Canonical Correlation Analysis Technique Research Articles

A novel fusion framework of deep bottleneck residual convolutional neural network for breast cancer classification from mammogram images.

With over 2.1 million new cases of breast cancer diagnosed annually, the incidence and mortality rate of this disease pose severe global health issues for women. Identifying the disease's influence is the only practical way to lessen it immediately. Numerous research works have developed automated methods using different medical imaging to identify BC. Still, the precision of each strategy differs based on the available resources, the issue's nature, and the dataset being used. We proposed a novel deep bottleneck convolutional neural network with a quantum optimization algorithm for breast cancer classification and diagnosis from mammogram images. Two novel deep architectures named three-residual blocks bottleneck and four-residual blocks bottle have been proposed with parallel and single paths. Bayesian Optimization (BO) has been employed to initialize hyperparameter values and train the architectures on the selected dataset. Deep features are extracted from the global average pool layer of both models. After that, a kernel-based canonical correlation analysis and entropy technique is proposed for the extracted deep features fusion. The fused feature set is further refined using an optimization technique named quantum generalized normal distribution optimization. The selected features are finally classified using several neural network classifiers, such as bi-layered and wide-neural networks. The experimental process was conducted on a publicly available mammogram imaging dataset named INbreast, and a maximum accuracy of 96.5% was obtained. Moreover, for the proposed method, the sensitivity rate is 96.45, the precision rate is 96.5, the F1 score value is 96.64, the MCC value is 92.97%, and the Kappa value is 92.97%, respectively. The proposed architectures are further utilized for the diagnosis process of infected regions. In addition, a detailed comparison has been conducted with a few recent techniques showing the proposed framework's higher accuracy and precision rate.

Driving Fatigue Detection Based on Hybrid Electroencephalography and Eye Tracking.

EEG-based unimodal method has demonstrated substantial success in the detection of driving fatigue. Nonetheless, the data from a single modality might be not sufficient to optimize fatigue detection due to incomplete information. To address this limitation and enhance the performance of driving fatigue detection, a novel multimodal architecture combining electroencephalography (EEG) and eye tracking data was proposed in this study. Specifically, EEG and eye tracking data were separately input into encoders, generating two one-dimensional (1D) features. Subsequently, these 1D features were fed into a cross-modal predictive alignment module to improve fusion efficiency and two 1D attention modules to enhance feature representation. Furthermore, the fused features were recognized by a linear classifier. To evaluate the effectiveness of the proposed multimodal method, comprehensive validation tasks were conducted, including intra-session, cross-session, and cross-subject evaluations. In the intra-session task, the proposed architecture achieves an exceptional average accuracy of 99.93%. Moreover, in the cross-session task, our method results in an average accuracy of 88.67%, surpassing the performance of EEG-only approach by 8.52%, eye tracking-only method by 5.92%, multimodal deep canonical correlation analysis (DCCA) technique by 0.42%, and multimodal deep generalized canonical correlation analysis (DGCCA) approach by 0.84%. Similarly, in the cross-subject task, the proposed approach achieves an average accuracy of 78.19%, outperforming EEG-only method by 5.87%, eye tracking-only approach by 4.21%, DCCA method by 0.55%, and DGCCA approach by 0.44%. The experimental results conclusively illustrate the superior effectiveness of the proposed method compared to both single modality approaches and canonical correlation analysis-based multimodal methods. Overall, this study provides a new and effective strategy for driving fatigue detection.

