A semiparametric mixed-effects model for censored longitudinal data

In longitudinal studies involving laboratory-based outcomes, repeated measurements can be censored due to assay detection limits. Linear mixed-effects (LMEs) models are a powerful tool to model the relationship between a response variable and covariates in longitudinal studies. However, the linear parametric form of linear mixed-effect models is often too restrictive to characterize the complex relationship between a response variable and covariates. More general and robust modeling tools, such as nonparametric and semiparametric regression models, have become increasingly popular in the last decade. In this article, we use semiparametric mixed models to analyze censored longitudinal data with irregularly observed repeated measures. The proposed model extends the censored linear mixed-effect model and provides more flexible modeling schemes by allowing the time effect to vary nonparametrically over time. We develop an Expectation-Maximization (EM) algorithm for maximum penalized likelihood estimation of model parameters and the nonparametric component. Further, as a byproduct of the EM algorithm, the smoothing parameter is estimated using a modified linear mixed-effects model, which is faster than alternative methods such as the restricted maximum likelihood approach. Finally, the performance of the proposed approaches is evaluated through extensive simulation studies as well as applications to data sets from acquired immune deficiency syndrome studies.

Nintedanib in idiopathic and secondary pleuroparenchymal fibroelastosis

Background Pleuroparenchymal fibroelastosis (PPFE) has a variable disease course with dismal prognosis in the majority of patients with no validated drug therapy. This study is to evaluate the effect of nintedanib in patients with idiopathic and secondary PPFE. Patients admitted to a tertiary care center (2010–2019) were included into this retrospective analysis if they had a multidisciplinary diagnosis of PPFE, had been followed-up for 3 months or more, and had lung function tests and chest CTs available for review. Changes in pulmonary function tests were assessed using non-parametric tests and linear mixed effect model. Lung volumes were measured with lobar segmentation using chest CT. Results Out of 21 patients with PPFE, nine had received nintedanib, six had received another treatment and another six patients were monitored without drug therapy. Annual FVC (% of predicted) relative decline was − 13.6 ± 13.4%/year before nintedanib and − 1.6 ± 6.02%/year during nintedanib treatment (p = 0.014), whereas no significant change in FVC% relative decline was found in patients receiving another treatment (− 13.25 ± 34 before vs − 16.61 ± 36.2%/year during treatment; p = 0.343). Using linear mixed effect model, the slope in FVC was − 0.97%/month (95% CI: − 1.42; − 0.52) before treatment and − 0.50%/month (95% CI: − 0.88; 0.13) on nintedanib, with a difference between groups of + 0.47%/month (95% CI: 0.16; 0.78), p = 0.004. The decline in the upper lung volumes measured by CT was − 233 mL/year ± 387 mL/year before nintedanib and − 149 mL/year ± 173 mL/year on nintedanib (p = 0.327). Nintedanib tolerability was unremarkable. Conclusion In patients with PPFE, nintedanib treatment might be associated with slower decline in lung function, paving the way for prospective, controlled studies.

Lamina cribrosa vessel and collagen beam networks are distinct

Our goal was to analyze the spatial interrelation between vascular and collagen networks in the lamina cribrosa (LC). Specifically, we quantified the percentages of collagen beams with/without vessels and of vessels inside/outside of collagen beams. To do this, the vasculature of six normal monkey eyes was labelled by perfusion post-mortem. After enucleation, coronal cryosections through the LC were imaged using fluorescence and polarized light microscopy to visualize the blood vessels and collagen beams, respectively. The images were registered to form 3D volumes. Beams and vessels were segmented, and their spatial interrelationship was quantified in 3D. We found that 22% of the beams contained a vessel (range 14% to 32%), and 21% of vessels were outside beams (13% to 36%). Stated differently, 78% of beams did not contain a vessel (68% to 86%), and 79% of vessels were inside a beam (64% to 87%). Individual monkeys differed significantly in the fraction of vessels outside beams (p<0.01 by linear mixed effect analysis), but not in the fraction of beams with vessels (p>0.05). There were no significant differences between contralateral eyes in the percent of beams with vessels and of vessels outside beams (p>0.05). Our results show that the vascular and collagenous networks of the LC in monkey are clearly distinct, and the historical notions that each LC beam contains a vessel and all vessels are within beams are inaccurate. We postulate that vessels outside beams may be relatively more vulnerable to mechanical compression by elevated IOP than are vessels shielded inside of beams.

