Principal component component analysis of persistent homology rank functions with case studies of spatial point patterns, sphere packing and colloids

Katharine Felicity Turner
Elsevier, 2016
Journal paper

Abstract

Persistent homology, while ostensibly measuring changes in topology, captures multiscale geometrical information. It is a natural tool for the analysis of point patterns. In this paper we explore the statistical power of the persistent homology rank functions. For a point pattern X we construct a filtration of spaces by taking the union of balls of radius a centred on points in X, X-a = U-x is an element of x B(x, a). The rank function beta(k)(X) : {(a, b) is an element of R-2 : a R is then defined by beta(k)(X)(a, b) = rank (l* : H-k(X-a) -> H-k(X-b)) where l* is the induced map on homology from the inclusion map on spaces. We consider the rank functions as lying in a Hilbert space and show that under reasonable conditions the rank functions from multiple simulations or experiments will lie in an affine subspace. This enables us to perform functional principal component analysis which we apply to experimental data from colloids at different effective temperatures and to sphere packings with different volume fractions. We also investigate the potential of rank functions in providing a test of complete spatial randomness of 2D point patterns using the distances to an empirically computed mean rank function of binomial point patterns in the unit square. (C) 2016 Elsevier B.V. All rights reserved.

Official source

https://infoscience.epfl.ch/record/225113?ln=en

About this result

This page is automatically generated and may contain information that is not correct, complete, up-to-date, or relevant to your search query. The same applies to every other page on this website. Please make sure to verify the information with EPFL's official sources.

Related concepts

Related publications

Related publications (6)

EEG source imaging of brain states using spatiotemporal regression

Christoph Michel, Dimitri Nestor Alice Van De Ville

Relating measures of electroencephalography (EEG) back to the underlying sources is a long-standing inverse problem. Here we propose a new method to estimate the EEG sources of identified electrophysiological states that represent spontaneous activity, or are evoked by a stimulus, or caused by disease or disorder. Our method has the unique advantage of seamlessly integrating a statistical significance of the source estimate while efficiently eliminating artifacts (e.g., due to eye blinks, eye movements, bad electrodes). After determining the electrophysiological states in terms of stable topographies using established methods (e.g.: ICA, PCA, k-means, epoch average), we propose to estimate these states' time courses through spatial regression of a General Linear Model (GLM). These time courses are then used to find EEG sources that have a similar time-course (using temporal regression of a second GLM). We validate our method using both simulated and experimental data. Simulated data allows us to assess the difference between source maps obtained by the proposed method and those obtained by applying conventional source imaging of the state topographies. Moreover, we use data from 7 epileptic patients (9 distinct epileptic foci localized by intracranial EEG) and 2 healthy subjects performing an eyes-open/eyes-closed task to elicit activity in the alpha frequency range. Our results indicate that the proposed EEG source imaging method accurately localizes the sources for each of the electrical brain states. Furthermore, our method is particularly suited for estimating the sources of EEG resting states or otherwise weak spontaneous activity states, a problem not adequately solved before. (C) 2014 Elsevier Inc. All rights reserved.

Elsevier2014

An experimental-numerical investigation of hydrothermal response in adhesively bonded composite structures

John Botsis, Luis Pablo Canal Casado, Hans Georg Limberger, Véronique Michaud, Roohollah Sarfaraz Khabbaz, Georgios Violakis

Water absorption and thermal response of adhesive composite joints were investigated by measurements and numerical simulations. Water diffusivity, saturation, swelling, and thermal expansion of the constituent materials and the joint were obtained from gravimetric experiments and strain measurements using embedded fiber Bragg grating (FBG) sensors. The mechanical response of these materials at different temperatures and water content was characterized by dynamic mechanical analysis. Thermal loading and water absorption in joint specimens were detected by monitoring the FBG wavelength shift caused by thermal expansion or water swelling. The measured parameters were used in finite element models to simulate the response of the embedded sensor. The good correlation of experimental data and simulations confirmed that the change in FBG wavelength could be accurately related to the thermal load or water absorption process. The suitability of the embedded FBG sensors for monitoring of water uptake in adhesive composite joints was demonstrated. (C) 2015 Elsevier Ltd. All rights reserved.

Elsevier2015

Covariance operators play a fundamental role in functional data analysis, providing the canonical means to analyse functional variation via the celebrated Karhunen-Loève expansion. These operators may themselves be subject to variation, for instance in contexts where multiple functional populations are to be compared. Statistical techniques to analyse such variation are intimately linked with the choice of metric on the space of such operators, as well as with their intrinsic infinite-dimensionality. We will show that we can identify the space of infinite-dimensional covariance operators equipped with the Procrustes size-and-shape metric from shape theory, with that of centred Gaussian processes, equipped with the Wasserstein metric of optimal transportation. We then describe key geometrical and topological aspects of the space of covariance operators endowed with the Procrustes metric. Through the notion of multicoupling of Gaussian measures, we establish existence, uniqueness and stability for the Fréchet mean of covariance operators with respect to the Procrustes metric. Furthermore, we will provide generative models that are canonical for such metric. We then turn to the problem of comparing several samples of stochastic processes with respect to their second-order structure, and we subsequently describe the main modes of variation in this second order structure. These two tasks are carried out via an Analysis of Variance (ANOVA) and a Principal Component Analysis (PCA) of covariance operators respectively. In order to perform ANOVA, we introduce a novel approach based on optimal (multi)transport and identify each covariance with an optimal transport map. These maps are then contrasted with the identity with respect to a norm-induced distance. The resulting test statistic, calibrated by permutation, outperforms the state-of-the-art in the functional case. If the null hypothesis postulating equality of the operators is rejected, thanks to a geometric interpretation of the transport maps we can construct a PCA on the tangent space with the aim of understanding the sample variability. Finally, we provide a further example of use of the optimal transport framework, by applying it to the problem of clustering of operators. Two different clustering algorithms are presented, one of which is innovative. The transportation ANOVA, PCA and clustering are validated both on simulated scenarios and real dataset.

EPFL2019