Spatial Relationships Research Articles

Frequent Alzheimer's disease neuropathological change in patients with glioblastoma.

The incidence of brain cancer and neurodegenerative diseases is increasing with a demographic shift towards aging populations. Biological parallels have been observed between glioblastoma and Alzheimer's disease (AD), which converge on accelerated brain aging. Here, we aimed to map the cooccurrence of AD neuropathological change (ADNC) in the tumor-adjacent cortex of patients with glioblastoma. Immunohistochemical screening of AD markers amyloid beta (Abeta), amyloid precursor protein (APP), and hyperphosphorylated tau (pTau) was conducted in 420 tumor samples of 205 patients. For each cortex area, we quantified ADNC, neurons, tumor cells, and microglia. Fifty-two percent of patients (N = 106/205) showed ADNC (Abeta and pTau, Abeta or pTau) in the tumor-adjacent cortex, with histological patterns widely consistent with AD. ADNC was positively correlated with patient age and varied spatially according to Thal phases and Braak stages. It decreased with increasing tumor cell infiltration (P < .0001) and was independent of frequent expression of APP in neuronal cell bodies (N = 182/205) and in tumor necrosis-related axonal spheroids (N = 195/205; P = .46). Microglia response was most present in tumor cell infiltration plus ADNC, being further modulated by patient age and sex. ADNC did not impact patient survival in the present cohort. Our findings highlight the frequent presence of ADNC in the glioblastoma vicinity, which was linked to patient age and tumor location. The cooccurrence of AD and glioblastoma seemed stochastic without clear spatial relation. ADNC did not impact patient survival in our cohort.

Tourism, Im/mobility and the University Collections in Norway, 1870–1914

This article explores the cultural collections at Det Kongelige Frederiks Universitet in Christiania (now Oslo) from the perspective of German-language tourists between the 1870s and 1914. It takes for its starting point the early history of tourism and museum collections as a story of the entanglement of the mobile and the immobile, of movement and stoppage. It argues that examining this entanglement provides insight into the complex processes of nation-building, which are formed in an interplay between tourist ascriptions and national self-images. Moreover, the museum objects’ spatial relations – the location and context in which they were exhibited – had a decisive impact on their perception and interpretation. Using guidebooks and travelogues as primary sources, the article discusses four of the most popular collections: the Viking ships, stave church portals, Sámi artefacts and objects brought back from the Gjøa expedition, in particular artefacts made by the Netsilingmiut/Nattilik.

A fast point cloud registration method based on spatial relations and features

Abstract Point cloud registration plays a crucial role in mobile robot localization, map building and three-dimensional (3D) model reconstruction. However, it remains challenged by issues such as compromised accuracy and sluggish efficiency, posing significant obstacles in achieving precise and timely alignments. Therefore, we propose a lightweight and fast point cloud registration method. Firstly, we mesh the 3D point cloud, compared with the traditional gridded point cloud method, it achieves initial point cloud registration by preserving the curvature characteristics of the internal point cloud, and utilizing the spatial relationship between grid cells and the quantitative relationship between the internal point cloud. Moreover, we adopt an iterative nearest point based on KD-Tree to realize the fine registration. So, our method does not necessitate intricate feature analysis and data training, and is resilient to similar transformations, non-uniform densities and noise. Finally, we conduct point cloud registration experiments using multiple publicly available point cloud datasets and compare them with several point cloud registration methods. The results demonstrate it is able to accomplish the point cloud registration quickly and exhibit high accuracy. More importantly, it maintains its efficacy and robustness even in the presence of noisy and defective point clouds.

Presence, absence, and spatial relations: An Interactional Ethnography of physical-virtual field-based learning through a sociomaterial lens

Field-based learning has been central to ecology education in supporting scientific inquiry and connecting learners with nature. However, pandemic-era campus closures have renewed debates regarding the role of virtual and immersive experiences. This Interactional Ethnography (IE) of a university-secondary school initiative in Hong Kong employs a sociomaterial lens to trace field-based learning interactions within and across physical and virtual spaces. Metaphors from social topology guided analysis of an ethnographic archive of classroom and field-based video records, learning artifacts, and interviews. Our analysis surfaces how and in what ways organisms act as agents in providing an essential anchor and recurring motif within and across interactions, contexts and settings in the enacted field-based design. While their ‘absence’ in the classroom and virtual environment constructs a sense of authenticity through the flickering metaphor of fire space, the material and semiotic resources in the field enable fluidity, opening possibilities for multiple, often serendipitous forms of relations, presences and dialogues that support diverse forms of knowledge and learning. By positioning field-based learning as varying sociomaterial assemblages centering on natural materials, we reconsider physical-virtual binaries and propose their designs as entangled human and non-human spatial relations that recognize and elevate the agency of natural materials.