Deep and Shallow Feature Fusion in Feature Score Level for Palmprint Recognition

Contactless palmprint recognition offers friendly customer experience due to its ability to operate without touching the recognition device under rigid constrained conditions. Recent palmprint recognition methods have shown promising accuracy; however, there still exist some issues that need to be further studied such as the limited discrimination of the single feature and how to effectively fuse deep features and shallow features. In this paper, deep features and shallow features are integrated into a unified framework using feature‐level and score‐level fusion methods. Specifically, deep feature is extracted by residual neural network (ResNet), and shallow features are extracted by principal component analysis (PCA), linear discriminant analysis (LDA), and competitive coding (CompCode). In feature‐level fusion stage, ResNet feature and PCA feature are dimensionally reduced and fused by canonical correlation analysis technique to achieve the fused feature for the next stage. In score‐level fusion stage, score information is embedded in the fused feature, LDA feature, and CompCode feature to obtain a more reliable and robust recognition performance. The proposed method achieves competitive performance on Tongji dataset and demonstrates more satisfying generalization capabilities on IITD and CASIA datasets. Comprehensive validation across three palmprint datasets confirms the effectiveness of our proposed deep and shallow feature fusion approach.

Forecast skill of Bangladesh summer monsoon rainfall in C3S and NMME models after calibration

This study assesses the dynamical seasonal predictions initialized in April and May for forecasting Bangladesh summer (June to September: JJAS) monsoon rainfall (BSMR) over the 1993–2016 period. The BSMR from nine models, sourced from the North American Multi-Model Ensemble (NMME) and the Copernicus Climate Change Service (C3S), undergoes a calibration process. This calibration employs the Canonical Correlation Analysis (CCA) technique to rectify biases within each individual model's ensemble mean BSMR data (referred as predictor or X variable). These corrections are made in comparison to observed BSMR (referred as predictand or Y variable), acquired from the Climate Hazards Group InfraRed Precipitation with Station data. Subsequently, the models that undergo calibration are amalgamated to construct a calibrated multi-model ensemble (CMME), which, in turn, facilitates the generation of a forecast for BSMR. The CCA correction brings about a significant improvement in root-mean-square error, underscoring the presence of correctable systematic biases in the raw model forecasts. However, these CCA corrections weakly enhance the skill (anomaly correlation) across the region. The scores that assess discrimination (the two-alternative forced-choice: 2AFC and the area under the relative operating characteristic curve: ROC for above/below normal BSMR) for tercile-based forecasts exceeded 50 % across a substantial portion of the region. This indicates a superior level of discrimination compared to what one would anticipate based on climatology. Using the CMME approach, a probabilistic forecast for the 2022 BSMR was generated and proved quite effective in capturing the observed 2022 BSMR tercile, which includes below-normal and above-normal categories of rainfall in the central-southern and northern regions of Bangladesh respectively. Furthermore, the absence of substantial skill improvements may be attributed to inaccuracies in the teleconnection patterns of the simulated first leading principal component (PC) time series of BSMR with the El Niño Southern Oscillation. In contrast, the second PC time series exhibits a similar connection to observations. These findings emphasize the importance and utility of statistical post-processing in producing reliable seasonal climate outlooks for the region.

Domain Adaptation Problem in Sketch Based Image Retrieval

In this article, we present two algorithms that discover the discriminative structures of sketches, given pairs of sketches and photos in sketch-based image retrieval (SBIR) scenarios. Unlike the existing approaches, we aim at the few-shot and domain adaptation (DA) problems, and set up a module with canonical correlation analysis (CCA) technique in our algorithms to improve retrieval performance. For single source domain settings, our first algorithm can effectively transfer a classifier trained on a known dataset to a new one. For multisource settings, our second algorithm sophisticatedly combines multisource domain data to yield a classifier on the target domain. To the best of our knowledge, these two works are the first research in SBIR field. Experiments on the Sketchy and TU-Berlin sketch benchmark datasets demonstrate the effectiveness of our algorithms and compelling performance. Compared with the state-of-the-art methods, our algorithms do not use text based semantic information, but achieve competitive results. Furthermore, experiments also turn out that feature representation by available trained deep networks has distinct advantages and combination with traditional machine learning methods brings substantial improvements against the state-of-the-art methods on image retrievals. The complete source code of the proposed algorithms will be released on gitHub: [ GitHub- ADAM0912/Domain-Adaptation-Problem-in-Sketch-Based-Image-Retrieval: The code repository for the articleDomain Adaptation Problem in Sketch Based Image Retrieval ] .

