Bayesian optimization for sparse neural networks with trainable activation functions.

In the literature on deep neural networks, there is considerable interest in developing activation functions that can enhance neural network performance. In recent years, there has been renewed scientific interest in proposing activation functions that can be trained throughout the learning process, as they appear to improve network performance, especially by reducing overfitting. In this paper, we propose a trainable activation function whose parameters need to be estimated. A fully Bayesian model is developed to automatically estimate from the learning data both the model weights and activation function parameters. An MCMC-based optimization scheme is developed to build the inference. The proposed method aims to solve the aforementioned problems and improve convergence time by using an efficient sampling scheme that guarantees convergence to the global maximum. The proposed scheme has been tested across a diverse datasets, encompassing both classification and regression tasks, and implemented in various CNN architectures to demonstrate its versatility and effectiveness. Promising results demonstrate the usefulness of our proposed approach in improving models accuracy due to the proposed activation function and Bayesian estimation of the parameters.

mid-DeepLabv3+: A Novel Approach for Image Semantic Segmentation Applied to African Food Dietary Assessments.

Recent decades have witnessed the development of vision-based dietary assessment (VBDA) systems. These systems generally consist of three main stages: food image analysis, portion estimation, and nutrient derivation. The effectiveness of the initial step is highly dependent on the use of accurate segmentation and image recognition models and the availability of high-quality training datasets. Food image segmentation still faces various challenges, and most existing research focuses mainly on Asian and Western food images. For this reason, this study is based on food images from sub-Saharan Africa, which pose their own problems, such as inter-class similarity and dishes with mixed-class food. This work focuses on the first stage of VBDAs, where we introduce two notable contributions. Firstly, we propose mid-DeepLabv3+, an enhanced food image segmentation model based on DeepLabv3+ with a ResNet50 backbone. Our approach involves adding a middle layer in the decoder path and SimAM after each extracted backbone feature layer. Secondly, we present CamerFood10, the first food image dataset specifically designed for sub-Saharan African food segmentation. It includes 10 classes of the most consumed food items in Cameroon. On our dataset, mid-DeepLabv3+ outperforms benchmark convolutional neural network models for semantic image segmentation, with an mIoU (mean Intersection over Union) of 65.20%, representing a +10.74% improvement over DeepLabv3+ with the same backbone.

Non-smooth Bayesian learning for artificial neural networks.

Artificial neural networks (ANNs) are being widely used in supervised machine learning to analyze signals or images for many applications. Using an annotated learning database, one of the main challenges is to optimize the network weights. A lot of work on solving optimization problems or improving optimization methods in machine learning has been proposed successively such as gradient-based method, Newton-type method, meta-heuristic method. For the sake of efficiency, regularization is generally used. When non-smooth regularizers are used especially to promote sparse networks, such as the ℓ 1 norm, this optimization becomes challenging due to non-differentiability issues of the target criterion. In this paper, we propose an MCMC-based optimization scheme formulated in a Bayesian framework. The proposed scheme solves the above-mentioned sparse optimization problem using an efficient sampling scheme and Hamiltonian dynamics. The designed optimizer is conducted on four (4) datasets, and the results are verified by a comparative study with two CNNs. Promising results show the usefulness of the proposed method to allow ANNs, even with low complexity levels, reaching high accuracy rates of up to 94 % . The proposed method is also faster and more robust concerning overfitting issues. More importantly, the training step of the proposed method is much faster than all competing algorithms.

Globally altered sleep patterns and physical activity levels by confinement in 5056 individuals: ECLB COVID-19 international online survey.

