Using models to predict the future: what to do when the data run out?

Although clinical trials often provide "best evidence" comparing the effectiveness of alternative management strategies, such evidence can be limited in duration or in the results reported, causing clinicians and policy analysts to wonder "what if?" Models of the clinical prognosis-often spanning patients' lifetimes (the "long haul")-are perhaps weaker evidence, but can help answer questions about the management of individual patients and place that best evidence into the context of clinical reality.     

Can host reaction animal models be used to predict and modulate skin regeneration?

The study of host reactions in the biomedical and tissue engineering (TE) fields is a key issue but somehow set aside where TE constructs are concerned. Every day new biomaterials and TE constructs are being developed and presented to the scientific community. The combination of cells and biomolecules with scaffolding materials, as TE constructs, make the isolation and the understanding of the effect of each one those elements over the overall host reaction difficult. Eventually, all variables influence the host reaction and the performance of the constructs. For this reason, current assessment of the in vivo performance of TE constructs follows individual approaches, using specific animal models to independently provide insights regarding the contribution of the biomaterials/scaffolds towards the host reaction, and of all the constructs regarding their functionality. Skin wound healing progress into tissue regeneration or repair is highly dependent on the specificities of the inflammatory stage, as demonstrated by comparison between fetal and adult mechanisms. Thus, it would be expected that insights acquired from host tissue reaction evaluation to biomaterials/scaffolds would be explored to predict healing progression and improve the functionality of skin TE constructs. The rational of this review is to make a comprehensive analysis of to what extent the knowledge obtained from the evaluation of in vivo host reactions to implantable biomaterials/scaffolds has been used in the design of skin TE strategies, by promoting tissue regeneration rather than repair. Copyright © 2016 John Wiley & Sons, Ltd.     

ℓ(1)-penalized linear mixed-effects models for high dimensional data with application to BCI

Recently, a novel statistical model has been proposed to estimate population effects and individual variability between subgroups simultaneously, by extending Lasso methods. We will for the first time apply this so-called ?(1)-penalized linear regression mixed-effects model for a large scale real world problem: we study a large set of brain computer interface data and through the novel estimator are able to obtain a subject-independent classifier that compares favorably with prior zero-training algorithms. This unifying model inherently compensates shifts in the input space attributed to the individuality of a subject. In particular we are now for the first time able to differentiate within-subject and between-subject variability. Thus a deeper understanding both of the underlying statistical and physiological structures of the data is gained.