Cluster analysis of high-dimensional data: A case study

Normal mixture models are often used to cluster continuous data. However, conventional approaches for fitting these models will have problems in producing nonsingular estimates of the component-covariance matrices when the dimension of the observations is large relative to the number of observations. In this case, methods such as principal components analysis (PCA) and the mixture of factor analyzers model can be adopted to avoid these estimation problems. We examine these approaches applied to the Cabernet wine data set of Ashenfelter (1999), considering the clustering of both the wines and the judges, and comparing our results with another analysis. The mixture of factor analyzers model proves particularly effective in clustering the wines, accurately classifying many of the wines by location.

The use of multivariate methods in the identification of subtypes of Alzheimer's disease: a comparison of principal components and cluster analysis

Two contrasting multivariate statistical methods, viz., principal components analysis (PCA) and cluster analysis were applied to the study of neuropathological variations between cases of Alzheimer's disease (AD). To compare the two methods, 78 cases of AD were analyzed, each characterised by measurements of 47 neuropathological variables. Both methods of analysis revealed significant variations between AD cases. These variations were related primarily to differences in the distribution and abundance of senile plaques (SP) and neurofibrillary tangles (NFT) in the brain. Cluster analysis classified the majority of AD cases into five groups which could represent subtypes of AD. However, PCA suggested that variation between cases was more continuous with no distinct subtypes. Hence, PCA may be a more appropriate method than cluster analysis in the study of neuropathological variations between AD cases.

Core scans and lithologic analysis of Subglacial Lake Whillans sediment cores

The hydrologic system beneath the Antarctic Ice Sheet is thought to influence both the dynamics and distribution of fast flowing ice streams, which discharge most of the ice lost by the ice sheet. Despite considerable interest in understanding this subglacial network and its affect on ice flow, in situ observations from the ice sheet bed are exceedingly rare. Here we describe the first sediment cores recovered from an active subglacial lake. The lake, known as Subglacial Lake Whillans, is part of a broader, dynamic hydrologic network beneath the Whillans Ice Stream in West Antarctica. Even though "floods" pass through the lake, the lake floor shows no evidence of erosion or deposition by flowing water. By inference, these floods must have insufficient energy to erode or transport significant volumes of sediment coarser than silt. Consequently, water flow beneath the region is probably incapable of incising continuous channels into the bed and instead follows preexisting subglacial topography and surface slope. Sediment on the lake floor consists of till deposited during intermittent grounding of the ice stream following flood events. The fabrics within the till are weaker than those thought to develop in thick deforming beds suggesting subglacial sediment fluxes across the ice plain are currently low and unlikely to have a large stabilizing effect on the ice stream's grounding zone.

Multivariate Techniques in Biomechanical Analysis

This work outlines the theoretical advantages of multivariate methods in biomechanical data, validates the proposed methods and outlines new clinical findings relating to knee osteoarthritis that were made possible by this approach. New techniques were based on existing multivariate approaches, Partial Least Squares (PLS) and Non-negative Matrix Factorization (NMF) and validated using existing data sets. The new techniques developed, PCA-PLS-LDA (Principal Component Analysis – Partial Least Squares – Linear Discriminant Analysis), PCA-PLS-MLR (Principal Component Analysis – Partial Least Squares –Multiple Linear Regression) and Waveform Similarity (based on NMF) were developed to address the challenging characteristics of biomechanical data, variability and correlation. As a result, these new structure-seeking technique revealed new clinical findings. The first new clinical finding relates to the relationship between pain, radiographic severity and mechanics. Simultaneous analysis of pain and radiographic severity outcomes, a first in biomechanics, revealed that the knee adduction moment’s relationship to radiographic features is mediated by pain in subjects with moderate osteoarthritis. The second clinical finding was quantifying the importance of neuromuscular patterns in brace effectiveness for patients with knee osteoarthritis. I found that brace effectiveness was more related to the patient’s unbraced neuromuscular patterns than it was to mechanics, and that these neuromuscular patterns were more complicated than simply increased overall muscle activity, as previously thought.

