Interpreting principal components in biomechanics: Representative extremes and single component reconstruction

Brandon, Scott C.E.; Graham, Ryan B.; Almosnino, Sivan; Sadler, Erin M.; Stevenson, Joan M.; Deluzio, Kevin J.

doi:10.1016/j.jelekin.2013.09.010

Cited by 99 publications

(60 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We computed PC scores for each trial and PC (Matlab R2012a; Brandon et al, 2013;Daffertshofer et al, 2004). For easier interpretation, we descriptively grouped PC loading vectors into sources of variation capturing landing sub phase characteristics (Fig.…”

Section: Discussionmentioning

confidence: 99%

“…Principal component analysis (PCA) has gained increasingly widespread application in biomechanical investigations, reducing relevant information from multi-dimensional signals into independent sources of variation (Brandon et al, 2013;Daffertshofer, Lamoth, Meijer, & Beek, 2004;Federolf, Boyer, & Andriacchi, 2013;Kipp & Palmieri-Smith, 2012;Richter, O'Connor, Marshall, & Moran, 2014). Given the outlined descriptions of movement coordination and synergies, PCA can provide interpretations in line with concepts from motor control (Latash, 2010;Li, 2006;Lohse et al, 2013;Scholz & Schoner, 1999), providing a multivariate time series measure of movement variability.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Neuromechanical synergies in single-leg landing reveal changes in movement control

Nordin

Dufek

2016

Human Movement Science

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Neuromechanical synergies in single-leg landing reveal changes in movement control

Nordin

Dufek

2016

Human Movement Science

View full text Add to dashboard Cite

“…Since our study was motivated by a clinical interest in defining gait‐related features characteristic of hip OA patients, we interpreted the 10 retained PCs by going back to the biomechanical gait variables. The clinical interpretation of each PC is guided by its corresponding source of variability that can be difference, magnitude or time‐shift . Difference variability identifies a variation in peak values of the waveform (i.e., amplitude) that can eventually lead to a different pattern; magnitude variability corresponds to a vertical shift of the complete waveform and is representative of an offset, whereas time‐shift variability illustrates a shift in the timing of an event within the waveform.…”

Section: Methodsmentioning

confidence: 99%

“…Scaling the loading vector using the 5th and 95th percentile of the PC scores and adding it to the mean waveform of controls combined with patients construct the lower and upper bands, respectively. The graphical representation of the perturbation around the mean waveform provides an easy‐to‐read method for single PC interpretation, also called single component reconstruction . A detailed interpretation of the three first PCs into gait‐related features is presented in the results section.…”

Section: Methodsmentioning

confidence: 99%

Biomechanical gait features associated with hip osteoarthritis: Towards a better definition of clinical hallmarks

Meyer

Corten

Fieuws

et al. 2015

Journal Orthopaedic Research

View full text Add to dashboard Cite

Critical appraisal of the literature highlights that the discriminative power of gait-related features in patients with hip osteoarthritis (OA) has not been fully explored. We aimed to reduce the number of gait-related features and define the most discriminative ones comparing the three-dimensional gait analysis of 20 patients with hip osteoarthritis (OA) with those of 17 healthy peers. First, principal component analysis was used to reduce the high-dimensional gait data into a reduced set of interpretable variables for further analysis, including tests for group differences. These differences were indicative for the selection of the top 10 variables to be included into linear discriminant analysis models (LDA). Our findings demonstrated the successful data reduction of hip osteoarthritic-related gait features with a high discriminatory power. The combination of the top variables into LDA models clearly separated groups, with a maximum misclassification error rate of 19%, estimated by cross-validation. Decreased hip/knee extension, hip flexion and internal rotation moment were gait features with the highest discriminatory power. This study listed the most clinically relevant gait features characteristics of hip OA. Moreover, it will help clinicians and physiotherapists understand the movement pathomechanics related to hip OA useful in the management and design of rehabilitation intervention.

show abstract

“…Other more complex analyses of EMG time-series include principal component analysis [Brandon et al, 2013], cross-correlation [Wren et al, 2006] or wavelet transform for time/frequency analysis [von Tscharner, 2000]. These methods may consider inter-muscle and/or time dependence yet these methods do not directly provide the necessary objective statistics with which a non-directed null hypothesis could be rejected.…”

Section: Time Dependencementioning

confidence: 99%

Statistical Parametric Mapping (SPM) for alpha-based statistical analyses of multi-muscle EMG time-series

Robinson

Vanrenterghem

Pataky

2015

Journal of Electromyography and Kinesiology

102

View full text Add to dashboard Cite

Multi-muscle EMG time-series are highly correlated and time dependent yet traditional statistical analysis of scalars from an EMG time-series fails to account for such dependencies. This paper promotes the use of SPM vector-field analysis for the generalised analysis of EMG time-series. We reanalysed a publicly available dataset of Young versus Adult EMG gait data to contrast scalar and SPM vector-field analysis. Independent scalar analyses of EMG data between 35-45% stance phase showed no statistical differences between the Young and Adult groups. SPM vector-field analysis did however identify statistical differences within this time period. As scalar analysis failed to consider the multi-muscle and time dependence of the EMG time-series it exhibited Type II error. SPM vector-field analysis on the other hand accounts for both dependencies whilst tightly controlling for Type I and Type II error making it highly applicable to EMG data analysis. Additionally SPM vector-field analysis is generalizable to linear and non-linear parametric and non-parametric statistical models, allowing its use under constraints that are common to electromyography and kinesiology.

show abstract

Interpreting principal components in biomechanics: Representative extremes and single component reconstruction

Cited by 99 publications

References 26 publications

Neuromechanical synergies in single-leg landing reveal changes in movement control

Neuromechanical synergies in single-leg landing reveal changes in movement control

Biomechanical gait features associated with hip osteoarthritis: Towards a better definition of clinical hallmarks

Statistical Parametric Mapping (SPM) for alpha-based statistical analyses of multi-muscle EMG time-series

Contact Info

Product

Resources

About