Spatial imbalance in the digital era: how the digital economy affects urban sprawl in China

ABSTRACT Economic and social activities in the digital age have changed business forms and spatial relations, and identifying evolution of the urban form in the digital age is the key to sustainable urban governance. Based on data from prefecture-level cities in China from 2010 to 2021, this study explores the impact of the digital economy on urban sprawl and the path to its realization, thus providing policy implications for urban governance in the digital economy era. The empirical results show that the digital economy significantly promotes urban sprawl and that the urban sprawl effect caused by the digital economy has regional and scale heterogeneity. Western and medium-sized cities show greater sprawl trends under the influence of the digital economy. Digital economy development can significantly affect residents’ lifestyles and lead to urban sprawl, such as encouraging green, digital, and suburban lifestyles. In addition, the digital economy significantly affects the demand for urban land space, such as the industrial division of labour patterns, urbanization processes, and industrial virtual agglomeration, which are indirect factors leading to urban sprawl.

Disentangling soil-based ecosystem services synergies, trade-offs, multifunctionality, and bundles: A case study at regional scale (NE Italy) to support environmental planning

The explicit use of ecosystem services (ESs) assessments has been called as a way to guide environmental decision making, yet the promise of the ES approach lies behind its potential. A way to consolidate the approach could be to introduce some aspects into the ESs assessments which might have been neglected so far. Such aspects are mainly: (1) a focus on the complex ESs relations (such as synergies and trade-offs) that can impact the supply of multiple SESs (soil ecosystem services), and (2) focus on potential drivers of SESs relations. We applied bivariate and multivariate approaches to SESs indicators derived from a solid pedological knowledge of the Emilia-Romagna study area in NE Italy. We focused on 7 SES: (1) habitat for soil organisms, (2) filtering and buffering capacity, (3) contribution to microclimate regulation, (4) carbon sequestration, (5) food provision potential, (6) water regulation, and (7) water storage capacity. These SESs were estimated through a combination of point observations, and pedotransfer functions (PTF) estimates spatialised over the area of interest with geostatistical simulation techniques. We found that SESs bivariate spatial relations could be categorised mainly in three types of patterns at regional scale, either: (1) synergistic SESs relations dominating at the region level, (2) trade-offs dominating, or (3) both kind of relations more or less equally frequent. Interestingly, in some cases the dominant regional SESs relation switched at a local level, and such switch was driven by soil properties. For the multivariate case (>2 SESs), two main results are highlighted. First, the combination of properties of some soils is so characteristic that they conform a single SESs bundle, as in the case of the rich SOM soils of alluvial origin in the NE of the region with low agricultural productivity, but high value in regulating SESs. Secondly, some SESs such as potential food provision and water regulation are more important than others to determine locations with high multi-services value at a regional level. This suggests that attention must be paid when ascribing high multi-services value locations as this is not independent of SESs relations. Overall, our results highlight the importance of soils in the potential supply of ESs and show that SESs relations are useful in the implementation of the concept in environmental assessments.

Box2Mask: Box-Supervised Instance Segmentation via Level-Set Evolution.

In contrast to fully supervised methods using pixel-wise mask labels, box-supervised instance segmentation takes advantage of simple box annotations, which has recently attracted increasing research attention. This paper presents a novel single-shot instance segmentation approach, namely Box2Mask, which integrates the classical level-set evolution model into deep neural network learning to achieve accurate mask prediction with only bounding box supervision. Specifically, both the input image and its deep features are employed to evolve the level-set curves implicitly, and a local consistency module based on a pixel affinity kernel is used to mine the local context and spatial relations. Two types of single-stage frameworks, i.e., CNN-based and transformer-based frameworks, are developed to empower the level-set evolution for box-supervised instance segmentation, and each framework consists of three essential components: instance-aware decoder, box-level matching assignment and level-set evolution. By minimizing the level-set energy function, the mask map of each instance can be iteratively optimized within its bounding box annotation. The experimental results on five challenging testbeds, covering general scenes, remote sensing, medical and scene text images, demonstrate the outstanding performance of our proposed Box2Mask approach for box-supervised instance segmentation. In particular, with the Swin-Transformer large backbone, our Box2Mask obtains 42.4% mask AP on COCO, which is on par with the recently developed fully mask-supervised methods.

