Individual Independent Component Analysis Research Articles

An integrated approach combining randomized kernel PCA, Gaussian mixture modeling and ICA for fault detection in non-linear processes

Principal component analysis (PCA) and independent component analysis (ICA), as well as their kernel extensions, have been widely applied in the past for industrial fault detection with Gaussian or non-Gaussian process data with linear or non-linear characteristics. Kernel-based techniques lead to computational complexity due to the high dimensionality of the dataset in the feature space. In this work, a randomization approach is used to obtain a low-rank approximation of the high-dimensional kernel matrix. A hybrid machine learning technique is proposed that integrates randomized kernel PCA (RKPCA) with ICA and Gaussian mixture modeling (GMM). The proposed approach, ICA-RKPCA-GMM, addresses the Gaussian and non-Gaussian characteristics of non-linear process data. Another hybrid algorithm combining three basic techniques of ICA, PCA and GMM is also developed (ICA-PCA-GMM). The fault detection performances of the proposed techniques (ICA-RKPCA-GMM and ICA-PCA-GMM) are compared with PCA, ICA, KPCA and combined ICA-PCA techniques by applying the techniques to two benchmark systems. Monitoring performances were evaluated by determining the false alarm rate and fault detection rate for different types of process and sensor faults. The simulation results show that the proposed ICA-RKPCA-GMM approach yields better results than individual ICA, PCA and KPCA techniques, the combined ICA-PCA and the proposed ICA-PCA-GMM technique.

Test-retest reliability of peak location in the sensorimotor network of resting state fMRI for potential rTMS targets.

Most stroke repetitive transcranial magnetic stimulation (rTMS) studies have used hand motor hotspots as rTMS stimulation targets; in addition, recent studies demonstrated that functional magnetic resonance imaging (fMRI) task activation could be used to determine suitable targets due to its ability to reveal individualized precise and stronger functional connectivity with motor-related brain regions. However, rTMS is unlikely to elicit motor evoked potentials in the affected hemisphere, nor would activity be detected when stroke patients with severe hemiplegia perform an fMRI motor task using the affected limbs. The current study proposed that the peak voxel in the resting-state fMRI (RS-fMRI) motor network determined by independent component analysis (ICA) could be a potential stimulation target. Twenty-one healthy young subjects underwent RS-fMRI at three visits (V1 and V2 on a GE MR750 scanner and V3 on a Siemens Prisma) under eyes-open (EO) and eyes-closed (EC) conditions. Single-subject ICA with different total number of components (20, 30, and 40) were evaluated, and then the locations of peak voxels on the left and right sides of the sensorimotor network (SMN) were identified. While most ICA RS-fMRI studies have been carried out on the group level, that is, Group-ICA, the current study performed individual ICA because only the individual analysis could guide the individual target of rTMS. The intra- (test-retest) and inter-scanner reliabilities of the peak location were calculated. The use of 40 components resulted in the highest test-retest reliability of the peak location in both the left and right SMN compared with that determined when 20 and 30 components were used for both EC and EO conditions. ICA with 40 components might be another way to define a potential target in the SMN for poststroke rTMS treatment.

Fusion analysis of resting-state networks and its application to Alzheimer's disease

Functional networks are extracted from resting-state functional magnetic resonance imaging data to explore the biomarkers for distinguishing brain disorders in disease diagnosis. Previous works have primarily focused on using a single Resting-State Network (RSN) with various techniques. Here, we apply fusion analysis of RSNs to capturing biomarkers that can combine the complementary information among the RSNs. Experiments are carried out on three groups of subjects, i.e., Cognition Normal (CN), Early Mild Cognitive Impairment (EMCI), and Alzheimer's Disease (AD) groups, which correspond to the three progressing stages of AD; each group contains 18 subjects. First, we apply group Independent Component Analysis (ICA) to extracting the Default Mode Network (DMN) and Dorsal Attention Network (DAN) for each subject group. Then, by obtaining the common DMN and DAN as templates for each group, we employ the individual ICA to extract the DMN and DAN for each subject. Finally, we fuse the DMNs and DANs to explore the biomarkers. The results show that (1) the templates generated by group ICA can extract the RSN for each subject by individual ICA effectively; (2) the RSNs combined with the fusion analysis can obtain more informative biomarkers than without fusion analysis; (3) the most different regions of DMN and DAN are found between CN and EMCI and between EMCI and AD, which show differences. For the DMN, the difference in the medial prefrontal cortex between the EMCI and AD is smaller than that between CN and EMCI, whereas that in the posterior cingulate between EMCI and AD is larger. As for the DAN, the difference in the intraparietal sulcus is smaller than that between CN and EMCI; (4) extracting DMN and DAN for each subject via the back reconstruction of group ICA is invalid.

