A Comparative Study of Source Separation Techniques for the Detection of Buried Defects in the EC NDE of Aeronautical Multi-Layered Lap-Joints

Source separation techniques applied to the detection of subsurface defects in the eddy current NDT of aeronautical lap-joints

BackgroundA key question when analyzing high throughput data is whether the information provided by the measured biological entities (gene, metabolite expression for example) is related to the experimental conditions, or, rather, to some interfering signals, such as experimental bias or artefacts. Visualization tools are therefore useful to better understand the underlying structure of the data in a 'blind' (unsupervised) way. A well-established technique to do so is Principal Component Analysis (PCA). PCA is particularly powerful if the biological question is related to the highest variance. Independent Component Analysis (ICA) has been proposed as an alternative to PCA as it optimizes an independence condition to give more meaningful components. However, neither PCA nor ICA can overcome both the high dimensionality and noisy characteristics of biological data.ResultsWe propose Independent Principal Component Analysis (IPCA) that combines the advantages of both PCA and ICA. It uses ICA as a denoising process of the loading vectors produced by PCA to better highlight the important biological entities and reveal insightful patterns in the data. The result is a better clustering of the biological samples on graphical representations. In addition, a sparse version is proposed that performs an internal variable selection to identify biologically relevant features (sIPCA).ConclusionsOn simulation studies and real data sets, we showed that IPCA offers a better visualization of the data than ICA and with a smaller number of components than PCA. Furthermore, a preliminary investigation of the list of genes selected with sIPCA demonstrate that the approach is well able to highlight relevant genes in the data with respect to the biological experiment.IPCA and sIPCA are both implemented in the R package mixomics dedicated to the analysis and exploration of high dimensional biological data sets, and on mixomics' web-interface.

Chapter 3 - A unified probabilistic model for independent and principal component analysis

General Background: Dimensionality reduction is a critical technique in image processing, especially for multispectral satellite imagery where data redundancy and computational complexity are prevalent challenges. Specific Background: Principal Component Analysis (PCA) and Independent Component Analysis (ICA) are two widely adopted methods for reducing dimensionality while preserving essential image information. Knowledge Gap: Despite their extensive usage, comparative assessments of their performance in multispectral image reconstruction, particularly in geospatial contexts, remain limited. Aims: This study aims to evaluate and compare the effectiveness of PCA and ICA in processing Landsat multispectral images of the Bighorn Basin by assessing image reconstruction fidelity. Results: The findings reveal that PCA outperforms ICA in reconstruction quality, achieving higher Peak Signal-to-Noise Ratio (PSNR) values (up to 27.78 dB) and lower Root Mean Square Error (RMSE), whereas ICA, though proficient in extracting statistically independent features, demonstrated lower fidelity (PSNR = 17.63 dB). Novelty: The work offers a rigorous, side-by-side quantitative analysis of PCA and ICA applied to real-world satellite data, highlighting variance behavior and reconstruction trade-offs. Implications: These insights inform the selection of dimensionality reduction techniques in remote sensing tasks—PCA for optimal reconstruction and noise elimination, and ICA for feature extraction based on statistical independence.Highlights: PCA provides superior image reconstruction accuracy with higher PSNR and lower RMSE. ICA excels in isolating statistically independent features for advanced analysis. PCA components show faster variance decay, making them efficient for compression. Keywords: Dimensionality Reduction, Satellite Imagery, Principal Component Analysis, Independent Component Analysis, Image Reconstruction

