Comment on "A depth-based global envelope test for comparing two groups of functions with applications to biomedical data" by S. Lopez-Pintado and K. Qian.

This is a note describing nonparametric methodology of functional tests in the functional general linear models, which is more rich than the methodology presented in the commented paper.

New methods for multiple testing in permutation inference for the general linear model.

Permutation methods are commonly used to test the significance of regressors of interest in general linear models (GLMs) for functional (image) data sets, in particular for neuroimaging applications as they rely on mild assumptions. Permutation inference for GLMs typically consists of three parts: choosing a relevant test statistic, computing pointwise permutation tests, and applying a multiple testing correction. We propose new multiple testing methods as an alternative to the commonly used maximum value of test statistics across the image. The new methods improve power and robustness against inhomogeneity of the test statistic across its domain. The methods rely on sorting the permuted functional test statistics based on pointwise rank measures; still, they can be implemented even for large data. The performance of the methods is demonstrated through a designed simulation experiment and an example of brain imaging data. We developed the R package GET, which can be used for the computation of the proposed procedures.

Point process models for sweat gland activation observed with noise.

The aim of this article is to construct spatial models for the activation of sweat glands for healthy subjects and subjects suffering from peripheral neuropathy by using videos of sweating recorded from the subjects. The sweat patterns are regarded as realizations of spatial point processes and two point process models for the sweat gland activation and two methods for inference are proposed. Several image analysis steps are needed to extract the point patterns from the videos and some incorrectly identified sweat gland locations may be present in the data. To take into account the errors, we either include an error term in the point process model or use an estimation procedure that is robust with respect to the errors.

Development and evaluation of spatial point process models for epidermal nerve fibers.

We propose two spatial point process models for the spatial structure of epidermal nerve fibers (ENFs) across human skin. The models derive from two point processes, Φb and Φe, describing the locations of the base and end points of the fibers. Each point of Φe (the end point process) is connected to a unique point in Φb (the base point process). In the first model, both Φe and Φb are Poisson processes, yielding a null model of uniform coverage of the skin by end points and general baseline results and reference values for moments of key physiologic indicators. The second model provides a mechanistic model to generate end points for each base, and we model the branching structure more directly by defining Φe as a cluster process conditioned on the realization of Φb as its parent points. In both cases, we derive distributional properties for observable quantities of direct interest to neurologists such as the number of fibers per base, and the direction and range of fibers on the skin. We contrast both models by fitting them to data from skin blister biopsy images of ENFs and provide inference regarding physiological properties of ENFs.

Second-order spatial analysis of epidermal nerve fibers.

Breakthroughs in imaging of skin tissue reveal new details on the distribution of nerve fibers in the epidermis. Preliminary neurologic studies indicate qualitative differences in the spatial patterns of nerve fibers based on pathophysiologic conditions in the subjects. Of particular interest is the evolution of spatial patterns observed in the progression of diabetic neuropathy. It appears that the spatial distribution of nerve fibers becomes more 'clustered' as neuropathy advances, suggesting the possibility of diagnostic prediction based on patterns observed in skin biopsies. We consider two approaches to establish statistical inference relating to this observation. First, we view the set of locations where the nerves enter the epidermis from the dermis as a realization of a spatial point process. Secondly, we treat the set of fibers as a realization of a planar fiber process. In both cases, we use estimated second-order properties of the observed data patterns to describe the degree and scale of clustering observed in the microscope images of blister biopsies. We illustrate the methods using confocal microscopy blister images taken from the thigh of one normal (disease-free) individual and two images each taken from the thighs of subjects with mild, moderate, and severe diabetes and report measurable differences in the spatial patterns of nerve entry points/fibers associated with disease status.