Joint, Partially-Joint, and Individual Independent Component Analysis in Multi-Subject fMRI Data.

Joint analysis of multi-subject brain imaging datasets has wide applications in biomedical engineering. In these datasets, some sources belong to all subjects (joint), a subset of subjects (partially-joint), or a single subject (individual). In this paper, this source model is referred to as joint/partially-joint/individual multiple datasets unidimensional (JpJI-MDU), and accordingly, a source extraction method is developed. We present a deflation-based algorithm utilizing higher order cumulants to analyze the JpJI-MDU source model. The algorithm maximizes a cost function which leads to an eigenvalue problem solved with thin-SVD (singular value decomposition) factorization. Furthermore, we introduce the JpJI-feature which indicates the spatial shape of each source and the amount of its jointness with other subjects. We use this feature to determine the type of sources. We evaluate our algorithm by analyzing simulated data and two real functional magnetic resonance imaging (fMRI) datasets. In our simulation study, we will show that the proposed algorithm determines the type of sources with the accuracy of 95% and 100% for 2-class and 3-class clustering scenarios, respectively. Furthermore, our algorithm extracts meaningful joint and partially-joint sources from the two real datasets, which are consistent with the existing neuroscience studies. Our results in analyzing the real datasets reveal that both datasets follow the JpJI-MDU source model. This source model improves the accuracy of source extraction methods developed for multi-subject datasets. The proposed joint blind source separation algorithm is robust and avoids parameters which are difficult to fine-tune.

Technical note: can resting state functional MRI assist in routine clinical diagnosis?

Despite some differences in clinical presentation, it is often difficult to differentiate between dementia with Lewy bodies (DLB), clinical Alzheimer’s dementia (AD) and Parkinson’s disease dementia. However, differentiation can be crucial, especially as patients with DLB characteristically have a hypersensitivity to most antiemetic and neuroleptic drugs as they affect the cholinergic and dopaminergic system, potentially leading to life-threatening catatonia, loss of cognitive function and muscle rigidity. The aim of this study is to evaluate if resting state (RS) functional MRI (fMRI) can be used in routine practice on a 1.5 T scanner to differentiate between AD and DLB on an individual basis. We age- and gender-matched a known DLB patient with an AD patient and a human control (HC). Individual independent component analysis was carried out. Region of interest seeds were chosen from the midcingulate and insula regions. Functional connectivity from insula to midcingulate and within the midcingulate network (part of the Salience network) was lower in DLB than AD or HC. RS-fMRI on a 1.5 T scanner, in a routine clinical setting, detected abnormal functional connectivity patterns and allowed differentiation of DLB and AD in a routine clinical setting. This is the first evaluation of RS-fMRI in a routine clinical setting. It shows that incorporating RS-fMRI into the clinical scanning protocol can assist in early diagnosis and likely assist in monitoring the natural history of the disease or disease modifying treatments.

Tumor Tissue Detection using Blood-Oxygen-Level-Dependent Functional MRI based on Independent Component Analysis

Accurate delineation of gliomas from the surrounding normal brain areas helps maximize tumor resection and improves outcome. Blood-oxygen-level-dependent (BOLD) functional MRI (fMRI) has been routinely adopted for presurgical mapping of the surrounding functional areas. For completely utilizing such imaging data, here we show the feasibility of using presurgical fMRI for tumor delineation. In particular, we introduce a novel method dedicated to tumor detection based on independent component analysis (ICA) of resting-state fMRI (rs-fMRI) with automatic tumor component identification. Multi-center rs-fMRI data of 32 glioma patients from three centers, plus the additional proof-of-concept data of 28 patients from the fourth center with non-brain musculoskeletal tumors, are fed into individual ICA with different total number of components (TNCs). The best-fitted tumor-related components derived from the optimized TNCs setting are automatically determined based on a new template-matching algorithm. The success rates are 100%, 100% and 93.75% for glioma tissue detection for the three centers, respectively, and 85.19% for musculoskeletal tumor detection. We propose that the high success rate could come from the previously overlooked ability of BOLD rs-fMRI in characterizing the abnormal vascularization, vasomotion and perfusion caused by tumors. Our findings suggest an additional usage of the rs-fMRI for comprehensive presurgical assessment.

ICA-based artifact removal diminishes scan site differences in multi-center resting-state fMRI.