Symptoms of psychological distress and disorder have been widely reported in people under quarantine during the COVID-19 pandemic; in addition to severe disruption of peoples' daily activity and sleep patterns. This study investigates the association between physical-activity levels and sleep patterns in quarantined individuals. An international Google online survey was launched in April 6th, 2020 for 12-weeks. Forty-one research organizations from Europe, North-Africa, Western-Asia, and the Americas promoted the survey through their networks to the general society, which was made available in 14 languages. The survey was presented in a differential format with questions related to responses "before" and "during" the confinement period. Participants responded to the Pittsburgh Sleep Quality Index (PSQI) questionnaire and the short form of the International Physical Activity Questionnaire. 5056 replies (59.4% female), from Europe (46.4%), Western-Asia (25.4%), America (14.8%) and North-Africa (13.3%) were analysed. The COVID-19 home confinement led to impaired sleep quality, as evidenced by the increase in the global PSQI score (4.37 ± 2.71 before home confinement vs. 5.32 ± 3.23 during home confinement) (p < 0.001). The frequency of individuals experiencing a good sleep decreased from 61% (n = 3063) before home confinement to 48% (n = 2405) during home confinement with highly active individuals experienced better sleep quality (p < 0.001) in both conditions. Time spent engaged in all physical-activity and the metabolic equivalent of task in each physical-activity category (i.e., vigorous, moderate, walking) decreased significantly during COVID-19 home confinement (p < 0.001). The number of hours of daily-sitting increased by ~2 hours/days during home confinement (p < 0.001). COVID-19 home confinement resulted in significantly negative alterations in sleep patterns and physical-activity levels. To maintain health during home confinement, physical-activity promotion and sleep hygiene education and support are strongly warranted.

Effects of home confinement on mental health and lifestyle behaviours during the COVID-19 outbreak: insights from the ECLB-COVID19 multicentre study.

Although recognised as effective measures to curb the spread of the COVID-19 outbreak, social distancing and self-isolation have been suggested to generate a burden throughout the population. To provide scientific data to help identify risk factors for the psychosocial strain during the COVID-19 outbreak, an international cross-disciplinary online survey was circulated in April 2020. This report outlines the mental, emotional and behavioural consequences of COVID-19 home confinement. The ECLB-COVID19 electronic survey was designed by a steering group of multidisciplinary scientists, following a structured review of the literature. The survey was uploaded and shared on the Google online survey platform and was promoted by thirty-five research organizations from Europe, North Africa, Western Asia and the Americas. Questions were presented in a differential format with questions related to responses "before" and "during" the confinement period. 1047 replies (54% women) from Western Asia (36%), North Africa (40%), Europe (21%) and other continents (3%) were analysed. The COVID-19 home confinement evoked a negative effect on mental wellbeing and emotional status (P < 0.001; 0.43 ≤ d ≤ 0.65) with a greater proportion of individuals experiencing psychosocial and emotional disorders (+10% to +16.5%). These psychosocial tolls were associated with unhealthy lifestyle behaviours with a greater proportion of individuals experiencing (i) physical (+15.2%) and social (+71.2%) inactivity, (ii) poor sleep quality (+12.8%), (iii) unhealthy diet behaviours (+10%), and (iv) unemployment (6%). Conversely, participants demonstrated a greater use (+15%) of technology during the confinement period. These findings elucidate the risk of psychosocial strain during the COVID-19 home confinement period and provide a clear remit for the urgent implementation of technology-based intervention to foster an Active and Healthy Confinement Lifestyle AHCL).

Convolutional Neural Network for Drowsiness Detection Using EEG Signals.

Drowsiness detection (DD) has become a relevant area of active research in biomedical signal processing. Recently, various deep learning (DL) researches based on the EEG signals have been proposed to detect fatigue conditions. The research presented in this paper proposes an EEG classification system for DD based on DL networks. However, the proposed DD system is mainly realized into two procedures; (i) data acquisition and (ii) model analysis. For the data acquisition procedure, two key steps are considered, which are the signal collection using a wearable Emotiv EPOC+ headset to record 14 channels of EEG, and the signal annotation. Furthermore, a data augmentation (DA) step has been added to the proposed system to overcome the problem of over-fitting and to improve accuracy. As regards the model analysis, a comparative study is also introduced in this paper to argue the choice of DL architecture and frameworks used in our DD system. In this sense, The proposed DD protocol makes use of a convolutional neural network (CNN) architecture implemented using the Keras library. The results showed a high accuracy value (90.42%) in drowsy/awake discrimination and revealed the efficiency of the proposed DD system compared to other research works.

