Characterizing brain dynamic functional connectivity (dFC) patterns from functional Magnetic Resonance Imaging (fMRI) data is of paramount importance in imaging neuroscience and medicine. Recently, graph neural network (GNN) models, combined with transformers or recurrent neural networks (RNNs), have shown great potential for modeling the dFC patterns. However, these methods face challenges in characterizing the modularity organization of brain networks and capturing varying dFC state patterns. To address these limitations, we propose dFCExpert, a novel method designed to learn robust representations of dFC patterns from fMRI data with modularity experts and state experts. Specifically, the modularity experts optimize multiple experts to characterize the brain modularity organization during graph feature learning process by combining GNN and mixture of experts (MoE), with each expert focusing on brain network nodes within the same functional network module. The state experts aggregate temporal dFC features into a set of distinct connectivity states using a soft prototype clustering method, providing insight into how these states support diverse brain functions and vary across brain conditions. Experiments on three large-scale fMRI datasets have demonstrated the superiority of our method over existing alternatives. The learned dFC representations not only enhance interpretability but also hold promise for advancing our understanding of brain function across a range of conditions, including brain development, sex differences, and Autism Spectrum Disorder. Our implementation is publicly available at https://github.com/MLDataAnalytics/dFCExperts.

To achieve fast and accurate cortical surface reconstruction from brain magnetic resonance images (MRIs), we develop a method to jointly reconstruct the inner (white-gray matter interface), outer (pial), and midthickness surfaces, regularized by their interdependence. Rather than reconstructing these surfaces separately without taking into consideration their interdependence as in most existing methods, our method learns three diffeomorphic deformations jointly to optimize the midthickness surface to lie halfway between the inner and outer cortical surfaces and simultaneously deforms it inward and outward towards the inner and outer cortical surfaces, respectively. The surfaces are encouraged to have a spherical topology by regularization terms for non-negativeness of the cortical thickness and symmetric cycle-consistency of the coupled surface deformations. The coupled reconstruction of cortical surfaces also facilitates an accurate estimation of the cortical thickness based on the diffeomorphic deformation trajectory of each vertex on the surfaces. Validation experiments have demonstrated that our method achieves state-of-the-art cortical surface reconstruction performance in terms of accuracy and surface topological correctness on large-scale MRI datasets, including ADNI, HCP, and OASIS. The code is available at: https://github.com/MLDataAnalytics/SurfNet.

A versatile medical image segmentation model applicable to images acquired with diverse equipment and protocols can facilitate model deployment and maintenance. However, building such a model typically demands a large, diverse, and fully annotated dataset, which is challenging to obtain due to the labor-intensive nature of data curation. To address this challenge, we propose a cost-effective alternative that harnesses multi-source data with only partial or sparse segmentation labels for training, substantially reducing the cost of developing a versatile model. We devise strategies for model self-disambiguation, prior knowledge incorporation, and imbalance mitigation to tackle challenges associated with inconsistently labeled multi-source data, including label ambiguity and modality, dataset, and class imbalances. Experimental results on a multi-modal dataset compiled from eight different sources for abdominal structure segmentation have demonstrated the effectiveness and superior performance of our method compared to state-of-the-art alternative approaches. We anticipate that its cost-saving features, which optimize the utilization of existing annotated data and reduce annotation efforts for new data, will have a significant impact in the field.

Accurate reconstruction of cortical surfaces from brain magnetic resonance images (MRIs) remains a challenging task due to the notorious partial volume effect in brain MRIs and the cerebral cortex's thin and highly folded patterns. Although many promising deep learning-based cortical surface reconstruction methods have been developed, they typically fail to model the interdependence between inner (white matter) and outer (pial) cortical surfaces, which can help generate cortical surfaces with spherical topology. To robustly reconstruct the cortical surfaces with topological correctness, we develop a new deep learning framework to jointly reconstruct the inner, outer, and their in-between (midthickness) surfaces and estimate cortical thickness directly from 3D MRIs. Our method first estimates the midthickness surface and then learns three diffeomorphic flows jointly to optimize the midthickness surface and deform it inward and outward to the inner and outer cortical surfaces respectively, regularized by topological correctness. Our method also outputs a cortex thickness value for each surface vertex, estimated from its diffeomorphic deformation trajectory. Our method has been evaluated on two large-scale neuroimaging datasets, including ADNI and OASIS, achieving state-of-the-art cortical surface reconstruction performance in terms of accuracy, surface regularity, and computation efficiency.

In contrast to most existing studies that typically characterize the developmental sex differences using analysis of variance or equivalently multiple linear regression, we present a parameter-free centralized multi-task learning method to identify sex specific and common resting state functional connectivity (RSFC) patterns underlying the brain development based on resting state functional MRI (rs-fMRI) data. Specifically, we design a novel multi-task learning model to characterize sex specific and common RSFC patterns in an age prediction framework by regarding the age prediction for males and females as separate tasks. Moreover, the importance of each task and the balance of these two patterns, respectively, are automatically learned in order to make the multi-task learning robust as well as free of tunable parameters, i.e., parameter-free for short. Our experimental results on synthetic datasets verified the effectiveness of our method with respect to prediction performance, and experimental results on rs-fMRI scans of 1041 subjects (651 males) of the Philadelphia Neurodevelopmental Cohort (PNC) showed that our method could improve the age prediction on average by 5.82% with statistical significance than the best alternative methods under comparison, in addition to characterizing the developmental sex differences in RSFC patterns.

