Graph Neural Networks (GNNs) have garnered widespread recognition in the identification of Autism Spectrum Disorder (ASD) owing to their remarkable adaptability to irregular patterns of Functional Brain Networks (FBNs). However, current methods for constructing FBNs generally employ a uniform modeling strategy to process neuroimaging data from different subjects, which fail to consider the heterogeneity of functional connectivity patterns among individuals adequately. In addition, existing methods tend to excessively focus on directly connected brain Regions of Interest (ROIs) when analyzing brain networks, underestimat\ing the importance of indirectly connected brain ROIs. At the same time, conventional approaches for identifying crucial brain regions may miss vital regions due to rigid threshold constraints. To address these issues, we propose Personalized Structure Preservation based GNN (PSP-GNN) for ASD diagnosis, which incorporates three aspects: 1) A personalized structure preservation strategy that constructs individualized brain networks by accounting for subject-specific variations; 2) A connection interaction-aware module designed to characterize interactions between directly and indirectly connected brain regions, providing comprehensive brain network representations; 3) A flexible brain region refinement technique based on Bernoulli sampling, which identifies salient brain regions without relying on pre-defined thresholds. Experimental results demonstrate the effectiveness of PSP-GNN in ASD diagnosis, highlighting its potential as a robust tool for future ASD diagnosis applications that combine FBNs and GNNs. Notably, the critical brain regions identified by PSP-GNN are consistent with established medical knowledge, suggesting their utility as potential biomarkers for clinical ASD diagnosis.

Few-shot learning (FSL) through the training of few labeled samples in the source domain and fine-tuning in target domain has gotten increasing attention in hyperspectral images (HSI) classification. However, in current FSL for HSI classification (HSIC), the number of classes trained in the source domain feature extractor is contingent upon the aligned class number with task classes, which restricts the availability and generalization of the transferable knowledge learned in the source domain. In this article, we propose a few-shot learning with class-number non-aligned training and feature differential network for hyperspectral image classification. Firstly, a class-number non-aligned FSL training framework on multiple independent sources is established, where each source trains respective classes, eliminating the need for aligning the number of classes with the target's. Secondly, in order to learn the feature brought by different domains at different scales, an attention-guided cross-scale feature differential network is constructed to obtain feature differentials between neighboring layers through a feature differential unit (FDU), which extracts detailed pixel's differential information and facilitates to exploit feature variations at different scales. Furthermore, to alleviate the training burden generated by multiple source domains from the non-aligned training strategy, a hybrid loss function is devised to augment the inter-class distance while reducing the intra-class distance. Experiments conducted on three public hyperspectral datasets demonstrate that the proposed model outperforms existing FSL methods for Hyperspectral image classification.

Individuals with Attention Deficit Hyperactivity Disorder (ADHD) exhibit intricate structural and functional interconnectivity across multiple brain regions. These patients demonstrate abnormal alterations in both respective modal brain regions and co-occurrent brain regions. Furthermore, there exist multi-level relationships between these abnormal brain structures and functions, encompassing hierarchical interactions between function-structural alterations as well as hierarchical progression from local regions to broader brain networks. However, most existing multi-modal ADHD classification approaches independently embed functional and structural data into separate spaces for information integration, often predominately focusing on uni-modal features. This approaches lead to a significant loss of features related to functiona-structural interaction relationships. Additionally, it is crucial for ADHD classification to accurately identify both uni-modal and co-occurrent abnormal alterations in brain regions which have hierarchical progression relationships. This study proposes a function-structural interaction multi-modal network with progressive and multi-level feature fusion (FSIPM) for ADHD classification. The main contributions are threefold: 1) An innovative function-structural interaction method is proposed to facilitate the mutual regulation of information across modalities, thereby relieving modal feature bias caused by integrated fusion. 2) A multi-level refinement framework is designed to promote the identification of both individual and co-occurrent abnormal brain regions. This progressive approach models the function-structural alterations of abnormal brain regions and the hierarchical relationships from local to brain networks, ensuring a deeper understanding of brain abnormalities. 3) Multi-level feature fusion aims to minimize the loss of details caused by consecutive sampling operations during the progressive process of the network, contributing to a more accurate and nuanced representation of ADHD-related brain alterations. Experimental results on the ADHD-200 and ABIDE I datasets demonstrate that FSIPM achieves competitive performance in ADHD classification while revealing uni-modal and co-occurrent altered brain regions that are consistent with clinical findings.

