Recently, Graph Neural Networks (GNNs) have gained widespread application in automatic brain network classification tasks, owing to their ability to directly capture crucial information in non-Euclidean structures. However, two primary challenges persist in this domain. First, within the realm of clinical neuro-medicine, signals from cerebral regions are inevitably contaminated with noise stemming from physiological or external factors. The construction of brain networks heavily relies on set thresholds and feature information within brain regions, making it susceptible to the incorporation of such noises into the brain topology. Additionally, the static nature of the artificially constructed brain network\'s adjacent structure restricts real-time changes in brain topology. Second, mainstream GNN-based approaches tend to focus solely on capturing information interactions of nearest neighbor nodes, overlooking high-order topology features. In response to these challenges, we propose an adaptive unsupervised Spatial-Temporal Dynamic Hypergraph Information Bottleneck (ST-DHIB) framework for dynamically optimizing brain networks. Specifically, adopting an information theory perspective, Graph Information Bottleneck (GIB) is employed for purifying graph structure, and dynamically updating the processed input brain signals. From a graph theory standpoint, we utilize the designed Hypergraph Neural Network (HGNN) and Bi-LSTM to capture higher-order spatial-temporal context associations among brain channels. Comprehensive patient-specific and cross-patient experiments have been conducted on two available datasets. The results demonstrate the advancement and generalization of the proposed framework.

Air pollution is a primary concern, causing around 7 million premature deaths annually, with traffic-related sources contributing 23 %-45 % of emissions. While several studies have surveyed vehicle emission models, they are either outdated or focus on specific data-driven models. This paper systematically reviews vehicle emission prediction models, comparing traditional approaches with data-driven emission models. The traditional emission models can be divided into average-speed, modal, and other models, noting their reliance on empirical assumptions and parameters that may not be universally applicable. In contrast, we delve into data-driven models utilizing dynamometer and on-road test data for time-series and spatial-temporal predictions. The application of these models is discussed across various scenarios, highlighting the progress and gap. We observed that traditional models, primarily estimating total traffic emissions in study regions, lack micro-level detail crucial for tailored decisions. The direct link between road emission model accuracy and input data quality poses challenges in disaggregating on-road vehicle emission inventories. Due to unique transportation instruments, traffic fleet components, and patterns, exploring the effects of emission-reduction policies in specific cities or regions is urgent. Vehicle characteristics, environmental conditions, traffic scenarios, and prediction scales are common effect factors, while instantaneous driving profiles prove effective in model calibration. In data-driven models, ANN outperforms in estimating emissions and performance of low-power diesel engines with errors not exceeding 5 %. However, no single data-driven method performed excellently in predicting all pollutants. Besides, integrated methods utilizing LSTM, GRU, and RNN outperform individual models. To enhance prediction accuracy considering the inherent connectivity of road networks and spatiotemporal variation patterns of vehicle emissions, GCN is an emerging approach for capturing spatial-temporal relationships based on remote sensing data. Moreover, limited data-driven studies have been performed to forecast particle matter emissions, the main contributors to urban pollution, calling for more attention for future research.

We study video inpainting, which aims to recover realistic textures from damaged frames. Recent progress has been made by taking other frames as references so that relevant textures can be transferred to damaged frames. However, existing video inpainting approaches neglect the ability of the model to extract information and reconstruct the content, resulting in the inability to reconstruct the textures that should be transferred accurately. In this paper, we propose a novel and effective spatial-temporal texture transformer network (STTTN) for video inpainting. STTTN consists of six closely related modules optimized for video inpainting tasks: feature similarity measure for more accurate frame pre-repair, an encoder with strong information extraction ability, embedding module for finding a correlation, coarse low-frequency feature transfer, refinement high-frequency feature transfer, and decoder with accurate content reconstruction ability. Such a design encourages joint feature learning across the input and reference frames. To demonstrate the advancedness and effectiveness of the proposed model, we conduct comprehensive ablation learning and qualitative and quantitative experiments on multiple datasets by using standard stationary masks and more realistic moving object masks. The excellent experimental results demonstrate the authenticity and reliability of the STTTN.