A supervised case-based reasoning approach for explainable thyroid nodule diagnosis

As an explainable experience-based artificial intelligence technique, case-based reasoning (CBR) has been widely used to help diagnose many diseases, but the application of CBR in the diagnosis of thyroid nodules (TDNs) is rarely studied. To fill this research gap, this paper proposes a supervised CBR approach to help diagnose TDNs. The proposed approach first investigates the correlation between the feature diagnoses of historical TDN cases and the corresponding overall diagnoses using the canonical correlation analysis technique. Then the learned canonical variables are used to reconstruct TDN cases. Based on the reconstructed historical case base, a classifier is constructed to provide pathological diagnosis predictions for new TDN cases. To explain these predictions with similar historical TDN cases, a convex optimization model is constructed to determine the similarity between historical TDN cases and new TDN cases. Finally, a weighted combination scheme is designed to generate an explainable pathological diagnosis for each new TDN case based on its similar historical TDN cases. The proposed approach not only avoids the burdensome parameter tuning task but also reduces the likelihood of retrieving noisy historical cases as similar cases of new cases with a supervised case retrieval process. Using a real diagnostic dataset collected from the ultrasound department of a local hospital, the effectiveness of the proposed approach in diagnosing TDNs is validated and its advantages are further highlighted by comparison with the traditional CBR approach and six mainstream machine learning models.

New media literacy and news trustworthiness: An application of importance–performance analysis

Previous studies have primarily focused on the detection of fake news using artificial intelligence or identifying it through analysis of politically charged topics and their implications. However, studies have rarely systematically analyzed the characteristics of fake news from the perspective of readers. Little empirical evidence exists demonstrating the positive role of the new media literacy (NML) in assisting adolescents in recognizing fake news. This study examined the relationship among Taiwanese students' level of NML and their performance and perception of the importance of using news trustworthiness (NT) attributes to discern fake news while interacting with media. The cross-sectional study was conducted through a questionnaire survey, and 563 responses were included in the analysis. Importance-performance analysis (IPA) and canonical correlation analysis techniques were used to validate the multidimensional relationships between NML and NT. From the IPA with data-centered quadrants and diagonal-based schemes, 31 of the 32 NT attributes were self-evaluated with lower performance than importance. Canonical correlation analysis showed just a slight correlation between NML and NT. This study seeks to provide insights into the strengths and weaknesses of students’ NML and NT skills, and its findings may help to further equip digital readers with the skillset needed to discern facts from falsehoods.

ISFun: an R package for integrative dimension reduction analysis.

In the analysis of high-dimensional omics data, dimension reduction techniques-including principal component analysis (PCA), partial least squares (PLS) and canonical correlation analysis (CCA)-have been extensively used. When there are multiple datasets generated by independent studies with compatible designs, integrative analysis has been developed and shown to outperform meta-analysis, other multidatasets analysis, and individual-data analysis. To facilitate integrative dimension reduction analysis in daily practice, we develop the R package iSFun, which can comprehensively conduct integrative sparse PCA, PLS and CCA, as well as meta-analysis and stacked analysis. The package can conduct analysis under the homogeneity and heterogeneity models and with the magnitude- and sign-based contrasted penalties. As a 'byproduct', this article is the first to develop integrative analysis built on the CCA technique, further expanding the scope of integrative analysis. The package is available at https://CRAN.R-project.org/package=iSFun. Supplementary materials are available at Bioinformatics online.

Seasonal Rainfall Prediction for Tamil Nadu Using Climate Predictability Tool

Seasonal rainfall prediction is potentially valuable input for agriculture and allied sectors. Climate Predictability Tool (CPT) is a computational statistical software and intended to facilitate the widespread development and dissemination of quality-controlled seasonal climate forecasts. In the present study, CPT was used to predict South West Monsoon (SWM) rainfall by utilizing the Canonical Correlation Analysis (CCA) technique. Results indicate that observed sea surface temperature (SST) from the ERSSTv5 model of March-April-May (MAM) months for the Nino 3.4 region exhibited a strong association with rainfall (the CCA score of 0.71). Thus, MAM SST was used to predict the SWM (2021) rainfall for 32 districts of Tamil Nadu. The prediction accuracy was tested with ROC score, Pearson correlation and 2AFC score. The good predictive skill of cross-validated hindcast in the below and the above-normal category of rainfall (1981-2020) was observed in 17 districts and poor discrimination was observed in 15 districts of Tamil Nadu.