The use of the non-invasively obtained fetal electrocardiogram (ECG) in fetal monitoring is complicated by the low signal-to-noise ratio (SNR) of ECG signals. Even after removal of the predominant interference (i.e. the maternal ECG), the SNR is generally too low for medical diagnostics, and hence additional signal processing is still required. To this end, several methods for exploiting the spatial correlation of multi-channel fetal ECG recordings from the maternal abdomen have been proposed in the literature, of which principal component analysis (PCA) and independent component analysis (ICA) are the most prominent. Both PCA and ICA, however, suffer from the drawback that they are blind source separation (BSS) techniques and as such suboptimum in that they do not consider a priori knowledge on the abdominal electrode configuration and fetal heart activity. In this paper we propose a source separation technique that is based on the physiology of the fetal heart and on the knowledge of the electrode configuration. This technique operates by calculating the spatial fetal vectorcardiogram (VCG) and approximating the VCG for several overlayed heartbeats by an ellipse. By subsequently projecting the VCG onto the long axis of this ellipse, a source signal of the fetal ECG can be obtained. To evaluate the developed technique, its performance is compared to that of both PCA and ICA and to that of augmented versions of these techniques (aPCA and aICA; PCA and ICA applied on preprocessed signals) in generating a fetal ECG source signal with enhanced SNR that can be used to detect fetal QRS complexes. The evaluation shows that the developed source separation technique performs slightly better than aPCA and aICA and outperforms PCA and ICA and has the main advantage that, with respect to aPCA/PCA and aICA/ICA, it performs more robustly. This advantage renders it favorable for employment in automated, real-time fetal monitoring applications.

Near-infrared spectroscopy (NIRS) is a method for noninvasive estimation of cerebral hemodynamic changes. Principal component analysis (PCA) and independent component analysis (ICA) can be used for decomposing a set of signals to underlying components. Our objective is to determine whether PCA or ICA is more efficient in identifying and removing scalp blood flow interference from multichannel NIRS signals. Concentration changes of oxygenated (HbO(2)) and deoxygenated (HbR) hemoglobin are measured on the forehead with multichannel NIRS during hyper- and hypocapnia. PCA and ICA are used separately to identify and remove signal contribution from extracerebral tissue, and the resulting estimates of cerebral responses are compared to the expected cerebral responses. Both methods were able to reduce extracerebral contribution to the signals, but PCA typically performs equal to or better than ICA. The improvement in 3-cm signal quality achieved with both methods is comparable to increasing the source-detector separation from 3 to 5 cm. Especially PCA appears to be well suited for use in NIRS applications where the cerebral activation is diffuse, such as monitoring of global cerebral oxygenation and hemodynamics. Performance differences between PCA and ICA could be attributed primarily to different criteria for identifying the surface effect.

Remote sensing, in recent days, has seen the emergence of hyperspectral sensors as the helping aspect to numerous applications. Hyperspectral remote sensing often contains data with narrow spectral bandwidth (10nm) that enables the feature identification and distinction of spectral similar features. Hyperion L1R data consists of 242 bands and out of which some bands contain noise (a pattern reorganization that hinders information). After removal of these sensor errors like bad bands and dropped line error, atmospheric correction using Fast Line-of-sight Atmospheric Analysis of Hypercubes (FLAASH) is carried out to get true ground reflectance. It was observed in the spectral reflectance curve of the obtained product is still affected with several noise and in order to remove this noise some corrections are required before atmospheric correction. In this study various noise reduction techniques like Minimum Noise Fraction (MNF), Principal Component Analysis (PCA) and Independent Component Analysis (ICA) are implemented to observe their effect on sensor error removed Hyperion data before carrying out atmospheric correction. At first noise reduction techniques (MNF, PCA and ICA) are applied separately to the Hyperion sensor error corrected product and then inverse of (MNF, PCA and ICA) are carried out respectively. It was observed that applying atmospheric correction after applying MNF & inverse MNF was giving better spectral profiles of the features (i.e. with less noise). From the above spectral profiles we can conclude that the data which is atmospherically corrected after applying MNF & inverse MNF is giving better results i.e. the spectral profiles of vegetation, water, urban and coal field is smooth and less noisy in comparison with the spectral profiles obtained by directly applying atmospheric correction and the spectral profiles obtained by applying PCA & inverse PCA and ICA &inverse ICA. After applying PCA & inverse PCA and then atmospheric correction the spectral profiles of the objects are smooth but not as good as those spectral profile obtained by applying MNF & inverse MNF. It was also observed that the nature of reflectance of the feature at the Short Wave Infrared (SWIR) region has abruptly increased after applying ICA & inverse ICA and atmospheric correction. So the atmospherically corrected data obtained after applying ICA & inverse ICA cannot be used for further processing. The atmospherically corrected data obtained after applying MNF & inverse MNF can be used for further hyperspectral data processing like feature identification and classification as it gives the true spectral nature of the features.