Detecting Injury Risk Factors with Algorithmic Models in Elite Women’s Pathway Cricket

This exploratory retrospective cohort analysis aimed to explore how algorithmic models may be able to identify important risk factors that may otherwise not have been apparent. Their association with injury was then assessed with more conventional data models. Participants were players registered on the England and Wales Cricket Board women’s international development pathway (n=17) from April 2018 to August 2019 aged between 14–23 years (mean 18.2±1.9) at the start of the study period. Two supervised learning techniques (a decision tree and random forest with traditional and conditional algorithms) and generalised linear mixed effect models explored associations between risk factors and injury. The supervised learning models did not predict injury (decision tree and random forest area under the curve [AUC] of 0.66 and 0.72 for conditional algorithms) but did identify important risk factors. The best-fitting generalised linear mixed effect model for predicting injury (Akaike Information Criteria [AIC]=843.94, conditional r-squared=0.58) contained smoothed differential 7-day load (P<0.001), average broad jump scores (P<0.001) and 20 m speed (P<0.001). Algorithmic models identified novel injury risk factors in this population, which can guide practice and future confirmatory studies can now investigate.

The role of existing language knowledge in bilingual and multilingual toddlers’ repetition of cross-linguistic and language-specific nonwords

Previous studies have shown that bilingual children typically score more poorly on nonword repetition (NWR) tasks than monolingual peers, which has been attributed to bilinguals’ lower proficiency in the language that the NWR task is based on. To enable fairer assessments of bilingual children, Cross-Linguistic NWR tasks (CL-NWR tasks) have been developed that are based on the linguistic properties of many languages. The aim of this study is to investigate whether young children’s performance on a CL-NWR is less dependent on existing knowledge of a specific language than performance on a Language-Specific (Dutch-based) NWR (LS-NWR). Bilingual and multilingual two- and three-year-olds (N = 216) completed a CL-NWR and LS-NWR, as well as a Dutch receptive vocabulary task. Parents reported the number of languages children spoke other than Dutch. Results of linear mixed-effect regressions showed that Dutch vocabulary scores related to performance on the CL-NWR task less strongly than to performance on the LS-NWR task. The number of non-Dutch languages spoken did not differentially relate to performance on the two tasks. These findings indicate that CL-NWR tasks – at least as used here – allow for more language-neutral NWR assessments within linguistically diverse samples, already at toddler age.

ML-LME: A Plant Growth Situation Analysis Model Using the Hierarchical Effect of Fractal Dimension

Rice plays an essential role in agricultural production as the most significant food crop. Automated supervision in the process of crop growth is the future development direction of agriculture, and it is also a problem that needs to be solved urgently. Productive cultivation, production and research of crops are attributed to increased automation of supervision in the growth. In this article, for the first time, we propose the concept of rice fractal dimension heterogeneity and define it as rice varieties with different fractal dimension values having various correlations between their traits. To make a comprehensive prediction of the rice growth, Machine Learning and Linear Mixed Effect (ML-LME) model is proposed to model and analyze this heterogeneity, which is based on the existing automatic measurement system RAP and introduces statistical characteristics of fractal dimensions as novel features. Machine learning algorithms are applied to distinguish the rice growth stages with a high degree of accuracy and to excavate the heterogeneity of rice fractal dimensions with statistical meaning. According to the information of growth stage and fractal dimension heterogeneity, a precise prediction of key rice phenotype traits can be received by ML-LME using a Linear Mixed Effect model. In this process, the value of the fractal dimension is divided into groups and then rices of different levels are respectively fitted to improve the accuracy of the subsequent prediction, that is, the heterogeneity of the fractal dimension. Afterwards, we apply the model to analyze the rice pot image. The research results show that the ML-LME model, which possesses the hierarchical effect of fractal dimension, performs more excellently in predicting the growth situation of plants than the traditional regression model does. Further comparison confirmed that the model we proposed is the first to consider the hierarchy structure of plant fractal dimension, and that consideration obviously strengthens the model on the ability of variation interpretation and prediction precision.