Functional brain network analysis has been a powerful tool for measuring brain function in normal and pathologic states based on resting state fMRI (rsfMRI) data. Recent advances in pattern recognition and sparse modeling have enabled us to characterize subject-specific functional brain networks and derive clinically useful biomarkers. In this paper, we briefly introduce our recent work in the development of functional brain network analytic techniques, including functional brain network modeling, pattern recognition of functional brain networks, as well as modeling heterogeneous patterns of functional connectivity. Finally, we discuss some current challenges that have received and are likely to receive more attention in the near future.

Frequency offset and channel estimation in cooperative orthogonal frequency division multiplexing (OFDM) systems is studied in this paper. We consider the scenario of two or more source nodes sharing the same relay, i.e., corelay cooperative communications, and a new preamble, which is central-symmetric in time-domain, is proposed to perform the frequency offset and channel estimation. The non-zero samples in the proposed preamble are sparsely distributed with two neighboring non-zero samples being separated by μ >; 1 zeros. As long as μ >; 2L - 1 is satisfied, the multipath interference can be effectively eliminated, where L stands for the channel order. Unlike [1], the proposed preamble has a much lower Peak-to-Average Power Ratio (PAPR). The interference among the multiple source nodes can also be eliminated by using a backoff modulation scheme on the proposed preamble in each source node, and the mean-square error (MSE) of the proposed Least-Square (LS) channel estimator can be minimized by ensuring the orthogonality among the source nodes. The Pairwise Error Probability (PEP) performance of the proposed system by considering both the frequency offset and channel estimation errors is also derived in this paper. For a given Signalto-Noise-Ratio (SNR), by keeping the total power consumption to the source nodes and the relay to be constant, the PEP can be minimized by adjusting the ratio between the power allocated to the source nodes and the total power.

Cell search as well as synchronization in the 3rd Generation Partnership Project (3GPP) Long Term Evolution (LTE) system is performed in each User Equipment (UE) by using both the Primary Synchronization Signal (PSS) and Secondary Synchronization Signal (SSS), and the overall synchronization performance is dominated heavily by a robust PSS detection. Conventional non-coherent detector can achieve a reliable PSS detection based on the near-perfect auto-correlation and crosscorrelation properties of Zadoff-Chu (ZC) sequences [1], but at the cost of a relatively high computational complexity. This paper proposes two improved PSS detectors, i.e., Almost Half-Complexity (AHC) and Central Self-Correlation (CSC) detectors, by exploiting the central-symmetric property of ZC sequences. The AHC detector has exactly the same detection accuracy as that of the conventional detector but with 50% complexity being saved, and the proposed CSC detector can further reduce its complexity to 50% that of the AHC detector, however, at a cost of a slight accurate degradation. In order to mitigate the potential failure risk due to a large frequency offset in the proposed algorithms while at the same time keep the PSS detection accuracy un-degraded, an improvement of CSC, i.e., CSCIns, is also proposed. The performance of CSCIns detector is independent of the frequency offset, and numerical results show that the 90% PSS acquisition time of CSCIns is well within a 55ms duration with Signal-to-Noise Ratio (SNR) of -10 dB.

This paper presents a method for classification of structural brain magnetic resonance (MR) images, by using a combination of deformation-based morphometry and machine learning methods. A morphological representation of the anatomy of interest is first obtained using a high-dimensional mass-preserving template warping method, which results in tissue density maps that constitute local tissue volumetric measurements. Regions that display strong correlations between tissue volume and classification (clinical) variables are extracted using a watershed segmentation algorithm, taking into account the regional smoothness of the correlation map which is estimated by a cross-validation strategy to achieve robustness to outliers. A volume increment algorithm is then applied to these regions to extract regional volumetric features, from which a feature selection technique using support vector machine (SVM)-based criteria is used to select the most discriminative features, according to their effect on the upper bound of the leave-one-out generalization error. Finally, SVM-based classification is applied using the best set of features, and it is tested using a leave-one-out cross-validation strategy. The results on MR brain images of healthy controls and schizophrenia patients demonstrate not only high classification accuracy (91.8% for female subjects and 90.8% for male subjects), but also good stability with respect to the number of features selected and the size of SVM kernel used.

This article introduces a parallel genetic algorithm for 3D medical image analysis, which was applied in model based segmentation and multimodality image registration. The experimental results show that the proposed method is very encouraging and promising.

Active model-based segmentation has frequently been used in medical image processing with considerable success. Although the active model-based method was initially viewed as an optimization problem, most researchers implement it as a partial differential equation solution. The advantages and disadvantages of the active model-based method are distinct: speed and stability. To improve its performance, a parallel genetic algorithm-based active model method is proposed and applied to segment the lateral ventricles from magnetic resonance brain images. First, an objective function is defined. Then one instance surface was extracted using the finite-difference method-based active model and used to initialize the first generation of a parallel genetic algorithm. Finally, the parallel genetic algorithm is employed to refine the result. We demonstrate that the method successfully overcomes numerical instability and is capable of generating an accurate and robust anatomic descriptor for complex objects in the human brain, such as the lateral ventricles.