A Vision on Extensions of Canonical Correlation Analysis for Data Interpretation Applications in Compatible Semantic Communication

Background: A comprehensive approach to Canonical Correlation Analysis (CCA) technique that explicitly enhances data interpretation by encountering semantic barriers in communication is proposed. Object: To the extent that there exist potential inconsistencies due to redundancy and misinterpretation of data attributes, compatibility with respect to data interpretation may defer. For a consolidated and technology dependent network infrastructure, the concept of inclusive CCA (such as linear CCA, sparse CCA and kernel CCA) further asserts the inclusion of statistical correlational analysis in semantic communication. Methods: A Singular Value Decomposition (SVD) based Latent Semantic Indexing (LSI) method is substantiated upon a linear dataset and simulation results are canonically analyzed for the same. Results: Favorably, the p-value analysis from the t-test validates the significance of the application of extensions of CCA in the field of semantic communication. Conclusion: Hence, CCA as a statistical technique incorporates both symmetric as well as asymmetric multivariate data analysis to help delineate the incompatibility caused due to subtle semantic- defects.

The effect of groupness constraint on the sensitivity and specificity of canonical correlation analysis, a multi-modal anatomical and functional MRI study

While neuroimaging studies reveal that several brain regions may participate in multiple groups (networks), this group overlap is neglected in multi-modal data fusion frameworks. Indeed, it is not clear how much “information” is lost due to this negligence. To study this issue, we present a group-structured sparse canonical correlation analysis (gssCCA) technique by utilizing groupness and sparsity constraints in a unified fusion framework. The approach allows to: 1) compare structures of disjoint and overlapping groups (networks); and 2) consider appropriate levels of overlap among groups. We use simulations to investigate the performance of the proposed approach and compare overlapping gssCCA, disjoint gssCCA, and ssCCA. The gains of considering an overlapping groupness constraint are significant since it can increase detection sensitivity and specificity of associations between multi-modal datasets. They also demonstrate that, even when the ROIs are assumed to be disjoint, the lost structural information is less than the conditions that we do not have any groupness information. We also apply the methods to experimental anatomical magnetic resonance imaging (MRI) and resting-state functional MRI (rs-fMRI) data of schizophrenia (SZ) patients and control subjects (CS). The results show that the first pair of canonical variates (CVs) capture better classification accuracy between SZ and CS with a correlation of 88 % (p-value of less than 1 × 10−6). The extracted CVs show the most correlated clusters between anatomical and functional datasets. These clusters explain differences between the two groups in their own modality that are maximally correlated with the corresponding clusters in the other modality.

Disjoint subspaces for common and distinct component analysis: Application to the fusion of multi-task FMRI data

BackgroundData-driven methods such as independent component analysis (ICA) makes very few assumptions on the data and the relationships of multiple datasets, and hence, are attractive for the fusion of medical imaging data. Two important extensions of ICA for multiset fusion are the joint ICA (jICA) and the multiset canonical correlation analysis and joint ICA (MCCA-jICA) techniques. Both approaches assume identical mixing matrices, emphasizing components that are common across the multiple datasets. However, in general, one would expect to have components that are both common across the datasets and distinct to each dataset. New methodWe propose a general framework, disjoint subspace analysis using ICA (DS-ICA), which identifies and extracts not only the common but also the distinct components across multiple datasets. A key component of the method is the identification of these subspaces and their separation before subsequent analyses, which helps establish better model match and provides flexibility in algorithm and order choice. ComparisonWe compare DS-ICA with jICA and MCCA-jICA through both simulations and application to multiset functional magnetic resonance imaging (fMRI) task data collected from healthy controls as well as patients with schizophrenia. ResultsThe results show DS-ICA estimates more components discriminative between healthy controls and patients than jICA and MCCA-jICA, and with higher discriminatory power showing activation differences in meaningful regions. When applied to a classification framework, components estimated by DS-ICA results in higher classification performance for different dataset combinations than the other two methods. ConclusionThese results demonstrate that DS-ICA is an effective method for fusion of multiple datasets.