The COVID-19 pandemic has had great impact on human health and social economy. Several studies examined spatial and temporal patterns of health risk factors associated with COVID-19, but population flow spillover effect has not been sufficiently considered. In this paper, a population flow-based spatial-temporal eigenvector filtering model (FLOW-ESTF) was developed to consider spatial-temporal patterns and population flow connectivity simultaneously. The proposed FLOW-ESTF method efficiently improved model prediction accuracy, which could help the government aware of the infection risk level and to make suitable control policies. The selected population flow spatial-temporal eigenvector contributed most to modeling and the visualization of corresponding eigenvector set helped to explore the underlying spatial-temporal patterns and pandemic transmission nodes. The model coefficients could reflect how health risk factors contribute the modeling of state-level COVID-19 weekly increased cases and how their influence changed through time, which could help people and government to better aware the potential health risks and to adjust control measures at different stage. The extracted population flow spatial-temporal eigenvector not only represents influence of population flow and its spillover effects but also represents some possible omitted health risk factors. This could provide an efficient path to solve the problem of spatial and temporal autocorrelation in COVID-19 modeling and an intuitive way to discover underlying spatial patterns, which will partially compensate for the problems of insufficient consideration of potential risk variables and missing data.

Aba\'s topography, weather, and climate make it prone to landslides, mudslides, and other natural disasters, which limit economic and social growth. Assessing and improving regional resilience is important to mitigate natural disasters and achieve sustainable development. In this paper, the entropy weight method is used to calculate the resilience of Aba under multi-hazard stress from 2010 to 2018 by combining the existing framework with the disaster resilience of the place (DROP) model. Then spatial-temporal characteristics are analyzed based on the coefficient of variation and exploratory spatial data analysis (ESDA). Finally, partial least squares (PLS) regression is used to identify the key influences on disaster resilience. The results show that (1) the disaster resilience in Aba increased from 2010 to 2018 but dropped in 2013 and 2017 due to large-scale disasters. (2) There are temporal and spatial differences in the level of development in each of the Aba counties. From 2010 to 2016, disaster resilience shows a significant positive spatial association and high-high (HH) aggregation in the east and low-low (LL) aggregation in the west. Then the spatial aggregation weakened after 2017. This paper proposes integrating regional development, strengthening the development level building, and emphasizing disaster management for Aba.

OBJECTIVE: Clarification of propagation patterns in amyotrophic lateral sclerosis (ALS) is challenging, but understanding these has implications for individual prognostication and clinical trial design. However, systematic knowledge in this area is lacking. The aim of this study was to characterize the spatial and temporal features of propagation patterns in ALS, and to evaluate the association between propagation patterns and survival.
METHODS: A cohort of 833 patients with ALS, diagnosed between January 2018 and December 2019 and followed to August 2021, was analysed. Spatial and temporal features of propagation patterns were determined based on the involved functional regions (bulbar, cervical, thoracic/respiratory and lumbar) in time order. The final propagation pattern was identified in patients with at least three functional regions involved. Kaplan-Meier analysis and Cox regression analysis were performed.
RESULTS: During a median follow-up of 21.2 months, 19 final propagation patterns were identified in 657 patients (78.9%). In survival analysis, we found that the earlier the respiratory functional region becomes involved, the higher the risk of death (time order: 1st: hazard ratio [HR], 3.35, 95% confidence interval [CI] 1.23-9.15; 2nd: HR 2.45, 95% CI 1.55-3.87; 3rd: HR 1.94, 95% CI 1.52-2.49), adjusting for age, sex, diagnostic delay, revised ALS Functional Rating Scale score, cognitive impairment and riluzole. Shorter interval time between involved regions was an independent adverse prognostic factor.
CONCLUSIONS: The propagation patterns of ALS are varied. The order in which the respiratory region becomes involved and the interval time between involvement of functional regions are predictors for prognosis.

