scholarly journals On identifying kinematic and muscle synergies: a comparison of matrix factorization methods using experimental data from the healthy population

2017 ◽  
Vol 117 (1) ◽  
pp. 290-302 ◽  
Author(s):  
Navid Lambert-Shirzad ◽  
H. F. Machiel Van der Loos
Keyword(s):  
Principal Component Analysis ◽  
Matrix Factorization ◽  
Muscle Activation ◽  
Nonnegative Matrix Factorization ◽  
Nonnegative Matrix ◽  
Principal Component ◽  
Component Analysis ◽  
Data Sets ◽  
Motor Synergies ◽  
Factorization Methods

Human motor behavior is highly goal directed, requiring the central nervous system to coordinate different aspects of motion generation to achieve the motion goals. The concept of motor synergies provides an approach to quantify the covariation of joint motions and of muscle activations, i.e., elemental variables, during a task. To analyze goal-directed movements, factorization methods can be used to reduce the high dimensionality of these variables while accounting for much of the variance in large data sets. Three factorization methods considered in this paper are principal component analysis (PCA), nonnegative matrix factorization (NNMF), and independent component analysis (ICA). Bilateral human reaching data sets are used to compare the methods, and advantages of each are presented and discussed. PCA and NNMF had a comparable performance on both EMG and joint motion data and both outperformed ICA. However, NNMF's nonnegativity condition for activation of basis vectors is a useful attribute in identifying physiologically meaningful synergies, making it a more appealing method for future studies. A simulated data set is introduced to clarify the approaches and interpretation of the synergy structures returned by the three factorization methods.NEW & NOTEWORTHY Literature on comparing factorization methods in identifying motor synergies using numerically generated, simulation, and muscle activation data from animal studies already exists. We present an empirical evaluation of the performance of three of these methods on muscle activation and joint angles data from human reaching motion: principal component analysis, nonnegative matrix factorization, and independent component analysis. Using numerical simulation, we also studied the meaning and differences in the synergy structures returned by each method. The results can be used to unify approaches in identifying and interpreting motor synergies.

scholarly journals Matrix Factorization Algorithms for the Identification of Muscle Synergies: Evaluation on Simulated and Experimental Data Sets

2006 ◽  
Vol 95 (4) ◽  
pp. 2199-2212 ◽  
Author(s):  
Matthew C. Tresch ◽  
Vincent C. K. Cheung ◽  
Andrea d'Avella
Keyword(s):  
Matrix Factorization ◽  
Muscle Activation ◽  
Nonnegative Matrix Factorization ◽  
Nonnegative Matrix ◽  
Muscle Synergies ◽  
Data Sets ◽  
Data Set ◽  
Activation Patterns ◽  
Muscle Activation Patterns ◽  
Factorization Algorithms

Several recent studies have used matrix factorization algorithms to assess the hypothesis that behaviors might be produced through the combination of a small number of muscle synergies. Although generally agreeing in their basic conclusions, these studies have used a range of different algorithms, making their interpretation and integration difficult. We therefore compared the performance of these different algorithms on both simulated and experimental data sets. We focused on the ability of these algorithms to identify the set of synergies underlying a data set. All data sets consisted of nonnegative values, reflecting the nonnegative data of muscle activation patterns. We found that the performance of principal component analysis (PCA) was generally lower than that of the other algorithms in identifying muscle synergies. Factor analysis (FA) with varimax rotation was better than PCA, and was generally at the same levels as independent component analysis (ICA) and nonnegative matrix factorization (NMF). ICA performed very well on data sets corrupted by constant variance Gaussian noise, but was impaired on data sets with signal-dependent noise and when synergy activation coefficients were correlated. Nonnegative matrix factorization (NMF) performed similarly to ICA and FA on data sets with signal-dependent noise and was generally robust across data sets. The best algorithms were ICA applied to the subspace defined by PCA (ICAPCA) and a version of probabilistic ICA with nonnegativity constraints (pICA). We also evaluated some commonly used criteria to identify the number of synergies underlying a data set, finding that only likelihood ratios based on factor analysis identified the correct number of synergies for data sets with signal-dependent noise in some cases. We then proposed an ad hoc procedure, finding that it was able to identify the correct number in a larger number of cases. Finally, we applied these methods to an experimentally obtained data set. The best performing algorithms (FA, ICA, NMF, ICAPCA, pICA) identified synergies very similar to one another. Based on these results, we discuss guidelines for using factorization algorithms to analyze muscle activation patterns. More generally, the ability of several algorithms to identify the correct muscle synergies and activation coefficients in simulated data, combined with their consistency when applied to physiological data sets, suggests that the muscle synergies found by a particular algorithm are not an artifact of that algorithm, but reflect basic aspects of the organization of muscle activation patterns underlying behaviors.

