposure; 2) NIC exposure; 3) ETOH exposure; 4) vehicle. Immobile behavior, an animal model of depressive behavior, was assessed in the forced swimming test (FST) while the anhedonic state was assessed in the sucrose preference test (SPT) by the end of exposure (PN45-47) as well as during short- (PN50-52) and long-term (PN75-77) withdrawal. In the FST, ETOH female mice showed a reduction in immobility time by the end of exposure while, during long-term withdrawal, immobility time was increased. Short-term withdrawal elicited an increase in immobility time only in female NIC mice. In the SPT, males from both NIC and NIC + ETOH groups showed increased sucrose consumption, suggesting a reward-craving effect during short-term withdrawal. During long-term withdrawal, NIC male mice showed an anhedonic effect. Adolescent nicotine, ethanol and nicotine + ethanol combined exposures during adolescence thus elicit gender-selective effects both during exposure and withdrawal that may contribute to the increased prevalence of depression among drug users.     

The lasso—a novel method for predictive covariate model building in nonlinear mixed effects models

Covariate models for population pharmacokinetics and pharmacodynamics are often built with a stepwise covariate modelling procedure (SCM). When analysing a small dataset this method may produce a covariate model that suffers from selection bias and poor predictive performance. The lasso is a method suggested to remedy these problems. It may also be faster than SCM and provide a validation of the covariate model. The aim of this study was to implement the lasso for covariate selection within NONMEM and to compare this method to SCM. In the lasso all covariates must be standardised to have zero mean and standard deviation one. Subsequently, the model containing all potential covariate–parameter relations is fitted with a restriction: the sum of the absolute covariate coefficients must be smaller than a value, t. The restriction will force some coefficients towards zero while the others are estimated with shrinkage. This means in practice that when fitting the model the covariate relations are tested for inclusion at the same time as the included relations are estimated. For a given SCM analysis, the model size depends on the P-value required for selection. In the lasso the model size instead depends on the value of t which can be estimated using cross-validation. The lasso was implemented as an automated tool using PsN. The method was compared to SCM in 16 scenarios with different dataset sizes, number of investigated covariates and starting models for the covariate analysis. Hundred replicate datasets were created by resampling from a PK-dataset consisting of 721 stroke patients. The two methods were compared primarily on the ability to predict external data, estimate their own predictive performance (external validation), and on the computer run-time. In all 16 scenarios the lasso predicted external data better than SCM with any of the studied P-values (5%, 1% and 0.1%), but the benefit was negligible for large datasets. The lasso cross-validation provided a precise and nearly unbiased estimate of the actual prediction error. On a single processor, the lasso was faster than SCM. Further, the lasso could run completely in parallel whereas SCM must run in steps. In conclusion, the lasso is superior to SCM in obtaining a predictive covariate model on a small dataset or on small subgroups (e.g. rare genotype). Run in parallel the lasso could be much faster than SCM. Using cross-validation, the lasso provides a validation of the covariate model and does not require the user to specify a P-value for selection.