Sleep Quality and Physical Activity as Predictors of Mental Wellbeing Variance in Older Adults during COVID-19 Lockdown: ECLB COVID-19 International Online Survey.

BACKGROUND: The COVID-19 lockdown could engender disruption to lifestyle behaviors, thus impairing mental wellbeing in the general population. This study investigated whether sociodemographic variables, changes in physical activity, and sleep quality from pre- to during lockdown were predictors of change in mental wellbeing in quarantined older adults. METHODS: A 12-week international online survey was launched in 14 languages on 6 April 2020. Forty-one research institutions from Europe, Western-Asia, North-Africa, and the Americas, promoted the survey. The survey was presented in a differential format with questions related to responses "pre" and "during" the lockdown period. Participants responded to the Short Warwick-Edinburgh Mental Wellbeing Scale, the Pittsburgh Sleep Quality Index (PSQI) questionnaire, and the short form of the International Physical Activity Questionnaire. RESULTS: Replies from older adults (aged >55 years, n = 517), mainly from Europe (50.1%), Western-Asia (6.8%), America (30%), and North-Africa (9.3%) were analyzed. The COVID-19 lockdown led to significantly decreased mental wellbeing, sleep quality, and total physical activity energy expenditure levels (all p < 0.001). Regression analysis showed that the change in total PSQI score and total physical activity energy expenditure (F(2, 514) = 66.41 p < 0.001) were significant predictors of the decrease in mental wellbeing from pre- to during lockdown (p < 0.001, R2: 0.20). CONCLUSION: COVID-19 lockdown deleteriously affected physical activity and sleep patterns. Furthermore, change in the total PSQI score and total physical activity energy expenditure were significant predictors for the decrease in mental wellbeing.

Psychological consequences of COVID-19 home confinement: The ECLB-COVID19 multicenter study.

BACKGROUND: Public health recommendations and government measures during the COVID-19 pandemic have enforced restrictions on daily-living. While these measures are imperative to abate the spreading of COVID-19, the impact of these restrictions on mental health and emotional wellbeing is undefined. Therefore, an international online survey (ECLB-COVID19) was launched on April 6, 2020 in seven languages to elucidate the impact of COVID-19 restrictions on mental health and emotional wellbeing. METHODS: The ECLB-COVID19 electronic survey was designed by a steering group of multidisciplinary scientists, following a structured review of the literature. The survey was uploaded and shared on the Google online-survey-platform and was promoted by thirty-five research organizations from Europe, North-Africa, Western-Asia and the Americas. All participants were asked for their mental wellbeing (SWEMWS) and depressive symptoms (SMFQ) with regard to "during" and "before" home confinement. RESULTS: Analysis was conducted on the first 1047 replies (54% women) from Asia (36%), Africa (40%), Europe (21%) and other (3%). The COVID-19 home confinement had a negative effect on both mental-wellbeing and on mood and feelings. Specifically, a significant decrease (p < .001 and Δ% = 9.4%) in total score of the SWEMWS questionnaire was noted. More individuals (+12.89%) reported a low mental wellbeing "during" compared to "before" home confinement. Furthermore, results from the mood and feelings questionnaire showed a significant increase by 44.9% (p < .001) in SMFQ total score with more people (+10%) showing depressive symptoms "during" compared to "before" home confinement. CONCLUSION: The ECLB-COVID19 survey revealed an increased psychosocial strain triggered by the home confinement. To mitigate this high risk of mental disorders and to foster an Active and Healthy Confinement Lifestyle (AHCL), a crisis-oriented interdisciplinary intervention is urgently needed.