Resting-state fMRI (R-fMRI) has shown considerable promise in providing potential biomarkers for diagnosis, prognosis and drug response across a range of diseases. Incorporating R-fMRI into multi-center studies is becoming increasingly popular, imposing technical challenges on data acquisition and analysis, as fMRI data is particularly sensitive to structured noise resulting from hardware, software, and environmental differences. Here, we investigated whether a novel clean up tool for structured noise was capable of reducing center-related R-fMRI differences between healthy subjects. We analyzed three Tesla R-fMRI data from 72 subjects, half of whom were scanned with eyes closed in a Philips Achieva system in The Netherlands, and half of whom were scanned with eyes open in a Siemens Trio system in the UK. After pre-statistical processing and individual Independent Component Analysis (ICA), FMRIB's ICA-based X-noiseifier (FIX) was used to remove noise components from the data. GICA and dual regression were run and non-parametric statistics were used to compare spatial maps between groups before and after applying FIX. Large significant differences were found in all resting-state networks between study sites before using FIX, most of which were reduced to non-significant after applying FIX. The between-center difference in the medial/primary visual network, presumably reflecting a between-center difference in protocol, remained statistically significant. FIX helps facilitate multi-center R-fMRI research by diminishing structured noise from R-fMRI data. In doing so, it improves combination of existing data from different centers in new settings and comparison of rare diseases and risk genes for which adequate sample size remains a challenge.

Brain functional networks extraction based on fMRI artifact removal: Single subject and group approaches.

Independent component analysis (ICA) has been widely applied to identify brain functional networks from multiple-subject fMRI. However, the best approach to handle artifacts is not yet clear. In this work, we study and compare two ICA approaches for artifact removal using simulations and real fMRI data. The first approach, recommended by the human connectome project, performs ICA on individual data to remove artifacts, and then applies group ICA on the cleaned data from all subjects. We refer to this approach as Individual ICA artifact Removal Plus Group ICA (TRPG). A second approach, Group Information Guided ICA (GIG-ICA), performs ICA on group data, and then removes the artifact group independent components (ICs), followed by individual subject ICA using the remaining group ICs as spatial references. Experiments demonstrate that GIG-ICA is more accurate in estimation of sources and time courses, more robust to data quality and quantity, and more reliable for identifying networks than IRPG.

Age-related differences in functional nodes of the brain cortex - a high model order group ICA study

Functional MRI measured with blood oxygen dependent (BOLD) contrast in the absence of intermittent tasks reflects spontaneous activity of so-called resting state networks (RSN) of the brain. Group level independent component analysis (ICA) of BOLD data can separate the human brain cortex into 42 independent RSNs. In this study we evaluated age-related effects from primary motor and sensory, and, higher level control RSNs. One hundred sixty-eight healthy subjects were scanned and divided into three groups: 55 adolescents (ADO, 13.2 ± 2.4 years), 59 young adults (YA, 22.2 ± 0.6 years), and 54 older adults (OA, 42.7 ± 0.5 years), all with normal IQ. High model order group probabilistic ICA components (70) were calculated and dual-regression analysis was used to compare 21 RSN's spatial differences between groups. The power spectra were derived from individual ICA mixing matrix time series of the group analyses for frequency domain analysis. We show that primary sensory and motor networks tend to alter more in younger age groups, whereas associative and higher level cognitive networks consolidate and re-arrange until older adulthood. The change has a common trend: both spatial extent and the low frequency power of the RSN's reduce with increasing age. We interpret these result as a sign of normal pruning via focusing of activity to less distributed local hubs.

Speech Signal Enhancement Based on MAP Algorithm in the ICA Space

This paper presents a novel maximum a posteriori (MAP) denoising algorithm based on the independent component analysis (ICA). We demonstrate that the employment of individual ICA transformations for signal and noise can provide the best estimate within the linear framework. The signal enhancement problem is categorized based on the distribution of signal and noise being Gaussian or non-Gaussian and the estimation rule is derived for each of the categories. Our theoretical analysis shows that under the assumption of a Gaussian noise the proposed algorithm leads to some well-known enhancement techniques, i.e., Wiener filter and sparse code shrinkage. The analysis of the denoising capability shows that the proposed algorithm is most efficient for non-Gaussian signals corrupted by a non-Gaussian noise. We employed the generalized Gaussian model (GGM) to model the distributions of speech and noise. Experimental evaluation is performed in terms of signal-to-noise ratio (SNR) and spectral distortion measure. Experimental results show that the proposed algorithms achieve significant improvement on the enhancement performance in both Gaussian and non-Gaussian noise.