The application of PCA as a movement pattern recognition technique: A proof of principle

Quantitative methods can help us understand how underlying attributes contribute to movement patterns. Applying principal components analysis (PCA) to whole-body motion data may provide an objective data-driven method to identify unique and statistically important movement patterns. Therefore, the primary purpose of this study was to determine if athletes’ movement patterns can be differentiated based on skill level or sport played using PCA. Motion capture data from 542 athletes performing three sport-screening movements (i.e. bird-dog, drop jump, T-balance) were analyzed. A PCA-based pattern recognition technique was used to analyze the data. Prior to analyzing the effects of skill level or sport on movement patterns, methodological considerations related to motion analysis reference coordinate system were assessed. All analyses were addressed as case-studies. For the first case study, referencing motion data to a global (lab-based) coordinate system compared to a local (segment-based) coordinate system affected the ability to interpret important movement features. Furthermore, for the second case study, where the interpretability of PCs was assessed when data were referenced to a stationary versus a moving segment-based coordinate system, PCs were more interpretable when data were referenced to a stationary coordinate system for both the bird-dog and T-balance task. As a result of the findings from case study 1 and 2, only stationary segment-based coordinate systems were used in cases 3 and 4. During the bird-dog task, elite athletes had significantly lower scores compared to recreational athletes for principal component (PC) 1. For the T-balance movement, elite athletes had significantly lower scores compared to recreational athletes for PC 2. In both analyses the lower scores in elite athletes represented a greater range of motion. Finally, case study 4 reported differences in athletes’ movement patterns who competed in different sports, and significant differences in technique were detected during the bird-dog task. Through these case studies, this thesis highlights the feasibility of applying PCA as a movement pattern recognition technique in athletes. Future research can build on this proof-of-principle work to develop robust quantitative methods to help us better understand how underlying attributes (e.g. height, sex, ability, injury history, training type) contribute to performance.

A principal component approach to space-based gravitational wave astronomy

The current approach to data analysis for the Laser Interferometry Space Antenna (LISA) depends on the time delay interferometry observables (TDI) which have to be generated before any weak signal detection can be performed. These are linear combinations of the raw data with appropriate time shifts that lead to the cancellation of the laser frequency noises. This is possible because of the multiple occurrences of the same noises in the different raw data. Originally, these observables were manually generated starting with LISA as a simple stationary array and then adjusted to incorporate the antenna's motions. However, none of the observables survived the flexing of the arms in that they did not lead to cancellation with the same structure. The principal component approach is another way of handling these noises that was presented by Romano and Woan which simplified the data analysis by removing the need to create them before the analysis. This method also depends on the multiple occurrences of the same noises but, instead of using them for cancellation, it takes advantage of the correlations that they produce between the different readings. These correlations can be expressed in a noise (data) covariance matrix which occurs in the Bayesian likelihood function when the noises are assumed be Gaussian. Romano and Woan showed that performing an eigendecomposition of this matrix produced two distinct sets of eigenvalues that can be distinguished by the absence of laser frequency noise from one set. The transformation of the raw data using the corresponding eigenvectors also produced data that was free from the laser frequency noises. This result led to the idea that the principal components may actually be time delay interferometry observables since they produced the same outcome, that is, data that are free from laser frequency noise. The aims here were (i) to investigate the connection between the principal components and these observables, (ii) to prove that the data analysis using them is equivalent to that using the traditional observables and (ii) to determine how this method adapts to real LISA especially the flexing of the antenna. For testing the connection between the principal components and the TDI observables a 10x 10 covariance matrix containing integer values was used in order to obtain an algebraic solution for the eigendecomposition. The matrix was generated using fixed unequal arm lengths and stationary noises with equal variances for each noise type. Results confirm that all four Sagnac observables can be generated from the eigenvectors of the principal components. The observables obtained from this method however, are tied to the length of the data and are not general expressions like the traditional observables, for example, the Sagnac observables for two different time stamps were generated from different sets of eigenvectors. It was also possible to generate the frequency domain optimal AET observables from the principal components obtained from the power spectral density matrix. These results indicate that this method is another way of producing the observables therefore analysis using principal components should give the same results as that using the traditional observables. This was proven by fact that the same relative likelihoods (within 0.3%) were obtained from the Bayesian estimates of the signal amplitude of a simple sinusoidal gravitational wave using the principal components and the optimal AET observables. This method fails if the eigenvalues that are free from laser frequency noises are not generated. These are obtained from the covariance matrix and the properties of LISA that are required for its computation are the phase-locking, arm lengths and noise variances. Preliminary results of the effects of these properties on the principal components indicate that only the absence of phase-locking prevented their production. The flexing of the antenna results in time varying arm lengths which will appear in the covariance matrix and, from our toy model investigations, this did not prevent the occurrence of the principal components. The difficulty with flexing, and also non-stationary noises, is that the Toeplitz structure of the matrix will be destroyed which will affect any computation methods that take advantage of this structure. In terms of separating the two sets of data for the analysis, this was not necessary because the laser frequency noises are very large compared to the photodetector noises which resulted in a significant reduction in the data containing them after the matrix inversion. In the frequency domain the power spectral density matrices were block diagonals which simplified the computation of the eigenvalues by allowing them to be done separately for each block. The results in general showed a lack of principal components in the absence of phase-locking except for the zero bin. The major difference with the power spectral density matrix is that the time varying arm lengths and non-stationarity do not show up because of the summation in the Fourier transform.

