3D spatially-adaptive canonical correlation analysis: Local and global methods

Temporal spatial differences observed by functional MRI and human intraoperative optical imaging

Blind source separation (BSS) is commonly used in functional magnetic resonance imaging (fMRI) data analysis. Recently, BSS models based on restricted Boltzmann machine (RBM), one of the building blocks of deep learning models, have been shown to improve brain network identification compared to conventional single matrix factorization models such as independent component analysis (ICA). These models, however, trained RBM on fMRI volumes, and are hence challenged by model complexity and limited training set. In this article, we propose to apply RBM to fMRI time courses instead of volumes for BSS. The proposed method not only interprets fMRI time courses explicitly to take advantages of deep learning models in latent feature learning but also substantially reduces model complexity and increases the scale of training set to improve training efficiency. Our experimental results based on Human Connectome Project (HCP) datasets demonstrated the superiority of the proposed method over ICA and the one that applied RBM to fMRI volumes in identifying task-related components, resulted in more accurate and specific representations of task-related activations. Moreover, our method separated out components representing intermixed effects between task events, which could reflect inherent interactions among functionally connected brain regions. Our study demonstrates the value of RBM in mining complex structures embedded in large-scale fMRI data and its potential as a building block for deeper models in fMRI data analysis.

Previous studies have shown that hemodynamic response overlap severely limits the maximum presentation rate with event-related functional MRI (fMRI) using fixed intertrial experimental designs. Here we demonstrate that the use of randomized experimental designs can largely overcome this limitation, thereby allowing for event-related fMRI experiments with extremely rapid presentation rates. In the first experiment, fMRI time courses were simulated using a fixed intertrial interval design with intervals of 16, 3, and 1 s, and using a randomized design having the same mean intertrial intervals. We found that using fixed intertrial interval designs the transient information decreased with decreasing intertrial intervals, whereas using randomized designs the transient information increased with decreasing mean intertrial intervals. In a second experiment, fMRI data were collected from two subjects using a randomized paradigm with visual hemifield stimuli presented randomly every 500 ms. Robust event-related activation maps and hemodynamic response estimates were obtained. These results demonstrate the feasibility of performing event-related fMRI experiments with rapid, randomized paradigms identical to those used in electrophysiological and behavioral studies, thereby expanding the applicability of event-related fMRI to a whole new range of cognitive neurosciences questions and paradigms.

Functional MRI evidence for disparate developmental processes underlying intelligence in boys and girls

In this paper, we present a grid-based framework for image retrieval. In order to represent the intricate composition of images, the grid-based approach partitions each image into blocks from which a feature representation is derived from the local low-level content. Since the background often dominates the subject in the foreground, a special query selection method was developed. It combines the salient region-of-interest/query-by-example paradigm with coarse segmentation to remove the irrelevant background regions. The proposed search method looks for similar features across all block positions and at several scales. Existing local grid-based methods are constrained by searching for objects in the same position as the query object. Using this framework, the spatial constraint can be eliminated, and steps toward scale invariance can be taken. Promising results show that the grid-based method performs better than global search.

Inertial odometry is a typical localization method that is widely and easily accessible in many devices. Pedestrian positioning can benefit from this approach based on inertial measurement unit (IMU) values embedded in smartphones. Fitting the inertial odometry outputs, namely step length and step heading of a human for instance, with spatial information is an ubiquitous way to correct for the cumulative noises. This so-called map-matching process can be achieved in several ways. In this paper, a novel real-time map-matching approach was developed, using a backtracking particle filter that benefits from the implemented geospatial analysis, which reduces the complexity of spatial queries and provides flexibility in the use of different kinds of spatial constraints. The goal was to generalize the algorithm to permit the use of any kind of odometry data calculated by different sensors and approaches as the input. Further research, development, and comparisons have been done by the easy implementation of different spatial constraints and use cases due to the modular structure. Additionally, a simple map-based optimization using transition areas between floors has been developed. The developed algorithm could achieve accuracies of up to 3 m at approximately the 90th percentile for two different experiments in a complex building structure.