Mapping in the Context of Primary Social and Science Education: Children, Participation and Spatial Power Relations

Abstract This paper focuses on how children perceive (public) space and the power dynamics within them. The concept of “Education for Spatial Citizenship” (Schulze et al., 2020) is used as a theoretical and methodological framework for promoting life-world-oriented and reflexive learning concerning children’s participation and their roles in public space. The article proposes a mapping-based participatory learning environment, in which children’s desires and visions intersect or conflict with their real-life spatial experiences. The empirical study presented in this article took place in Essen, Germany, and was conducted with a total of 40 primary school children, divided into seven groups. Although the study is rooted in geography education and primary social and science education, the findings of this research are also relevant for civic education and social science education. They offer a broader perspective on student-teacher relationships and interactions in educational settings.

Hider: A Hyperspectral Image Denoising Transformer With Spatial-Spectral Constraints for Hybrid Noise Removal.

Hyperspectral image (HSI) qualities are limited by a mixture of Gaussian noise, impulse noise, stripes, and deadlines during the sensor imaging process, resulting in weak application performance. To enhance HSI qualities, methods based on convolutional neural networks have been successively applied to restore clean data from the observed data. However, the architecture of these methods lacks spectral and spatial constraints, and the convolution operators have limited receptive fields and inflexible model inferences. Thus, in this study, we propose an efficient end-to-end transformer, named HSI denoising transformer (Hider), for mixed HSI noise removal. First, a U-shaped 3-D transformer architecture is built for multiscale feature aggregation. Second, a multihead global spectral attention module within the spectral transformer block is designed to excavate information in different spectral patterns. Finally, an additional locally enhanced cross-spatial attention module within the spatial-spectral transformer block is constructed to build the long-range spatial relationship to avoid the high computational complexity of global spatial self-attention. Through the imposition of global correlations along spectrum and spatial self-similarity constraints on the transformer, our proposed Hider aims to capture long-range spatial contextual information and cluster objects with the same spectral pattern for HSI denoising. To verify the effectiveness and efficiency of Hider, we conducted extensive simulated and real experiments. The denoising results on both simulated and real-world datasets show that Hider achieves superior evaluation metrics and visual assessments compared with other state-of-the-art methods.

VR-HandNet: A Visually and Physically Plausible Hand Manipulation System in Virtual Reality.

This study aims to allow users to perform dexterous hand manipulation of objects in virtual environments with hand-held VR controllers. To this end, the VR controller is mapped to the virtual hand and the hand motions are dynamically synthesized when the virtual hand approaches an object. At each frame, given the information about the virtual hand, VR controller input, and hand-object spatial relations, the deep neural network determines the desired joint orientations of the virtual hand model in the next frame. The desired orientations are then converted into a set of torques acting on hand joints and applied to a physics simulation to determine the hand pose at the next frame. The deep neural network, named VR-HandNet, is trained with a reinforcement learning-based approach. Therefore, it can produce physically plausible hand motion since the trial-and-error training process can learn how the interaction between hand and object is performed under the environment that is simulated by a physics engine. Furthermore, we adopted an imitation learning paradigm to increase visual plausibility by mimicking the reference motion datasets. Through the ablation studies, we validated the proposed method is effectively constructed and successfully serves our design goal. A live demo is demonstrated in the supplementary video.

Effectiveness assessment of recent large vision-language models

The advent of large vision-language models (LVLMs) represents a remarkable advance in the quest for artificial general intelligence. However, the models’ effectiveness in both specialized and general tasks warrants further investigation. This paper endeavors to evaluate the competency of popular LVLMs in specialized and general tasks, respectively, aiming to offer a comprehensive understanding of these novel models. To gauge their effectiveness in specialized tasks, we employ six challenging tasks in three different application scenarios: natural, healthcare, and industrial. These six tasks include salient/camouflaged/transparent object detection, as well as polyp detection, skin lesion detection, and industrial anomaly detection. We examine the performance of three recent open-source LVLMs, including MiniGPT-v2, LLaVA-1.5, and Shikra, on both visual recognition and localization in these tasks. Moreover, we conduct empirical investigations utilizing the aforementioned LVLMs together with GPT-4V, assessing their multi-modal understanding capabilities in general tasks including object counting, absurd question answering, affordance reasoning, attribute recognition, and spatial relation reasoning. Our investigations reveal that these LVLMs demonstrate limited proficiency not only in specialized tasks but also in general tasks. We delve deep into this inadequacy and uncover several potential factors, including limited cognition in specialized tasks, object hallucination, text-to-image interference, and decreased robustness in complex problems. We hope that this study can provide useful insights for the future development of LVLMs, helping researchers improve LVLMs for both general and specialized applications.

