Generalized Bayes minimax estimation of the normal mean matrix with unknown covariance matrix

This paper addresses the problem of estimating the normal mean matrix in the case of unknown covariance matrix. This problem is solved by considering generalized Bayesian hierarchical models. The resulting generalized Bayes estimators with respect to an invariant quadratic loss function are shown to be matricial shrinkage equivariant estimators and the conditions for their minimaxity are given.     

Bayesian Variable and Transformation Selection in Linear Regression

This article suggests a method for variable and transformation selection based on posterior probabilities. Our approach allows for consideration of all possible combinations of untransformed and transformed predictors along with transformed and untransformed versions of the response. To transform the predictors in the model, we use a change-point model, or “change-point transformation,” which can yield more interpretable models and transformations than the standard Box–Tidwell approach. We also address the problem of model uncertainty in the selection of models. By averaging over models, we account for the uncertainty inherent in inference based on a single model chosen from the set of models under consideration. We use a Markov chain Monte Carlo model composition (MC³) method which allows us to average over linear regression models when the space of models under consideration is very large. This considers the selection of variables and transformations at the same time. In an example, we show that model averaging improves predictive performance as compared with any single model that might reasonably be selected, both in terms of overall predictive score and of the coverage of prediction intervals. Software to apply the proposed methodology is available via StatLib.     

Posterior cordectomy and subtotal arytenoidectomy for the treatment of bilateral vocal fold immobility: Functional results

We report vocal and respiratory results following endoscopic CO₂ laser therapy for bilateral vocal fold immobility in adduction. Two techniques were used: posterior cordectomy (PC) and subtotal arytenoidectomy (SA). Respiratory improvement was demonstrated by the peak expiratory flow/peak inspiratory flow ratio (PEF/PIF, normal = 1), which was less than 2 for 83% of patients following PC and for 81% following SA. As for vocal results, there were no significant quantitative differences between the two techniques. Mean maximum phonation time (/a/) was 6.8 ± 2.6 s after SA and 7.8 ± 1.6 s following PC. The phonation quotient was 288 ± 116 ml/s after SA and 304 ± 92 ml/s after PC. Mean vocal intensity was 62 ± 4 dB after SA and 59 ± 3 dB after PC. Vocal quality was measured by high-resolution vocal frequency analysis, as represented by a histogram. Peaks corresponding to fundamental frequency and first harmonics were preserved in more than 60% of patients in the two groups. Vocal preservation is better when the paralyzed folds are in the paramedian position, with the possibility of adduction (Gerhardt syndrome). SA is performed in our procedure, though it is longer and more difficult to perform than PC. PC often requires two procedures to achieve satisfactory results.     

图象复原的新后验方法

本文就某系统的动态图象的复原,阐述了复原的主要技术过程。提出了一种新的后验模型,即退化信息不是从退化图象本身中提取,而是从给定样本的一系列退化象中提取,从而可以用线性空不变系统的求解模型来处理非线性空变系统的图象复原问题。本文给出了用此方法所获得的处理结果。     

正态分布的共轭分布及贝叶斯估计

本文介绍含有一个或两个未知参数的正态分布Ｎ（μ，）的共轭分布，以及对正态总体的未知参数进行估计的贝叶斯方法。     

Differences Between Flexible and Rigid Endoscopy in Assessing the Posterior Glottic Chink

Objective

To determine whether different modalities of laryngeal examination produce differences in the assessment of the posterior glottic chink (PGC), and whether the prevalence of PGC differs by gender.

Introduction

The PGC has been described as a triangular laryngeal space between the posterior laryngeal wall and the vocal processes during glottic closure found commonly in human females, but less often in males. The purpose of this study was not only to identify whether there are gender differences in prevalence of posterior glottic but also to determine whether there is a difference in detection of this configuration dependent on the modality of laryngeal imaging, specifically flexible nasopharyngolaryngoscopy (FNPL) versus rigid laryngoscopy (RL).

Methods

A review of 104 consecutive initial laryngeal examinations was performed. All patients underwent both flexible laryngoscopy (FL) and RL performed under stroboscopic light. Patients with immobile vocal folds, masses causing glottic gaps, atrophy, or severe muscle tension dysphonia causing an inability to fully visualize the entire length of the vocal fold were excluded. In the remaining patients, the posterior glottic configuration showing a posterior chink in relation to the vocal process was graded on a 0–4 scale (called the Posterior Glottic Closure Score [PGCS]); 0 was used to indicate a closed glottis and 4 the most open configuration without creating a complete glottic gap. PGCSs for males were compared with those of females, and the PGCSs obtained by flexible nasopharyngolaryngoscopy was compared with RL.