PCA and ICA processing methods for removal of artifacts and noise in electrocardiograms: A survey and comparison

A comparison of PCA and ICA in geochemical pattern recognition of soil data: The case of Cyprus

A Comparison of ECG Signal Pre-processing Using FrFT, FrWT and IPCA for Improved Analysis

A comparative study of independent component analysis with principal component analysis in geological objects identification. Part II: A case study of Pinghe District, Fujian, China

Principal component analysis (PCA) is used for ECG data compression, denoising and decorrelation of noisy and useful ECG components or signals. In this study, a comparative analysis of independent component analysis (ICA) and PCA for correction of ECG signals is carried out by removing noise and artifacts from various raw ECG data sets. PCA and ICA scatter plots of various chest and augmented ECG leads and their combinations are plotted to examine the varying orientations of the heart signal. In order to qualitatively illustrate the recovery of the shape of the ECG signals with high fidelity using ICA, corrected source signals and extracted independent components are plotted. In this analysis, it is also investigated if difference between the two kurtosis coefficients is positive than on each of the respective channels and if we get a super-Gaussian signal, or a sub-Gaussian signal. The efficacy of the combined PCA–ICA algorithm is verified on six channels V1, V3, V6, AF, AR and AL of 12-channel ECG data. ICA has been utilized for identifying and for removing noise and artifacts from the ECG signals. ECG signals are further corrected by using statistical measures after ICA processing. PCA scatter plots of various ECG leads give different orientations of the same heart information when considered for different combinations of leads by quadrant analysis. The PCA results have been also obtained for different combinations of ECG leads to find correlations between them and demonstrate that there is significant improvement in signal quality, i.e., signal-to-noise ratio is improved. In this paper, the noise sensitivity, specificity and accuracy of the PCA method is evaluated by examining the effect of noise, base-line wander and their combinations on the characteristics of ECG for classification of true and false peaks.

Independent component analysis (ICA): A statistical approach to the analysis of superimposed rock paintings

Independent principal component analysis for simulation of soil water content and bulk density in a Canadian Watershed

Spatial independent component analysis (ICA) applied to functional magnetic resonance imaging (fMRI) data identifies functionally connected networks by estimating spatially independent patterns from their linearly mixed fMRI signals. Several multi-subject ICA approaches estimating subject-specific time courses (TCs) and spatial maps (SMs) have been developed, however there has not yet been a full comparison of the implications of their use. Here, we provide extensive comparisons of four multi-subject ICA approaches in combination with data reduction methods for simulated and fMRI task data.For multi-subject ICA, the data first undergo reduction at the subject and group levels using principal component analysis (PCA). Comparisons of subject-specific, spatial concatenation, and group data mean subject-level reduction strategies using PCA and probabilistic PCA (PPCA) show that computationally intensive PPCA is equivalent to PCA, and that subject-specific and group data mean subject-level PCA are preferred because of well-estimated TCs and SMs. Second, aggregate independent components are estimated using either noise free ICA or probabilistic ICA (PICA). Third, subject-specific SMs and TCs are estimated using back-reconstruction. We compare several direct group ICA (GICA) back-reconstruction approaches (GICA1-GICA3) and an indirect back-reconstruction approach, spatio-temporal regression (STR, or dual regression). Results show the earlier group ICA (GICA1) approximates STR, however STR has contradictory assumptions and may show mixed-component artifacts in estimated SMs. Our evidence-based recommendation is to use GICA3, introduced here, with subject-specific PCA and noise-free ICA, providing the most robust and accurate estimated SMs and TCs in addition to offering an intuitive interpretation.