Disparity in temporal and spatial relationships between resting-state electrophysiological and fMRI signals.

Resting-state brain networks (RSNs) have been widely applied in health and disease, but the interpretation of RSNs in terms of the underlying neural activity is unclear. To address this fundamental question, we conducted simultaneous recordings of whole-brain resting-state functional magnetic resonance imaging (rsfMRI) and electrophysiology signals in two separate brain regions of rats. Our data reveal that for both recording sites, spatial maps derived from band-specific local field potential (LFP) power can account for up to 90% of the spatial variability in RSNs derived from rsfMRI signals. Surprisingly, the time series of LFP band power can only explain to a maximum of 35% of the temporal variance of the local rsfMRI time course from the same site. In addition, regressing out time series of LFP power from rsfMRI signals has minimal impact on the spatial patterns of rsfMRI-based RSNs. This disparity in the spatial and temporal relationships between resting-state electrophysiology and rsfMRI signals suggests that electrophysiological activity alone does not fully explain the effects observed in the rsfMRI signal, implying the existence of an rsfMRI component contributed by "electrophysiology-invisible" signals. These findings offer a novel perspective on our understanding of RSN interpretation.

TRANSIENT SCENERY FROM THE ARTISTIC PERSPECTIVE

This paper explores the transient landscape from an artistic perspective. “Transience” has always been a topic of much discussion in art research, and in recent years, many artists have focused their research on transient landscapes. In this paper, the methodology of “Graham Gibbs’ Reflective Cycle model” is adopted, and through the descriptive analysis of the works of several artists, it triggers an in-depth reflection on the manifestation of spatial and temporal relations under the phenomenon of “transient landscape”, as well as the real meaning of the artist’s practice. In this paper, we first introduce the context of the study, using the Graham Gibbs’ Reflective Cycle model as a guide to illustrate the work of several artists’ practices, and reflect on the topic from the artists’ point of view. This paper focuses on the concept of “transient” and its origins, and explores its philosophical implications. The contribution of this paper is to deepen the understanding of transient landscapes through an artistic perspective and to recognize the possibility of hidden spatio-temporal relationships between them.

Гендерная и коммуникативная специфика двойной пространственной ориентации говорящего в виртуальной реальности

A clear and unambiguous rendering of spatial relations is the key cognitive function of communication. Spatial relationships reflect the categories and experience shared by all people. The paper introduces the first-of-its-kind attempt to describe the dual spatial orientation of the speaker in virtual reality as depending on two factors: the communicative situation and the gender. Dual spatial orientation combines egocentric, allocentric, or geocentric orientations and their varieties. The combination depends on various factors. The experiment in virtual environment involved 24 informants and five scenarios (scenes) with different communicative parameters. It yielded 725 responses with combinations of double orientations, which went through the IS Semograph. The combination of allocentric and geocentric orientation domineered; the extra-linguistic gender factor and some communicative parameters affected the frequency of combinations of dual spatial orientation. The core of the speaker’s dual spatial orientation in virtual reality combined allocentric and geocentric orientations to the topos of GE \ T + AL and AL + GE \ T.