Hyperspectral images (HSI) have spectral variability, which leads to spectral dependence in adjacent and non-adjacent regions, and this dependence is essential for the classification of regions with mixed pixels. Current neural architecture search (NAS) methods have achieved significant advantages in HSI classification, but these methods cannot capture spectral dependence in non-adjacent regions because only use feedforward connections. Meanwhile, the cost of the search process in NAS is proportional to the scale of the search space, which limits the expansion of the search space. To address these issues, we propose a sparse-coding-inspired feedback neural architecture search (SCIF-NAS) method for HSI classification. Firstly, we view HSI samples as sequences and introduce a feedback mechanism in NAS to model the spectral dependence of non-adjacent regions to mitigate the effects of spectral variation. Secondly, we design several feedforward operations according to the characteristics of HSI, to form the search space together with feedback operations. Meanwhile, a sparse-coding-inspired NAS accelerated strategy is introduced to alleviate the search time burden caused by the expansion of search space. Thirdly, we integrate center loss with cross-entropy loss to construct a hybrid loss function that helps to obtain a better classification boundary. Finally, we conduct experiments on three popular HSI benchmarks, which show that SCIF-NAS outperforms the state-of-the-art methods in HSI classification.

Few-shot learning (FSL) has achieved promising performance in hyperspectral image classification (HSIC) with few labeled samples by designing a proper embedding feature extractor. However, the performance of embedding feature extractors relies on the design of efficient deep convolutional neural network architectures, which heavily depends on the expertise knowledge. Particularly, FSL requires extracting discriminative features effectively across different domains, which makes the construction even more challenging. In this paper, we propose a novel neural architecture search-based FSL model for HSI classification, called HCFSL-NAS. Three novel strategies are proposed in this work. First, a neural architecture search-based embedding feature extractor is developed to the FSL in HSIC, whose search space includes a group of proposed multi-scale convolutions with channel attention. Second, a multi-source learning framework is employed to aggregate abundant heterogeneous and homogeneous source data, which enables the powerful generalization of network to the HSIC with only few labeled samples. Finally, the pointwise-based cross-entropy loss and the pairwise-based adaptive sparse loss are jointly optimized to maximize inter-class distance and minimize the distance within a class simultaneously. Experimental results on four publicly hyperspectral data sets demonstrate that HCFSL-NAS outperforms both the exiting FSL methods and supervised learning methods for HSI classification with only few labeled samples. Code is available at: https://github.com/xh-captain/HCFSL-NAS.

Modeling functional brain networks (FBNs) for attention deficit hyperactivity disorder (ADHD) has sparked significant interest since the abnormal functional connectivity is discovered in certain functional magnetic resonance imaging (fMRI)-based brain regions compared to typical developmental control (TC) individuals. However, existing models for modeling FBNs generally use dimensionality reduction techniques to process the high dimensional input data, which results in confusion and an inaccurate representation of voxel interactions between spatially close brain regions, causing misdiagnosis of the disease. To address these issues, we propose a spatial preservation-based neural architecture search (SP-NAS) for FBNs modeling in ADHD. The main work includes three-fold: 1) A spatial preservation module is designed to embed original spatial information into dimensionality reduction data, addressing the challenge of a large number of parameters in the original data and mitigating disease misdiagnosis resulting from voxel confusion between different brain regions caused by dimensionality reduction. 2) A search space using more suitable search operations is constructed to efficiently extract spatial-temporal interaction characteristics of fMRI data in ADHD while narrowing the search space. 3) Cross-regional association differences between ADHD and TC groups are explored for ADHD auxiliary diagnosis since the abnormal activation regions of ADHD relative to TC on the brain regions and the abnormal connectivity between the lesion brain regions are identified. Model validation results on the ADHD-200 dataset show that the FBNs obtained from SP-NAS not only achieve competitive results in ADHD diagnosis but also reveal abnormal connections in the lesion regions of ADHD consistent with clinical diagnosis.