COVID-19 Home Confinement Negatively Impacts Social Participation and Life Satisfaction: A Worldwide Multicenter Study.

Public health recommendations and governmental measures during the new coronavirus disease (COVID-19) pandemic have enforced numerous restrictions on daily living including social distancing, isolation, and home confinement. While these measures are imperative to mitigate spreading of COVID-19, the impact of these restrictions on psychosocial health is undefined. Therefore, an international online survey was launched in April 2020 to elucidate the behavioral and lifestyle consequences of COVID-19 restrictions. This report presents the preliminary results from more than one thousand responders on social participation and life satisfaction. METHODS: Thirty-five research organizations from Europe, North-Africa, Western Asia, and the Americas promoted the survey through their networks to the general society, in 7 languages (English, German, French, Arabic, Spanish, Portuguese, and Slovenian). Questions were presented in a differential format with questions related to responses "before" and "during" confinement conditions. RESULTS: 1047 participations (54% women) from Asia (36%), Africa (40%), Europe (21%), and others (3%) were included in the analysis. Findings revealed psychosocial strain during the enforced COVID-19 home confinement. Large decreases (p < 0.001) in the amount of social activity through family (-58%), friends/neighbors (-44.9%), or entertainment (-46.7%) were triggered by the enforced confinement. These negative effects on social participation were also associated with lower life satisfaction (-30.5%) during the confinement period. Conversely, the social contact score through digital technologies significantly increased (p < 0.001) during the confinement period with more individuals (+24.8%) being socially connected through digital technology. CONCLUSION: These preliminary findings elucidate the risk of psychosocial strain during the early COVID-19 home confinement period in 2020. Therefore, in order to mitigate the negative psychosocial effects of home confinement, implementation of national strategies focused on promoting social inclusion through a technology-based solution is strongly suggested.

Effects of COVID-19 Home Confinement on Eating Behaviour and Physical Activity: Results of the ECLB-COVID19 International Online Survey.

BACKGROUND: Public health recommendations and governmental measures during the COVID-19 pandemic have resulted in numerous restrictions on daily living including social distancing, isolation and home confinement. While these measures are imperative to abate the spreading of COVID-19, the impact of these restrictions on health behaviours and lifestyles at home is undefined. Therefore, an international online survey was launched in April 2020, in seven languages, to elucidate the behavioural and lifestyle consequences of COVID-19 restrictions. This report presents the results from the first thousand responders on physical activity (PA) and nutrition behaviours. METHODS: Following a structured review of the literature, the "Effects of home Confinement on multiple Lifestyle Behaviours during the COVID-19 outbreak (ECLB-COVID19)" Electronic survey was designed by a steering group of multidisciplinary scientists and academics. The survey was uploaded and shared on the Google online survey platform. Thirty-five research organisations from Europe, North-Africa, Western Asia and the Americas promoted the survey in English, German, French, Arabic, Spanish, Portuguese and Slovenian languages. Questions were presented in a differential format, with questions related to responses "before" and "during" confinement conditions. RESULTS: 1047 replies (54% women) from Asia (36%), Africa (40%), Europe (21%) and other (3%) were included in the analysis. The COVID-19 home confinement had a negative effect on all PA intensity levels (vigorous, moderate, walking and overall). Additionally, daily sitting time increased from 5 to 8 h per day. Food consumption and meal patterns (the type of food, eating out of control, snacks between meals, number of main meals) were more unhealthy during confinement, with only alcohol binge drinking decreasing significantly. CONCLUSION: While isolation is a necessary measure to protect public health, results indicate that it alters physical activity and eating behaviours in a health compromising direction. A more detailed analysis of survey data will allow for a segregation of these responses in different age groups, countries and other subgroups, which will help develop interventions to mitigate the negative lifestyle behaviours that have manifested during the COVID-19 confinement.