(088) CONTRALATERAL SYMMETRY IN THE TOPOGRAPHY OF SELF-REPORTED VULVAR PAIN

Abstract Introduction Vulvar pain is a transdiagnostic symptom that is associated with diverse end organs, neurological, and musculoskeletal mechanisms. A better understanding of the neural mechanisms behind the spreading/worsening of vulvar pain across different sites on the vulva, could guide future research, classification, and treatment of vulvar pain-related conditions. Objective Examine the topographic correlations of vulvar pain and provide insights regarding the involvement of neural pathways underlying the topographic distribution of such pain. Methods Self-reports of sexual and genital health were collected from 2199 women through online questionnaires, as part of a larger study. Efforts were made to recruit a heterogenous sample, oversampling for women with genito-pelvic pain. Participants were asked about the presence and location of vulvar and/or vaginal pain. In addition, participants were asked about history and types of non-genital injuries. If a participant reported chronic vulvar pain, they were then presented with a drawing of a vulva divided into 24 areas, asking them to indicate the location(s) of their pain (localization of vaginal pain not included). Correlations were conducted on the presence of pain across vulvar areas that were contralaterally distributed (horizontal) vs spatially close but ipsilaterally distributed (vertical and radial), and distinguishing between women with vs without a history of back and/or pelvic injury. Results 911 women reported experiencing chronic vulvar or vulvar-vaginal pain. Results from correlations across vulvar areas are reported in Figure 1, 2 and 3, distinguishing between women with and without a history of back/pelvic injury. In both groups, horizontal correlations across left and right sides of the same tissue/structure were the strongest (Figure 1), whereas areas that were spatially close on the ipsilateral side (vertical and radial) showed weaker correlations (Figures 2 &amp; 3). Statistical tests confirmed differences in correlations both between-groups and within-groups, across different spatial relations (horizontal, vertical, radial). Conclusions The observed pattern of topographic correlations suggests specific constraints on the mechanisms involved in the spreading of vulvar pain, showing a horizontally symmetrical distribution of correlations (within tissue/structure), as opposed to a proximity-based distribution of correlations. This appears to reflect symmetrically contralateral sensitization within the same structure, even if the areas were relatively distant (e.g., right and left labia minora). According to our results, this applies more to women without a history of back/pelvic injury, compared to women with a history back/pelvic injury. Consistent with past research showing that a temporary nociceptive stimulus induces hyperalgesia and allodynia in the ipsilateral and symmetrically contralateral site of noxious stimulation (Shenker et al.,2008), our results suggest that similar neural mechanism(s) might be involved in chronic vulvar pain. The pattern of correlations found could be mediated by chronic pain-induced activation of the commissural fibers that connect, and perhaps sprout into, the symmetrically contralateral sides at the level of the dorsal horn. This would constitute a form of central sensitization (Latremoliere &amp; Woolf, 2007). Whether a comparable central sensitization process also occurs at the level of the sensory cortex (or other brain regions), or whether different forms of synaptic potentiation at the neural level are involved, must also be considered. Disclosure Any of the authors act as a consultant, employee or shareholder of an industry for: Adamo Bioscience.

HIGHRISE BUILDINGS IN HOUSING CONSTRUCTION: ECONOMIC EFFICIENCY VS SPATIAL BALANCE

&amp;lt;p class=&amp;quot;ABSTRACT&amp;quot;&amp;gt;Nowadays, many growing cities around the world face very intensive construction in the housing sector. The pressure of the construction sector is reflected in the insistence on increasing the number of floors of residential buildings due to economic profitability, while the principles of spatial balance and sustainability are questionable. Therefore, this research seeks to discover which minimum number of floors of residential buildings ensures the economic profitability of the investment in housing construction. The methodological approach, based on the relevant principles of urban economy, ensures the development of a model for empirical testing of spatial and economic relations. The research polygon is the residential module from the regulatory plan of the residential settlement located in the area of a fairground in Banja Luka - Bosnia and Herzegovina.&amp;lt;/p&amp;gt;

Enhanced Encrypted Traffic Analysis Leveraging Graph Neural Networks and Optimized Feature Dimensionality Reduction

With the continuously growing requirement for encryption in network environments, web browsers are increasingly employing hypertext transfer protocol security. Despite the increase in encrypted malicious network traffic, the encryption itself limits the data accessible for analyzing such behavior. To mitigate this, several studies have examined encrypted network traffic by analyzing metadata and payload bytes. Recent studies have furthered this approach, utilizing graph neural networks to analyze the structural data patterns within malicious encrypted traffic. This study proposed an enhanced encrypted traffic analysis leveraging graph neural networks which can model the symmetric or asymmetric spatial relations between nodes in the traffic network and optimized feature dimensionality reduction. It classified malicious network traffic by leveraging key features, including the IP address, port, CipherSuite, MessageLen, and JA3 features within the transport-layer-security session data, and then analyzed the correlation between normal and malicious network traffic data. The proposed approach outperformed previous models in terms of efficiency, using fewer features while maintaining a high accuracy rate of 99.5%. This demonstrates its research value as it can classify malicious network traffic with a high accuracy based on fewer features.