Increasing attention has been paid to attention-deficit hyperactivity disorder (ADHD)-assisted diagnosis using functional brain networks (FBNs) since FBNs-based ADHD diagnosis can not only extract the functional connectivities from FBNs as potential biomarkers for brain disease classification, but also identify the focal regions of disease. Therefore, modeling FBNs has become a key topic for ADHD diagnosis via resting state functional magnetic resonance imaging (rfMRI). However, the dominant models either ignore the strong regional correlation between adjacent time series or fail to capture the long-distance dependency (LDD) in imaging series. To address the issues, we propose a multi-head attention-based region-enhancement model (MAREM) for ADHD classification. Firstly, a multi-head attention mechanism with region-enhancement is designed to represent the FBNs, where region-enhancement module are designed to process strong regional correlation between adjacent time series. Secondly, multi-head attention is used to map the region information of each time point into different subspaces for establishing global dependencies in imaging series. Thirdly, the proposed model is applied to the ADHD-200 dataset for classification. The results show the proposed model’s out-performance of the state-of-the-art in both classification accuracy and generalization ability. Furthermore, we identify several brain networks that have been considered to be associated with ADHD in clinical studies.

Spatio-temporal modelling based on resting-state functional magnetic resonance imaging (rsfMRI) of ADHD has been a major concern in the neuroimaging community, given the differences in the role of brain regions between attention deficit hyperactivity disorder (ADHD) patients versus typical developmental control group (TC). Several spatio-temporal deep learning models are proposed for rsfMRI, however, due to the high dimensionality and few samples of brain data, most models use dimension-reduced data as input for modelling, which suffer from the loss of original spatial relationships in the brain data. Although Recurrent Neural Network (RNN) and Attention mechanism (Attention) proposed in recent years can extract local correlations and long-distance dependency (LDD), the spatio-temporal relationships they rely on have lost their original high-dimensional spatial relevance. Therefore, a spatial preservation-based autoencoder for modelling ADHD functional brain networks (FBNs) is proposed by embedding the spatial information and combining both RNN and Transformer to address the issue that the dimension-reduced data cannot preserve the original high-dimensional spatial correlations. Firstly, a spatial preservation module is designed to fill the gap between the original data and the dimension-reduced data. Secondly, the dimension reduction module and feature extraction module are designed to improve the representation of spatio-temporal correlations. Thirdly, the extracted FBNs are applied to the disease classification on the ADHD-200 dataset, which show the model’s effectiveness in classifying ADHD compared with the state-of-the-art methods. Finally, we investigate the differences in regional correlations between ADHD and TC.

Hyperspectral image (HSI) classification based on neural architecture search (NAS) is a currently attractive frontier as it not only automatically searches complex neural network architecture, but also avoids professional knowledge and experience design, and alleviates the lacking of generalization ability as well when dealing with a new classification task. However, the existing HSI classification based on NAS has some drawbacks: 1) A huge number of training parameters and high calculations are inductive to over-fitting and high complexity. 2) Efficient operators are lacking in the search space which can distinguish spatial locations and spectral features in different bands. Furthermore, as the category samples in HSI data show a serious long-tail distribution phenomenon, HSI classification remains challenging. To address these issues, we propose a lightweight HSI classification model LMSS-NAS integrating multi-scale spectral-spatial attention. The main work includes three-fold: 1) In order to reduce the number of model parameters and promote spectral-spatial feature fusion, a new lightweight efficient search space is designed, which consists of three equivalent lightweight convolution operators with multiple receptive fields. 2) To fully use the spectral-spatial correlation of HSI, a cube-to-pixel classification framework is designed to mine the local spatial and spectral context. 3) Focal loss and label smoothing loss in computer vision tasks are jointly migrated to LMSS-NAS to improve the unbalanced samples’ classification and model robustness. Experimental results on four public hyperspectral data sets show that the proposed method can achieve competitive classification performance as well as low computational cost. Code is available at: https://github.com/xh-captain/LMSS-NAS.

The compressed sensing (CS) method can reconstruct images with a small amount of under-sampling data, which is an effective method for fast magnetic resonance imaging (MRI). As the traditional optimization-based models for MRI suffered from non-adaptive sampling and shallow” representation ability, they were unable to characterize the rich patterns in MRI data. In this paper, we propose a CS MRI method based on iterative shrinkage threshold algorithm (ISTA) and adaptive sparse sampling, called DSLS-ISTA-Net. Corresponding to the sampling and reconstruction of the CS method, the network framework includes two folders: the sampling sub-network and the improved ISTA reconstruction sub-network which are coordinated with each other through end-to-end training in an unsupervised way. The sampling sub-network and ISTA reconstruction sub-network are responsible for the implementation of adaptive sparse sampling and deep sparse representation respectively. In the testing phase, we investigate different modules and parameters in the network structure, and perform extensive experiments on MR images at different sampling rates to obtain the optimal network. Due to the combination of the advantages of the model-based method and the deep learning-based method in this method, and taking both adaptive sampling and deep sparse representation into account, the proposed networks significantly improve the reconstruction performance compared to the art-of-state CS-MRI approaches.