Assessment of water pollution induced by anthropogenic activities on zooplankton community in Mariout Lake using statistical simulation

The growing increase in population and urbanization expansion leads to corresponding increase in urban effluents, industrial and agriculture that discharged into the aquatic environment. The present paper is an attempt to estimate the water quality characteristics and zooplankton community of different water bodies in some human activities effects of Mariout Lake subjected to various types of pollutants by using statistical simulation. During the period beginning from March 2019 to February 2020, ten representative surface water samples were collected seasonally from Mariout Lake for determination of physico-chemical parameters, namely temperature, biological oxygen demand, salinity, hydrogen ion concentration, ammonia, dissolved oxygen, reactive phosphate, and chlorophyll were estimated. The results of the monitoring stations were analyzed by multivariate statistical techniques to measure relationship among physical and chemical parameters by principal component analysis (PCA) and canonical correlation analysis (CCA) techniques. A total of 78 species were from zooplankton population, additional to fish eggs and the larval stages of different groups. Rotifer formed 48 species, copepods formed 8 species additional nauplii larvae and copepodite stage, protozoans (6 species: one of each of Protozoan, Ciliatea, tintinnids, and Sessilida), Cladocera 5 species, Ostracoda 4 species, Insecta 2 larvea, Mollusca 1 species and one larvae, while Annelida, Nematoda, Cirripeda, Molluscs, Annelida, Amphipoda, and fish eggs were only represented by one organism each. The annual average values of physico-chemical parameters were 24.26 °C for temperature; 4.07 ± 1.28‰ for salinity; 8.015 ± 0.22 for pH; 9.07 ± 9.56 mg/L for DO; 19.49 ± 5.22 mg/L for BOD; 0.54 ± 0.27 mg/L foe ammonia; 0.426 μM for reactive phosphate; and 6.312 μg/L for chlorophyll-a. From the calculations of WQI, the lakes were categorized into good, bad, and poor water quality. Increasing pollution rate of Mariout Lake will create large areas of oxygen-depleted (hypoxic) dead zones, which affect fish, zooplankton, and other organisms in the lake. Verifying the ability of CCA and PCA techniques was achieved by correlation methods and simple regression respectively.

Targeting interventions for HIV testing and treatment uptake: An attitudinal and behavioural segmentation of men aged 20–34 in KwaZulu-Natal and Mpumalanga, South Africa

Despite recent improvements, men still have worse HIV outcomes than women in South Africa. This study describes how young men form distinct behavioural and attitudinal subgroups, and is intended to inform the design of targeted interventions to encourage HIV testing and initiation on antiretroviral therapy. Data were collected using a cross-sectional survey with questions on men’s attitudes, beliefs and behaviours around HIV/AIDS. A total of 2,019 men were randomly sampled from eight district municipalities in KwaZulu-Natal and Mpumalanga provinces between October 2018 and January 2019. Men were eligible to participate if they were aged 20–34, Black African, had an education level below university graduation, were aware of HIV and were willing to disclose whether they had tested for HIV. Each participant responded to a questionnaire asking about their demographic characteristics, reported sexual behaviour, engagement with HIV testing and treatment services, alcohol consumption, HIV knowledge, attitudes to gender equity and reported level of depressive symptoms. Data were analysed using canonical correlation, hierarchical clustering and factor analysis techniques to produce five groups of men. The results were synthesised using Human Centred Design principles to suggests areas for potential intervention for each segment. The results showed that men vary based on their attitudes to gender and masculinity, use of alcohol, testing and treatment behaviour, HIV-related fears and preferences for testing modalities. Segment 1 (21%) avoids the topic of HIV, perhaps fearful of the impact on his life. Segment 2 (23%) is well connected to his community and has social concerns about HIV. Segment 3 (15%) struggles with more distal determinants of HIV acquisition such as unemployment and poor mental health. Segment 4 (25%) has concerns about the lifestyle changes that would be required if he were HIV positive. Segment 5 (16%) has a strong traditional mindset and is fearful of the ramifications of HIV in his community. The results will be used to design targeted interventions to increase HIV testing and treatment rates among young men in South Africa. Further research is required to understand the impact of interventions designed in this way.