The reliable localization for capsule endoscopes is an open research topic. In this study, a low-frequency magnetic localization method for capsule endoscopes with an integrated active coil is proposed. The spatial constraint, the limited battery capacity and the ferromagnetic battery shell of commercial capsules were considered. The generated magnetic flux density was evaluated depending on the distance to the coil and the maximal detectable range was determined. Twelve sensors were arranged in rings and by comparing the measured magnetic flux density with the analytic dipole model, the position and orientation of the coil were reconstructed. The results revealed that the ferromagnetic shell increases the magnetic moment of the coil by approximately a factor of 2.4. Moreover, the mean position and orientation errors were 0.5 mm and 0.3°. Furthermore, by using only the three closest sensors to the coil, a similar localization performance was achieved. Therefore, it was concluded that it is a feasible approach to choose the sensors, which measure the strongest signal to address the problem of the maximal detectable range of magnetic sensors. Moreover, by considering the limited battery capacity, the localization with the proposed coil must be conducted in short time intervals instead of continuously.

This study proposes a new hybrid algorithm for solving small to large-scale continuous global optimization problems. It comprises evolutionary computation algorithm featuring a novel adaptive elitism strategy and a sequential quadratic programming algorithm; combined in a collaborative portfolio with a validation procedure. The sequential quadratic programming is a gradient based local search method designed to derive effective search directions by using exact Hessians obtained via a vectorized forward accumulation of derivatives technique. The proposed hybrid design aim was to ensure that the two algorithms complement each other by effectively exploring and exploiting the problem search space. Experimental results justify that an adept hybridization of evolutionary algorithms with a suitable local search method could yield a robust and efficient means of solving wide range of global optimization problems.

Imaging brain function with simultaneous BOLD and viscoelasticity contrast: fMRI/fMRE

Since it has become possible to acquire EEG und fMRI data simultaneously (EEG-correlated fMRI, EEG/fMRI), this technique has been applied mainly to four fields: 1) mapping of fMRI activation associated with interictal epileptiform discharges (IED) in patients with epilepsy, 2) mapping of fMRI activation associated with physiological EEG activity, like alpha rhythm; 3) monitoring vigilance and sleep stages during fMRI experiments; and 4) simultaneous acquisition of fMRI data and evoked potentials. Here, EEG/fMRI studies of IED and alpha- and beta-rhythm are presented. Several studies have shown that EEG-correlated fMRI is a practicable method to be applied to patients with epilepsy showing frequent IED on scalp EEG. However, the clinical interpretation of the fMRI maps remains difficult; mainly because EEG/fMRI results have not been systematically validated with a gold standard, which are intracranial recordings and outcome after epilepsy surgery. Currently, EEG-correlated fMRI has to be considered as a research tool providing insights to the pathophysiological processes underlying epileptic disorders and to the effects of epileptiform activity on cognition. It remains unclear if the method will also be used for the routine clinical work-up of epilepsy patients in the future. EEG/fMRI studies of alpha- and beta-rhythm in normal controls have shown that spontaneous fluctuations of the band power during rest are associated with fMRI signal changes, possibly representing alterations of attention-modulating networks (retrosplenial and fronto-parietal cortices). Similar fMRI activation patterns are found during generalized and focal IED in epileptic patients and might represent an fMRI signature of the cognitive effects of IED.

Time-resolved fMRI seeks to elucidate neuronal activity during a single execution of a mental task, which corresponds typically to a timescale of seconds. However, this is also the timescale of the hemodynamic response, which delays and blurs the signal in time. In order to distinguish the temporal characteristics of the neuronal activity from that of the hemodynamic response, which is often vaguely known, we recorded a set of fMRI time courses under conditions of a varying behavioral parameter, and correlated this parameter to the width of the fMRI response. For the task under investigation, the mental rotation of three-dimensional objects, we found that the activation in the parietal lobe is related to an aspect of the task that is described by the reaction time (for example, the very act of mental rotation), and not only to aspects of the task that are constant from trial to trial, such as the visual presentation at the beginning or the decision at the end of the task.