Covid-19 pandemic by the "real-time" monitoring: the Tunisian case and lessons for global epidemics in the context of 3PM strategies.

Covid-19 is neither the first nor the last viral epidemic which societies around the world are, were and will be affected by. Which lessons should be taken from the current pandemic situation? The Covid-19 disease is still not well characterised, and many research teams all over the world are working on prediction of the epidemic scenario, protective measures to populations and sub-populations, therapeutic and vaccination issues, amongst others. Contextually, countries with currently low numbers of Covid-19-infected individuals such as Tunisia are intended to take lessons from those countries which already reached the exponential phase of the infection distribution as well as from those which have the exponential phase behind them and record a minor number of new cases such as China. To this end, in Tunisia, the pandemic wave has started with a significant delay compared with Europe, the main economic partner of the country. In this paper, we do analyse the current pandemic situation in this country by studying the infection evolution and considering potential protective strategies to prevent a pandemic scenario. The model is predictive based on a large number of undetected Covid-19 cases that is particularly true for some country regions such as Sfax. Infection distribution and mortality rate analysis demonstrate a highly heterogeneous picture over the country. Qualitative and quantitative comparative analysis leads to a conclusion that the reliable "real-time" monitoring based on the randomised laboratory tests is the optimal predictive strategy to create the most effective evidence-based preventive measures. In contrast, lack of tests may lead to incorrect political decisions causing either unnecessary over-protection of the population that is risky for a long-term economic recession, or under-protection of the population leading to a post-containment pandemic rebound. Recommendations are provided in the context of advanced predictive, preventive and personalised (3P) medical approach.

Spatio-spectral regularization to improve magnetic resonance spectroscopic imaging quantification.

Magnetic resonance spectroscopic imaging (MRSI) is a non-invasive technique able to provide the spatial distribution of relevant biochemical compounds commonly used as biomarkers of disease. Information provided by MRSI can be used as a valuable insight for the diagnosis, treatment and follow-up of several diseases such as cancer or neurological disorders. Obtaining accurate metabolite concentrations from in vivo MRSI signals is a crucial requirement for the clinical utility of this technique. Despite the numerous publications on the topic, accurate quantification is still a challenging problem due to the low signal-to-noise ratio of the data, overlap of spectral lines and the presence of nuisance components. We propose a novel quantification method, which alleviates these limitations by exploiting a spatio-spectral regularization scheme. In contrast to previous methods, the regularization terms are not expressed directly on the parameters being sought, but on appropriate transformed domains. In order to quantify all signals simultaneously in the MRSI grid, while introducing prior information, a fast proximal optimization algorithm is proposed. Experiments on synthetic MRSI data demonstrate that the error in the estimated metabolite concentrations is reduced by a mean of 41% with the proposed scheme. Results on in vivo brain MRSI data show the benefit of the proposed approach, which is able to fit overlapping peaks correctly and to capture metabolites that are missed by single-voxel methods due to their lower concentrations. Copyright © 2016 John Wiley & Sons, Ltd.

Sparse EEG Source Localization Using Bernoulli Laplacian Priors.

Source localization in electroencephalography has received an increasing amount of interest in the last decade. Solving the underlying ill-posed inverse problem usually requires choosing an appropriate regularization. The usual l2 norm has been considered and provides solutions with low computational complexity. However, in several situations, realistic brain activity is believed to be focused in a few focal areas. In these cases, the l2 norm is known to overestimate the activated spatial areas. One solution to this problem is to promote sparse solutions for instance based on the l1 norm that are easy to handle with optimization techniques. In this paper, we consider the use of an l0 + l1 norm to enforce sparse source activity (by ensuring the solution has few nonzero elements) while regularizing the nonzero amplitudes of the solution. More precisely, the l0 pseudonorm handles the position of the nonzero elements while the l1 norm constrains the values of their amplitudes. We use a Bernoulli-Laplace prior to introduce this combined l0 + l1 norm in a Bayesian framework. The proposed Bayesian model is shown to favor sparsity while jointly estimating the model hyperparameters using a Markov chain Monte Carlo sampling technique. We apply the model to both simulated and real EEG data, showing that the proposed method provides better results than the l2 and l1  norms regularizations in the presence of pointwise sources. A comparison with a recent method based on multiple sparse priors is also conducted.