Feature concatenation for scene classification of VHR images based on canonical correlation analysis.

The recent advancements of technology in remote sensing enable us to get very high-resolution images (VHR). To do scene classification in these images turned significant and become a challenging problem due to the lack of availability of adequate labelled data. We get over fitting problems by training a limited amount of labelled data. Considering the features obtained by deep learning convolutional nets as inputs we address this issue. Here we utilize the existing VGG16, Alex Net frameworks as a feature extractor to extricate informative features from the authentic VHR Images. Succeeding, discuss feature concatenation and classification framework based on Canonical Correlation Analysis (CCA). This strategy uses the correspondence of two feature vectors of discriminant information and eradicates redundant information within the features. This permits a more effective approach than conventional extraction techniques exclusively. The experimental results demonstrate that the feature concatenation strategy based on the CCA technique produces good informative features and accomplishes a higher accuracy with much dimension reduction than exclusively using the raw deep features. We use a 0.3 sub-meter resolution UC MERCED data set to explore our approach.

The Relationship between Vocational Personality Types and Vocational Values

In this study, the relationship between high school students’ vocational personality types and vocational values was examined. The study data were collected from 488 high school students going to schools in Niğde city center. In the study, Vocational Personality Types Inventory and Vocational Value Inventory were administered. Canonical correlation analysis technique was used to analyze the data. The study results revealed that the canonical model was significant, and when the data on canonical functions were evaluated separately, three canonical functions were found significant. The common variance shared between the data sets of the canonical variables showing that there was a significant relationship between occupational personality types and occupational values were 44% for the first canonical function, 30% for the second canonical function and 11% for the third canonical function, respectively. The findings were discussed in the light of the related literature.

Tensor network factorizations: Relationships between brain structural connectomes and traits

Advanced brain imaging techniques make it possible to measure individuals’ structural connectomes in large cohort studies non-invasively. Given the availability of large scale data sets, it is extremely interesting and important to build a set of advanced tools for structural connectome extraction and statistical analysis that emphasize both interpretability and predictive power. In this paper, we developed and integrated a set of toolboxes, including an advanced structural connectome extraction pipeline and a novel tensor network principal components analysis (TN-PCA) method, to study relationships between structural connectomes and various human traits such as alcohol and drug use, cognition and motion abilities. The structural connectome extraction pipeline produces a set of connectome features for each subject that can be organized as a tensor network, and TN-PCA maps the high-dimensional tensor network data to a lower-dimensional Euclidean space. Combined with classical hypothesis testing, canonical correlation analysis and linear discriminant analysis techniques, we analyzed over 1100 scans of 1076 subjects from the Human Connectome Project (HCP) and the Sherbrooke test-retest data set, as well as 175 human traits measuring different domains including cognition, substance use, motor, sensory and emotion. The test-retest data validated the developed algorithms. With the HCP data, we found that structural connectomes are associated with a wide range of traits, e.g., fluid intelligence, language comprehension, and motor skills are associated with increased cortical-cortical brain structural connectivity, while the use of alcohol, tobacco, and marijuana are associated with decreased cortical-cortical connectivity. We also demonstrated that our extracted structural connectomes and analysis method can give superior prediction accuracies compared with alternative connectome constructions and other tensor and network regression methods.