scholarly journals Diffuse Optical Imaging Using Decomposition Methods

2012 ◽  
Vol 2012 ◽  
pp. 1-12 ◽  
Author(s):  
Binlin Wu ◽  
M. Alrubaiee ◽  
W. Cai ◽  
M. Xu ◽  
S. K. Gayen
Keyword(s):  
Experimental Data ◽  
Optical Imaging ◽  
Matrix Factorization ◽  
Nonnegative Matrix Factorization ◽  
Nonnegative Matrix ◽  
Principal Component ◽  
Matrix Decomposition ◽  
Decomposition Methods ◽  
Component Analysis ◽  
Diffuse Optical Imaging

Diffuse optical imaging (DOI) for detecting and locating targets in a highly scattering turbid medium is treated as a blind source separation (BSS) problem. Three matrix decomposition methods, independent component analysis (ICA), principal component analysis (PCA), and nonnegative matrix factorization (NMF) were used to study the DOI problem. The efficacy of resulting approaches was evaluated and compared using simulated and experimental data. Samples used in the experiments included Intralipid-10% or Intralipid-20% suspension in water as the medium with absorptive or scattering targets embedded.

scholarly journals Collaborative Filtering Recommendation Using Nonnegative Matrix Factorization in GPU-Accelerated Spark Platform

2021 ◽  
Vol 2021 ◽  
pp. 1-15
Author(s):  
Bing Tang ◽  
Linyao Kang ◽  
Li Zhang ◽  
Feiyan Guo ◽  
Haiwu He
Keyword(s):  
Collaborative Filtering ◽  
Processing Speed ◽  
Matrix Factorization ◽  
Nonnegative Matrix Factorization ◽  
Nonnegative Matrix ◽  
Experimental Results ◽  
Computational Time ◽  
Data Sets ◽  
Heterogeneous Cluster ◽  
The Matrix

Nonnegative matrix factorization (NMF) has been introduced as an efficient way to reduce the complexity of data compression and its capability of extracting highly interpretable parts from data sets, and it has also been applied to various fields, such as recommendations, image analysis, and text clustering. However, as the size of the matrix increases, the processing speed of nonnegative matrix factorization is very slow. To solve this problem, this paper proposes a parallel algorithm based on GPU for NMF in Spark platform, which makes full use of the advantages of in-memory computation mode and GPU acceleration. The new GPU-accelerated NMF on Spark platform is evaluated in a 4-node Spark heterogeneous cluster using Google Compute Engine by configuring each node a NVIDIA K80 CUDA device, and experimental results indicate that it is competitive in terms of computational time against the existing solutions on a variety of matrix orders. Furthermore, a GPU-accelerated NMF-based parallel collaborative filtering (CF) algorithm is also proposed, utilizing the advantages of data dimensionality reduction and feature extraction of NMF, as well as the multicore parallel computing mode of CUDA. Using real MovieLens data sets, experimental results have shown that the parallelization of NMF-based collaborative filtering on Spark platform effectively outperforms traditional user-based and item-based CF with a higher processing speed and higher recommendation accuracy.

Principal Component Analysis of Hydrological Data

2010 ◽  
pp. 364-388
Author(s):  
Petr Praus
Keyword(s):  
Water Quality ◽  
Principal Component Analysis ◽  
Drinking Water ◽  
Ground Water ◽  
Principal Component ◽  
Component Analysis ◽  
Data Sets ◽  
Hydrological Data ◽  
First Case

In this chapter the principals and applications of principal component analysis (PCA) applied on hydrological data are presented. Four case studies showed the possibility of PCA to obtain information about wastewater treatment process, drinking water quality in a city network and to find similarities in the data sets of ground water quality results and water-related images. In the first case study, the composition of raw and cleaned wastewater was characterised and its temporal changes were displayed. In the second case study, drinking water samples were divided into clusters in consistency with their sampling localities. In the case study III, the similar samples of ground water were recognised by the calculation of cosine similarity, the Euclidean and Manhattan distances. In the case study IV, 32 water-related images were transformed into a large image matrix whose dimensionality was reduced by PCA. The images were clustered using the PCA scatter plots.