Results

Fifty-two patients were included in the study. Twenty-four of the patients were male, and 28 were female. The average age of the patients was 48.4 years (SD ± 17.35), and the range was 15–81 years. On RL, eight males had a PGCS 1–4, that is, evidence of PGC, and 23 females had a PGCS 1–4. On flexible nasopharyngolaryngoscopy, only four males had a PGCS 1–4 and 24 females had PGCS 1–4. Twenty-two females had a PGC detected by both modalities, and the PGCS was significantly higher using RL (2.73 ± 0.70 vs 2.14 ± 0.834) than FL. For the four males in which PGC was detected by both modalities, there was no statistical significance when comparing the PGCS between RL and FL (1.75 ± 0.96 vs 1.75 ± 0.5, P < 0.05). When comparing only males and females who had a PGC (PGCS 1–4), females had a higher PGCS (2.65 ± 0.78) than males (1.75 ± 0.71, P < 0.05) on RL, indicating a more open posterior glottis in females. On flexible examination, there was no difference detected in the average PGCS, 2.08 ± 0.83 for females and males 1.75 ± 0.50. PGCs were more common in younger (age 43 years) than older (age 54–56 years) subjects for both laryngoscopic modalities.

Conclusion

From this pilot study, we determined that there is a difference in male and female PGC prevalence and size. PGC is more common in females than males. Prevalence (or the detection rate) is about the same with RL and FL in females, but higher with RL than FL in males. The average score of the glottic opening, when present, was statistically significantly different between RL and FL in females but not in males. Furthermore, females had a larger PGCS on both modalities when compared with males, although this difference was only found to be statistically significant on RL; and complete glottic closure was more common in older than in younger subjects.     

基于动脉自旋标记技术的正常大脑后循环血流分析

目的 利用动脉自旋标记技术对正常大脑后循环进行MRI 灌注成像,测量正常大脑后循环血流量,提供正常大脑后循环供血区血流量参考值。方法 对60 例志愿者进行MRI 常规扫描和3D-ASL 灌注扫描,获得血流量图,分别测得双侧枕叶、颞叶后部、小脑半球及脑干血流量（cerebral blood flow,CBF）,结果进行统计分析。结果 测得血流量,双侧枕叶为（64.45±12.72）ml/（min·100g）,颞叶后部为（63.28±14.97）ml/（min·100g）,小脑半球为（61.75±13.54）ml/（min·100g）,脑干为（53.64±12.71）ml/（min·100g）,男性与女性的血流量未见差异;不同年龄段的血流量存在差异。结论 3D-ASL 能定量测量后循环血流量,对脑梗死的病理生理学的进一步研究有着重要价值。     

地声参数及传播损失不确定性估计与建模

地声参数的不确定性对水声传播具有重要的影响。通过贝叶斯理论建立水声环境不确定性推理模型,理论推导了地声参数的似然函数以及地声参数和传播损失的后验概率密度,并采用MCMC(Markov Chain Monte Carlo)进行了仿真计算,给出了地声参数的二维后验联合概率密度和一维边缘概率密度,在此基础上对传播损失的不确定性进行了估计,得到了传播损失80%的可信区间。仿真和实验结果表明,该方法适用于地声参数反演和不确定性估计,并能获取因地声参数不确定性导致的传播损失不确定性估计。     

Disrupted cortical connectivity theory as an explanatory model for autism spectrum disorders

Recent findings of neurological functioning in autism spectrum disorder (ASD) point to altered brain connectivity as a key feature of its pathophysiology. The cortical underconnectivity theory of ASD (Just et al., 2004) provides an integrated framework for addressing these new findings. This theory suggests that weaker functional connections among brain areas in those with ASD hamper their ability to accomplish complex cognitive and social tasks successfully. We will discuss this theory, but will modify the term underconnectivity to ‘disrupted cortical connectivity’ to capture patterns of both under- and over-connectivity in the brain. In this paper, we will review the existing literature on ASD to marshal supporting evidence for hypotheses formulated on the disrupted cortical connectivity theory. These hypotheses are: 1) underconnectivity in ASD is manifested mainly in long-distance cortical as well as subcortical connections rather than in short-distance cortical connections; 2) underconnectivity in ASD is manifested only in complex cognitive and social functions and not in low-level sensory and perceptual tasks; 3) functional underconnectivity in ASD may be the result of underlying anatomical abnormalities, such as problems in the integrity of white matter; 4) the ASD brain adapts to underconnectivity through compensatory strategies such as overconnectivity mainly in frontal and in posterior brain areas. This may be manifested as deficits in tasks that require frontal–parietal integration. While overconnectivity can be tested by examining the cortical minicolumn organization, long-distance underconnectivity can be tested by cognitively demanding tasks; and 5) functional underconnectivity in brain areas in ASD will be seen not only during complex tasks but also during task-free resting states. We will also discuss some empirical predictions that can be tested in future studies, such as: 1) how disrupted connectivity relates to cognitive impairments in skills such as Theory-of-Mind, cognitive flexibility, and information processing; and 2) how connection abnormalities relate to, and may determine, behavioral symptoms hallmarked by the triad of Impairments in ASD. Furthermore, we will relate the disrupted cortical connectivity model to existing cognitive and neural models of ASD.