The spatiotemporal representation of task-based brain functional networks is a key topic in functional magnetic resonance image (fMRI) research. At present, deep learning has been more powerful and flexible in brain functional network research than traditional methods. However, the dominant deep learning models failed in capturing the long-distance dependency (LDD) in task-based fMRI images (tfMRI) due to the time correlation among different task stimuli, the nature between temporal and spatial dimensions, which resulting in inaccurate brain pattern extraction. To address this issue, this paper proposes a spatiotemporal attention neural architecture search (NAS) model for task-based brain functional networks representation (SA-NAS-BFNR), where attention mechanism and gate recurrent unit (GRU) are integrated into a novel framework and GRU structure is searched by the differentiable neural architecture search. This model can not only achieve meaningful brain functional networks (BFNs) by addressing the LDD, but also simplify the existing recurrent structure models in tfMRI. Experiments show that the proposed model is capable of improving the fitting ability between time series and task stimulus sequence, and extracting the BFNs effectively as well.

Compressive sensing magnetic resonance (MR) imaging is aimed at achieving high‐quality MR image reconstruction by undersampling K‐space data. It is crucial to explore prior information since compressive sensing MR imaging relies heavily on some prior assumptions, such as signal's sparse property. In this study, in order to explore the prior information fully, an improved MR image reconstruction model based on compressive sensing theory is proposed, named reference image MR imaging with adaptive tight frame. In the proposed model, an adaptive tight frame is involved to explore the sparse prior information adapt to MR images and the similarity prior information to the target image. Meanwhile, improved adaptive weighting parameters are used to trade off the sparsity between the regions with much similarity and that of little similarity. In addition, the smoothing‐based fast iterative shrinkage‐threshold algorithm is utilised to tackle the optimisation problem so as to speed up imaging. The experimental results demonstrate that the proposed MR image reconstruction method outperforms some state‐of‐the‐art methods in terms of quantitative results.

Compressed sensing with sparse frame representation has received much greater attention than the orthonormal bases for its practical application in signal processing. One can expect exact recovery from undersampled data via an ‐minimisation under some proper conditions imposed on the sensing/measurement matrix and sparse representation matrix. In this study, the authors first introduce a recovery condition named by B ‐ GRIP , which is actually a generalisation of the well‐known restricted isometry property that most of the compressed sensing problems depend on. Under this condition, they expand the performance analysis of compressed sensing problem by an ‐minimisation problem for via the optimal dual frame. Finally, they present an iterative algorithm concerning the restoration of compressed sensing signals based on ‐minimisation for as well as the convergence of the algorithm, and prove that the iterative sequence can not only approximate the original signal, but be the exact one.

Salient object detection is a fundamental problem and has been received a great deal of attention in computer vision. Recently, deep learning model became a powerful tool for image feature extraction. In this study, the authors propose a multi‐scale deep neural network (MSDNN) for salient object detection. The proposed model first extracts global high‐level features and context information over the whole source image with the recurrent convolutional neural network. Then several stacked deconvolutional layers are adopted to get the multi‐scale feature representation and obtain a series of saliency maps. Finally, the authors investigate a fusion convolution module to build a final pixel level saliency map. The proposed model is extensively evaluated on six salient object detection benchmark datasets. Results show that the authors’ deep model significantly outperforms other 12 state‐of‐the‐art approaches.

Geometric constraint problem can be transformed to an optimization problem. We can solve the problem with niche improved particle swarm. Classical particle swarm optimization is likely to be trapped into local minima as well as premature. A niche improved particle swarm optimization (NIPSO) based on niche theory was developed. After the update of the particle velocity and position, the outlier particle was identified in the NIPSO by comparing the niche number of every particle, with which the crossover and selection operators were employed sequent for those particles, whose personal best values were less than that of the outlier particle. The experiment shows that it can improve the geometric constraint solving efficiency and possess better convergence property than the compared algorithms.