EVALUATION AND MODELLING OF GROUND WATER QUALITY DATA OF ALLAHABAD CITY BY ENVIRONMETRIC METHODS

2016 ◽  
Vol 2 (4) ◽  
pp. 211
Author(s):  
Girdhari Lal Chaurasia ◽  
Mahesh Kumar Gupta ◽  
Praveen Kumar Tandon
Keyword(s):  
Water Quality ◽  
Principal Component Analysis ◽  
Cluster Analysis ◽  
Factor Analysis ◽  
Principal Component ◽  
Component Analysis ◽  
Sampling Location ◽  
Data Sets ◽  
Multivariate Statistical ◽  
Positive Loading

Water is an essential resource for all the organisms, plants and animals including the human beings. It is the backbone for agricultural and industrial sectors and all the small business units. Increase in human population and economic activities have tremendously increased the demand for large-scale suppliers of fresh water for various competing end users.The quality evaluation of water is represented in terms of physical, chemical and Biological parameters. A particular problem in the case of water quality monitoring is the complexity associated with analyzing the large number of measured variables. The data sets contain rich information about the behavior of the water resources. Multivariate statistical approaches allow deriving hidden information from the data sets about the possible influences of the environment on water quality. Classification, modeling and interpretation of monitored data are the most important steps in the assessment of water quality. The application of different multivariate statistical techniques, such as cluster analysis (CA), principal component analysis (PCA) and factor analysis (FA) help to identify important components or factors accounting for most of the variances of a system. In the present study water samples were analyzed for various physicochemical analyses by different methods following the standards of APHA, BIS and WHO and were subjected to further statistical analysis viz. the cluster analysis to understand the similarity and differences among the various sampling stations.  Three clusters were found. Cluster 1 was marked with 3 sampling locations 1, 3 & 5; Cluster-2 was marked with sampling location-2 and cluster-3 was marked with sampling location-4. Principal component analysis/factor analysis is a pattern reorganization technique which is used to assess the correlation between the observations in terms of different factors which are not observable. Observations correlated either positively or negatively, are likely to be affected by the same factors while the observations which are not correlated are influenced by different factors. In our study three factors explained 99.827% of variances. F1 marked  51.619% of total variances, high positive strong loading with TSS, TS, Temp, TDS, phosphate and moderate with electrical conductivity with loading values of 0.986, 0.970, 0.792, 0.744, 0.695,  0.701, respectively. Factor 2 marked 27.236% of the total variance with moderate positive loading with total alkalinity & temp. with loading values 0.723 & 0.606 respectively. It also explained the moderate negative loading with conductivity, TDS, and chloride with loading values -0.698, -0.690, -0.582. Factor F 3 marked 20.972 % of the variances with positive loading with PH, chloride, and phosphate with strong loading of pH 0.872 and moderate positive loading with chloride and phosphate with loading values 0.721, and 0.569 respectively. 

Identification of Excitation Source Number Using Principal Component Analysis

2011 ◽  
Vol 199-200 ◽  
pp. 850-857
Author(s):  
Jian Chao Dong ◽  
Tie Jun Yang ◽  
Xin Hui Li ◽  
Zhi Jun Shuai ◽  
You Hong Xiao
Keyword(s):  
Principal Component Analysis ◽  
Signal To Noise Ratio ◽  
Multiple Input Multiple Output ◽  
Principal Component ◽  
Relevant Information ◽  
Component Analysis ◽  
Data Sets ◽  
Blind Signal ◽  
Input Multiple Output ◽  
Source Number

Principal component analysis (PCA), serving as one of the basic blind signal processing techniques, is extensively employed in all forms of analysis for extracting relevant information from confusing data sets. The principle of PCA is explained in this paper firstly, then the simulation and experiment are carried out to a simply supported beam rig, and PCA is used in frequency domain to identify sources number of several cases. Meanwhile principal components (PCs) contribution coefficient and signal to noise ratio between neighboring PCs (neighboring SNR) are introduced to cutoff minor components quantificationally. The results show that when observation number is equal to or larger than source number and additive noise is feebleness, accurate prediction of the number of uncorrelated excitation sources in a multiple input multiple output system could be obtained by principal component analysis.

Sign in / Sign up

Export Citation Format

Share Document