Detection Of Explosives On The Surface Of Banknotes By Raman Hyperspectral Imaging And Independent Component Analysis.

The aim of this study was to develop a methodology using Raman hyperspectral imaging and chemometric methods for identification of pre- and post-blast explosive residues on banknote surfaces. The explosives studied were of military, commercial and propellant uses. After the acquisition of the hyperspectral imaging, independent component analysis (ICA) was applied to extract the pure spectra and the distribution of the corresponding image constituents. The performance of the methodology was evaluated by the explained variance and the lack of fit of the models, by comparing the ICA recovered spectra with the reference spectra using correlation coefficients and by the presence of rotational ambiguity in the ICA solutions. The methodology was applied to forensic samples to solve an automated teller machine explosion case. Independent component analysis proved to be a suitable method of resolving curves, achieving equivalent performance with the multivariate curve resolution with alternating least squares (MCR-ALS) method. At low concentrations, MCR-ALS presents some limitations, as it did not provide the correct solution. The detection limit of the methodology presented in this study was 50μgcm(-2).

Long-term social recognition memory in adult male rats: factor analysis of the social and non-social behaviors

A modified version of the intruder-resident paradigm was used to investigate if social recognition memory lasts at least 24 h. One hundred and forty-six adult male Wistar rats were used. Independent groups of rats were exposed to an intruder for 0.083, 0.5, 2, 24, or 168 h and tested 24 h after the first encounter with the familiar or a different conspecific. Factor analysis was employed to identify associations between behaviors and treatments. Resident rats exhibited a 24-h social recognition memory, as indicated by a 3- to 5-fold decrease in social behaviors in the second encounter with the same conspecific compared to those observed for a different conspecific, when the duration of the first encounter was 2 h or longer. It was possible to distinguish between two different categories of social behaviors and their expression depended on the duration of the first encounter. Sniffing the anogenital area (49.9% of the social behaviors), sniffing the body (17.9%), sniffing the head (3%), and following the conspecific (3.1%), exhibited mostly by resident rats, characterized social investigation and revealed long-term social recognition memory. However, dominance (23.8%) and mild aggression (2.3%), exhibited by both resident and intruders, characterized social agonistic behaviors and were not affected by memory. Differently, sniffing the environment (76.8% of the non-social behaviors) and rearing (14.3%), both exhibited mostly by adult intruder rats, characterized non-social behaviors. Together, these results show that social recognition memory in rats may last at least 24 h after a 2-h or longer exposure to the conspecific.

A chemometric study on the analgesic activity of cannabinoid compounds using SDA, KNN and SIMCA methods

The supervised pattern recognition methods K-Nearest Neighbors (KNN), stepwise discriminant analysis (SDA), and soft independent modelling of class analogy (SIMCA) were employed in this work with the aim to investigate the relationship between the molecular structure of 27 cannabinoid compounds and their analgesic activity. Previous analyses using two unsupervised pattern recognition methods (PCA-principal component analysis and HCA-hierarchical cluster analysis) were performed and five descriptors were selected as the most relevants for the analgesic activity of the compounds studied: R (3) (charge density on substituent at position C(3)), Q (1) (charge on atom C(1)), A (surface area), log P (logarithm of the partition coefficient) and MR (molecular refractivity). The supervised pattern recognition methods (SDA, KNN, and SIMCA) were employed in order to construct a reliable model that can be able to predict the analgesic activity of new cannabinoid compounds and to validate our previous study. The results obtained using the SDA, KNN, and SIMCA methods agree perfectly with our previous model. Comparing the SDA, KNN, and SIMCA results with the PCA and HCA ones we could notice that all multivariate statistical methods classified the cannabinoid compounds studied in three groups exactly in the same way: active, moderately active, and inactive.