Person re-identification by utilizing hierarchical spatial relation reasoning

Deep learning-based methods have become increasingly popular in person re-identification (re-id) as they enhance the discriminative features while suppressing extraneous or irrelevant details. However, these methods face a potential limitation in overlooking the relationships between semantic regions focusing solely on salient features or pixel-level matching information. As a result, the model performance becomes susceptible to adverse factors such as pose variations, misalignment, and background clutter. To overcome this issue, we propose a novel deep relational learning network (DRLNet) that learns relationship knowledge between semantic regions and “local–global” information to improve discriminative feature learning effectively. We manually establish the relation between local regions using a simple partition strategy that captures global structural information and improves relation-aware attention learning. Dividing images into uniform partitions allows each part-level feature to represent itself and the entire image area, facilitating the learning of more refined and discerning relationships. Our approach's efficacy is evaluated on three datasets, where experimental results exhibit its superiority. Significantly, our approach achieves a new state-of-the-art Rank-1 accuracy of 84.7%, 80.9%, and 83.5% on CUHK03 (Labeled), CUHK03 (Detected), and MSMT17 datasets, respectively, outperforming current state-of-the-art methods. In conclusion, our proposed DRLNet demonstrates its potential as a highly effective solution to overcoming the limitations of deep learning-based methods by providing superior means of learning discriminative features while incorporating relationships between semantic regions.

Test Input Prioritization for 3D Point Clouds

3D point cloud applications have become increasingly prevalent in diverse domains, showcasing their efficacy in various software systems. However, testing such applications presents unique challenges due to the high-dimensional nature of 3D point cloud data and the vast number of possible test cases. Test input prioritization has emerged as a promising approach to enhance testing efficiency by prioritizing potentially misclassified test cases during the early stages of the testing process. Consequently, this enables the early labeling of critical inputs, leading to a reduction in the overall labeling cost. However, applying existing prioritization methods to 3D point cloud data is constrained by several factors: (1) inadequate consideration of crucial spatial information, and (2) susceptibility to noises inherent in 3D point cloud data. In this article, we propose PCPrior , the first test prioritization approach specifically designed for 3D point cloud test cases. The fundamental concept behind PCPrior is that test inputs closer to the decision boundary of the model are more likely to be predicted incorrectly. To capture the spatial relationship between a point cloud test and the decision boundary, we propose transforming each test (a point cloud) into a low-dimensional feature vector, toward indirectly revealing the underlying proximity between a test and the decision boundary. To achieve this, we carefully design a group of feature generation strategies, and for each test input, we generate four distinct types of features, namely spatial features, mutation features, prediction features, and uncertainty features. Through a concatenation of the four feature types, PCPrior assembles a final feature vector for each test. Subsequently, a ranking model is employed to estimate the probability of misclassification for each test based on its feature vector. Finally, PCPrior ranks all tests based on their misclassification probabilities. We conducted an extensive study based on 165 subjects to evaluate the performance of PCPrior, encompassing both natural and noisy datasets. The results demonstrate that PCPrior outperforms all of the compared test prioritization approaches, with an average improvement of 10.99% to 66.94% on natural datasets and 16.62% to 53% on noisy datasets.

Exploring Brain Effective Connectivity Networks Through Spatiotemporal Graph Convolutional Models.

Learning brain effective connectivity networks (ECN) from functional magnetic resonance imaging (fMRI) data has gained much attention in recent years. With the successful applications of deep learning in numerous fields, several brain ECN learning methods based on deep learning have been reported in the literature. However, current methods ignore the deep temporal features of fMRI data and fail to fully employ the spatial topological relationship between brain regions. In this article, we propose a novel method for learning brain ECN based on spatiotemporal graph convolutional models (STGCM), named STGCMEC, in which we first adopt the temporal convolutional network to extract the deep temporal features of fMRI data and utilize the graph convolutional network to update the spatial features of each brain region by aggregating information from neighborhoods, which makes the features of brain regions more discriminative. Then, based on such features of brain regions, we design a joint loss function to guide STGCMEC to learn the brain ECN, which includes a task prediction loss and a graph regularization loss. The experimental results on a simulated dataset and a real Alzheimer's disease neuroimaging initiative (ADNI) dataset show that the proposed STGCMEC is able to better learn brain ECN compared with some state-of-the-art methods.