Flexible multivariate hemodynamics fMRI data analyses and simulations with PyHRF.

As part of fMRI data analysis, the pyhrf package provides a set of tools for addressing the two main issues involved in intra-subject fMRI data analysis: (1) the localization of cerebral regions that elicit evoked activity and (2) the estimation of activation dynamics also known as Hemodynamic Response Function (HRF) recovery. To tackle these two problems, pyhrf implements the Joint Detection-Estimation framework (JDE) which recovers parcel-level HRFs and embeds an adaptive spatio-temporal regularization scheme of activation maps. With respect to the sole detection issue (1), the classical voxelwise GLM procedure is also available through nipy, whereas Finite Impulse Response (FIR) and temporally regularized FIR models are concerned with HRF estimation (2) and are specifically implemented in pyhrf. Several parcellation tools are also integrated such as spatial and functional clustering. Parcellations may be used for spatial averaging prior to FIR/RFIR analysis or to specify the spatial support of the HRF estimates in the JDE approach. These analysis procedures can be applied either to volume-based data sets or to data projected onto the cortical surface. For validation purpose, this package is shipped with artificial and real fMRI data sets, which are used in this paper to compare the outcome of the different available approaches. The artificial fMRI data generator is also described to illustrate how to simulate different activation configurations, HRF shapes or nuisance components. To cope with the high computational needs for inference, pyhrf handles distributing computing by exploiting cluster units as well as multi-core machines. Finally, a dedicated viewer is presented, which handles n-dimensional images and provides suitable features to explore whole brain hemodynamics (time series, maps, ROI mask overlay).

Spatio-temporal wavelet regularization for parallel MRI reconstruction: application to functional MRI.

BACKGROUND: Parallel magnetic resonance imaging (MRI) is a fast imaging technique that helps acquiring highly resolved images in space/time. Its performance depends on the reconstruction algorithm, which can proceed either in the k-space or in the image domain. OBJECTIVE AND METHODS: To improve the performance of the widely used SENSE algorithm, 2D regularization in the wavelet domain has been investigated. In this paper, we first extend this approach to 3D-wavelet representations and the 3D sparsity-promoting regularization term, in order to address reconstruction artifacts that propagate across adjacent slices. The resulting optimality criterion is convex but nonsmooth, and we resort to the parallel proximal algorithm to minimize it. Second, to account for temporal correlation between successive scans in functional MRI (fMRI), we extend our first contribution to 3D + t acquisition schemes by incorporating a prior along the time axis into the objective function. RESULTS: Our first method (3D-UWR-SENSE) is validated on T1-MRI anatomical data for gray/white matter segmentation. The second method (4D-UWR-SENSE) is validated for detecting evoked activity during a fast event-related functional MRI protocol. CONCLUSION: We show that our algorithm outperforms the SENSE reconstruction at the subject and group levels (15 subjects) for different contrasts of interest (motor or computation tasks) and two parallel acceleration factors (R = 2 and R = 4) on 2 × 2 × 3 MM(3) echo planar imaging (EPI) images.

Hybrid sparse regularization for magnetic resonance spectroscopy.