Objectives: This was a long-term retrospective study, aiming to understand the temporal and spatial trend of cirrhosis in Wuwei from 1995 to 2016, explore its spatio-temporal aggregation, and find out the high incidence areas. To provide theoretical basis for the formulation of comprehensive prevention and treatment strategy of cirrhosis in Wuwei. Methods: Herein, we extracted data of cirrhosis patients who were treated in 12 sentinel hospitals in Wuwei from their medical records. We used SAS and Joinpoint Regression Program for data analysis, SaTScan 9.4 software for clustering area detection, and ArcGIS 10.2 software for geographical distribution mapping. Results: Among 3308 patients with liver cirrhosis (average age, 55.34 years) included in this study, 15.9% were aged 50-54 years. The majority were men (2716, 65.8%), with a sex ratio of 1.92:1 and peasants by occupation (1369, 60.3%). The basic social medical insurance system covered the healthcare costs of 1271 patients (63%). A Joinpoint regression analysis done for 1995-2016 revealed an increase in the standardized cirrhosis rate [average annual percent change (AAPC) = 16.7% (95% CI, 10.2-23.5%)] with three joinpoints in 2010, 2013, and 2016. The annual percent change (APC) from 1995 to 2010 was 11.13% (95% CI: 6.5-16.0), and APC from 2010 to 2013 was 66.48% (95% CI:16.0-138.9); conversely, from 2013 to 2016, APC was 4.4% (95% CI, -7.5-17.8%). Hongshagang Town showed the highest average incidence. Each township showed a gradual increase in the incidence after 2010. The results revealed that in each township, liver cirrhosis incidence had some spatial aggregation and was nonrandom. Four liver cirrhosis clusters were noted in 75 townships in Wuwei. Data were gathered from 2011 to 2016. Conclusions: From 1995 to 2016, the incidence of cirrhosis in Wuwei still showed an increasing trend, but the growth rate slowed down since 2013. In Wuwei, the rate of standardization of cirrhosis in female patients increased steadily and faster than in male patients. It is necessary to strengthen the diagnosis, treatment, prevention, and control measures of cirrhosis-related diseases. The results of spatial scanning, basic spatial distribution, aggregation time, and time trend analysis were consistent.

Skeleton-based human action recognition has made great progress, especially with the development of a graph convolution network (GCN). The most important work is ST-GCN, which automatically learns both spatial and temporal patterns from skeleton sequences. However, this method still has some imperfections: only short-range correlations are appreciated, due to the limited receptive field of graph convolution. However, long-range dependence is essential for recognizing human action. In this work, we propose the use of a spatial-temporal relative transformer (ST-RT) to overcome these defects. Through introducing relay nodes, ST-RT avoids the transformer architecture, breaking the inherent skeleton topology in spatial and the order of skeleton sequence in temporal dimensions. Furthermore, we mine the dynamic information contained in motion at different scales. Finally, four ST-RTs, which extract spatial-temporal features from four kinds of skeleton sequence, are fused to form the final model, multi-stream spatial-temporal relative transformer (MSST-RT), to enhance performance. Extensive experiments evaluate the proposed methods on three benchmarks for skeleton-based action recognition: NTU RGB+D, NTU RGB+D 120 and UAV-Human. The results demonstrate that MSST-RT is on par with SOTA in terms of performance.

Geometry-based properties and characteristics of drug molecules play an important role in drug development for virtual screening in computational chemistry. The 3D characteristics of molecules largely determine the properties of the drug and the binding characteristics of the target. However, most of the previous studies focused on 1D or 2D molecular descriptors while ignoring the 3D topological structure, thereby degrading the performance of molecule-related prediction. Because it is very time-consuming to use dynamics to simulate molecular 3D conformer, we aim to use machine learning to represent 3D molecules by using the generated 3D molecular coordinates from the 2D structure.
We proposed Drug3D-Net, a novel deep neural network architecture based on the spatial geometric structure of molecules for predicting molecular properties. It is grid-based 3D convolutional neural network with spatial-temporal gated attention module, which can extract the geometric features for molecular prediction tasks in the process of convolution. The effectiveness of Drug3D-Net is verified on the public molecular datasets. Compared with other deep learning methods, Drug3D-Net shows superior performance in predicting molecular properties and biochemical activities.
https://github.com/anny0316/Drug3D-Net.
Supplementary data are available online at https://academic.oup.com/bib.

Exploring the spatial patterns and temporal dynamics of human brain activity has been of great interest, in the quest to better understand connectome-scale brain networks. Though modeling spatial and temporal patterns of functional brain networks have been researched for a long time, the development of a unified and simultaneous spatial-temporal model has yet to be realized. For instance, although some deep learning methods have been proposed recently in order to model functional brain networks, most of them can only represent either spatial or temporal perspective of functional Magnetic Resonance Imaging (fMRI) data and rarely model both domains simultaneously. Due to the recent success in applying sequential auto-encoders for brain decoding, in this paper, we propose a deep sparse recurrent auto-encoder (DSRAE) to be applied unsupervised to learn spatial patterns and temporal fluctuations of brain networks at the same time. The proposed DSRAE was evaluated and validated based on three tasks of the publicly available Human Connectome Project (HCP) fMRI dataset, resulting with promising evidence. To the best of our knowledge, the proposed DSRAE is among the early efforts in developing unified models that can extract connectome-scale spatial-temporal networks from 4D fMRI data simultaneously.