Determining water storage depletion within Iran by assimilating GRACE data into the W3RA hydrological model

Abstract Groundwater depletion, due to both unsustainable water use and a decrease in precipitation, has been reported in many parts of Iran. In order to analyze these changes during the recent decade, in this study, we assimilate Terrestrial Water Storage (TWS) data from the Gravity Recovery And Climate Experiment (GRACE) into the World-Wide Water Resources Assessment (W3RA) model. This assimilation improves model derived water storage simulations by introducing missing trends and correcting the amplitude and phase of seasonal water storage variations. The Ensemble Square-Root Filter (EnSRF) technique is applied, which showed stable performance in propagating errors during the assimilation period (2002–2012). Our focus is on sub-surface water storage changes including groundwater and soil moisture variations within six major drainage divisions covering the whole Iran including its eastern part (East), Caspian Sea, Centre, Sarakhs, Persian Gulf and Oman Sea, and Lake Urmia. Results indicate an average of -8.9 mm/year groundwater reduction within Iran during the period 2002 to 2012. A similar decrease is also observed in soil moisture storage especially after 2005. We further apply the canonical correlation analysis (CCA) technique to relate sub-surface water storage changes to climate (e.g., precipitation) and anthropogenic (e.g., farming) impacts. Results indicate an average correlation of 0.81 between rainfall and groundwater variations and also a large impact of anthropogenic activities (mainly for irrigations) on Iran’s water storage depletions.

Assessment of mental stress effects on prefrontal cortical activities using canonical correlation analysis: an fNIRS-EEG study.

This paper presents an investigation about the effects of mental stress on prefrontal cortex (PFC) subregions using simultaneous measurement of functional Near-Infrared Spectroscopy (fNIRS) and Electroencephalography (EEG) signals. The aim is to explore canonical correlation analysis (CCA) technique to study the relationship among the bi-modality signals in mental stress assessment, and how we could fuse the signals for better accuracy in stress detection. Twenty-five male healthy subjects participated in the study while performing mental arithmetic task under control and stress (under time pressure with negative feedback) conditions. The fusion of brain signals acquired by fNIRS-EEG was performed at feature-level using CCA by maximizing the inter-subject covariance across modalities. The CCA result discovered the associations across the modalities and estimated the components responsible for these associations. The experiment results showed that mental stress experienced by this cohort of subjects is subregion specific and localized to the right ventrolateral PFC subregion. These suggest the right ventrolateral PFC as a suitable candidate region to extract biomarkers as performance indicators of neurofeedback training in stress coping.

Structured and Sparse Canonical Correlation Analysis as a Brain-Wide Multi-Modal Data Fusion Approach.

Multi-modal data fusion has recently emerged as a comprehensive neuroimaging analysis approach, which usually uses canonical correlation analysis (CCA). However, the current CCA-based fusion approaches face problems like high-dimensionality, multi-collinearity, unimodal feature selection, asymmetry, and loss of spatial information in reshaping the imaging data into vectors. This paper proposes a structured and sparse CCA (ssCCA) technique as a novel CCA method to overcome the above problems. To investigate the performance of the proposed algorithm, we have compared three data fusion techniques: standard CCA, regularized CCA, and ssCCA, and evaluated their ability to detect multi-modal data associations. We have used simulations to compare the performance of these approaches and probe the effects of non-negativity constraint, the dimensionality of features, sample size, and noise power. The results demonstrate that ssCCA outperforms the existing standard and regularized CCA-based fusion approaches. We have also applied the methods to real functional magnetic resonance imaging (fMRI) and structural MRI data of Alzheimer's disease (AD) patients (n = 34) and healthy control (HC) subjects (n = 42) from the ADNI database. The results illustrate that the proposed unsupervised technique differentiates the transition pattern between the subject-course of AD patients and HC subjects with a p-value of less than 1×10-6 . Furthermore, we have depicted the brain mapping of functional areas that are most correlated with the anatomical changes in AD patients relative to HC subjects.