Magnetic resonance spectroscopy imaging (MRSI) is a powerful non-invasive tool for characterising markers of biological processes. This technique extends conventional MRI by providing an additional dimension of spectral information describing the abnormal presence or concentration of metabolites of interest. Unfortunately, in vivo MRSI suffers from poor signal-to-noise ratio limiting its clinical use for treatment purposes. This is due to the combination of a weak MR signal and low metabolite concentrations, in addition to the acquisition noise. We propose a new method that handles this challenge by efficiently denoising MRSI signals without constraining the spectral or spatial profiles. The proposed denoising approach is based on wavelet transforms and exploits the sparsity of the MRSI signals both in the spatial and frequency domains. A fast proximal optimization algorithm is then used to recover the optimal solution. Experiments on synthetic and real MRSI data showed that the proposed scheme achieves superior noise suppression (SNR increase up to 60%). In addition, this method is computationally efficient and preserves data features better than existing methods.

Fast joint detection-estimation of evoked brain activity in event-related FMRI using a variational approach.

In standard within-subject analyses of event-related functional magnetic resonance imaging (fMRI) data, two steps are usually performed separately: detection of brain activity and estimation of the hemodynamic response. Because these two steps are inherently linked, we adopt the so-called region-based joint detection-estimation (JDE) framework that addresses this joint issue using a multivariate inference for detection and estimation. JDE is built by making use of a regional bilinear generative model of the BOLD response and constraining the parameter estimation by physiological priors using temporal and spatial information in a Markovian model. In contrast to previous works that use Markov Chain Monte Carlo (MCMC) techniques to sample the resulting intractable posterior distribution, we recast the JDE into a missing data framework and derive a variational expectation-maximization (VEM) algorithm for its inference. A variational approximation is used to approximate the Markovian model in the unsupervised spatially adaptive JDE inference, which allows automatic fine-tuning of spatial regularization parameters. It provides a new algorithm that exhibits interesting properties in terms of estimation error and computational cost compared to the previously used MCMC-based approach. Experiments on artificial and real data show that VEM-JDE is robust to model misspecification and provides computational gain while maintaining good performance in terms of activation detection and hemodynamic shape recovery.

Variational solution to the joint detection estimation of brain activity in fMRI.

We address the issue of jointly detecting brain activity and estimating underlying brain hemodynamics from functional MRI data. We adopt the so-called Joint Detection Estimation (JDE) framework that takes spatial dependencies between voxels into account. We recast the JDE into a missing data framework and derive a Variational Expectation-Maximization (VEM) algorithm for its inference. It follows a new algorithm that has interesting advantages over the previously used intensive simulation methods (Markov Chain Monte Carlo, MCMC): tests on artificial data show that the VEM-JDE is more robust to model mis-specification while additional tests on real data confirm that it achieves similar performance in much less computation time.

A wavelet-based regularized reconstruction algorithm for SENSE parallel MRI with applications to neuroimaging.

To reduce scanning time and/or improve spatial/temporal resolution in some Magnetic Resonance Imaging (MRI) applications, parallel MRI acquisition techniques with multiple coils acquisition have emerged since the early 1990s as powerful imaging methods that allow a faster acquisition process. In these techniques, the full FOV image has to be reconstructed from the resulting acquired undersampled k-space data. To this end, several reconstruction techniques have been proposed such as the widely-used SENSitivity Encoding (SENSE) method. However, the reconstructed image generally presents artifacts when perturbations occur in both the measured data and the estimated coil sensitivity profiles. In this paper, we aim at achieving accurate image reconstruction under degraded experimental conditions (low magnetic field and high reduction factor), in which neither the SENSE method nor the Tikhonov regularization in the image domain give convincing results. To this end, we present a novel method for SENSE-based reconstruction which proceeds with regularization in the complex wavelet domain by promoting sparsity. The proposed approach relies on a fast algorithm that enables the minimization of regularized non-differentiable criteria including more general penalties than a classical ℓ(1) term. To further enhance the reconstructed image quality, local convex constraints are added to the regularization process. In vivo human brain experiments carried out on Gradient-Echo (GRE) anatomical and Echo Planar Imaging (EPI) functional MRI data at 1.5T indicate that our algorithm provides reconstructed images with reduced artifacts for high reduction factors.