Event Abstract Back to Event A combination model of ICA and sparsity prior with respect to fMRI signal analysis Nizhuan Wang1* and Weiming Zeng1 1 Shanghai maritime University, College of Information Engineering, China The blood oxygen level dependent (BOLD) functional magnetic resonance imaging (fMRI) data effective analysis is a challenging task because of the complexity of neuron activity in human brain. Independent component analysis (ICA) has been widely used to investigate the functional connectivity of fMRI data, which assumes that the sources of functional regions are independent statistically. However, in the method, the spatial sparsity feature has usually been ignored, which is intrinsic property of the fMRI sources. In 2001, Alexander et al. demonstrated that the optimal sparse representation could significantly benefit the de-noising of fMRI time courses [1]. Recently, Daubechies et al. also demonstrated that the sparsity was more general, intuitive and promising assumption for BOLD fMRI signals [2], whose sources were highly clustering and centering located. In this study, a combination model of FastICA and sparsity prior with respect to fMRI signal analysis, named SFICA, is presented. In this model, the sparse decomposition is performed on the fMRI signal using the wavelet packet decomposition, which yields wavelet tree nodes with different degree of sparsity [3]. Based on these wavelet tree nodes, a common fuzzy C-means clustering is applied to form the optimal sparse representation set of fMRI signal, which can be as an input of FastICA. Compared with FastICA, hybrid data experiment in our study demonstrated this combination model had better spatial source recovery performance on the ground of receiver operating characteristic (ROC) analysis (shown in Fig.1A). For task-related and resting state experimental tests, according to the skewness analysis and correlation variance analysis, almost all the functional networks discovered by this combination model were more symmetric with less skewness values in spatial domain and more consistent in temporal domain than the ones by FastICA. For simplicity, the partial experimental results corresponding to task-related experimental tests were showed in the Fig.1B. Fig. 1: (A) Hybrid data experimental results: ROC curves of simulated signals S1 and S2 corresponding to anterior regions and the more posterior regions respectively, where the z-score ranges from 0.5 to 2.0; (B) Task-related data experimental results: the visual networks of one subject detected by the FastICA and SFICA methods corresponding to the task-related visual stimulus. The corresponding skewness values are labeled and the spatial differences of the visual networks corresponding to the two method are marked by circle shapes with distinct colors in this figure. Figure 1 Acknowledgements Research supported by the National Natural Science Foundation of China (Grant No.31170952), the Innovation Program of Shanghai Municipal Education Commission (Grant No.11ZZ143) , the Program for Professor of Special Appointment (Eastern Scholar) at Shanghai Institutions of Higher Learning and the Program for Graduate Special Endowment Fund for Innovative developing of Shanghai Maritime University (Grant No. yc2011021).

In this paper we construct an atlas that captures functional characteristics of a cognitive process from a population of individuals. The functional connectivity is encoded in a low-dimensional embedding space derived from a diffusion process on a graph that represents correlations of fMRI time courses. The atlas is represented by a common prior distribution for the embedded fMRI signals of all subjects. The atlas is not directly coupled to the anatomical space, and can represent functional networks that are variable in their spatial distribution. We derive an algorithm for fitting this generative model to the observed data in a population. Our results in a language fMRI study demonstrate that the method identifies coherent and functionally equivalent regions across subjects.

Detection of signal synchronizations in resting-state fMRI datasets

It is a common practice to evaluate the reproducibility of fMRI at the group level. However, for clinical applications of fMRI, where the focus is on reproducibility of single individuals, the high test-retest reliability that is sometimes reported for group-based measures can be misleading. On the level of single subjects, reproducibility of fMRI is still far too low for clinical applications, not even meeting the standards to use fMRI for scientific purposes. The goal of this work is to enhance the poor single-subject time course reproducibility of fMRI. For this purpose, we have developed a framework for post-processing fMRI signals using Savitzky-Golay (SG) filters in conjunction with general linear model (GLM) based data cleaning. The parameters of these filters were trained to be the optimal ones based on a dataset of working memory relevant signals. By employing our data-driven filtering framework, we successfully improve the average reproducibility correlation of a single fMRI time course from r = 0.26 (as obtained with a conventional statistical parametric mapping (SPM) data cleaning pipeline) to a fair level of r = 0.41. Additionally, we are able to enhance the average connectivity correlation from r = 0.44 to r = 0.54. Our conclusion is that signal post-processing with a data-driven SG filter framework may substantially improve time course reproducibility compared to conventional denoising pipelines. As a conservative estimate, we conjecture that roughly 10-30% of the population may benefit from optimized fMRI pipelines in a clinical setting depending on the measure of interest while this number was nihil for conventional fMRI pipelines.