Multiple Subregions Research Articles

Sub-region based histogram analysis of amide proton transfer-weighted MRI for predicting tumor budding grade in rectal adenocarcinoma: a prospective study.

To explore the sub-regional histogram features of amide proton transfer-weighted (APTw) MRI, compared with those of diffusion-weighted imaging (DWI), in predicting the tumor budding (TB) grade of rectal cancer (RC). This study prospectively enrolled 74 patients with pathologically confirmed RC, who underwent APTw MRI before surgery from July 2022 to March 2023. Hematoxylin-eosin staining was used for TB scoring. K-means clustering (K = 4-6) was applied to obtain multiple sub-regions (n = 3-5), and corresponding histogram features (including mean, standard deviation, minimum, maximum, and 10th, 25th, 50th, 75th, and 90th quantile) of APT and apparent diffusion coefficient (ADC) maps were extracted and filtered using stepwise regression. When K = 5, the K-means clustering is four sub-regions, showing the best prediction for TB grade compared to K = 4 or 6. When K = 5, there were significantly higher histogram features of the APT map in sub-regions 3 and 4 in the high TB grade group compared to the low-intermediate TB grade group. Receiver operating characteristic (ROC) curve and internal validation suggested that the predictive efficiency of the model was highest when K = 5, with AUC, sensitivity, specificity, accuracy, and kappa values of 0.92, 93%, 71%, 87%, and 0.65, respectively. There were no significant differences in the histogram features of each sub-region in the ADC map (p > 0.05). The sub-regional histogram features of APTw images can help to distinguish the heterogeneous regions of RC, which can be used to predict the TB grade of RC. Question Can the sub-regional histogram features of APTw MRI predict the tumor budding (TB) grade of rectal cancer (RC)? Findings Differences exist in histogram features of APT map subregions between high and low-intermediate TB grade groups; subregions of the APT map have different predictive abilities. Clinical relevance APT-weighted imaging might outperform DWI in predicting TB grade in RC.

Comparative transcriptomic rhythms in the mouse and human prefrontal cortex.

Alterations in multiple subregions of the human prefrontal cortex (PFC) have been heavily implicated in psychiatric diseases. Moreover, emerging evidence suggests that circadian rhythms in gene expression are present across the brain, including in the PFC, and that these rhythms are altered in disease. However, investigation into the potential circadian mechanisms underlying these diseases in animal models must contend with the fact that the human PFC is highly evolved and specialized relative to that of rodents. Here, we use RNA sequencing to lay the groundwork for translational studies of molecular rhythms through a sex-specific, cross species comparison of transcriptomic rhythms between the mouse medial PFC (mPFC) and two subregions of the human PFC, the anterior cingulate cortex (ACC) and the dorsolateral PFC (DLPFC). We find that while circadian rhythm signaling is conserved across species and subregions, there is a phase shift in the expression of core clock genes between the mouse mPFC and human PFC subregions that differs by sex. Furthermore, we find that the identity of rhythmic transcripts is largely unique between the mouse mPFC and human PFC subregions, with the most overlap (20%, 236 transcripts) between the mouse mPFC and the human ACC in females. Nevertheless, we find that basic biological processes are enriched for rhythmic transcripts across species, with key differences between regions and sexes. Together, this work highlights both the evolutionary conservation of transcriptomic rhythms and the advancement of the human PFC, underscoring the importance of considering cross-species differences when using animal models.

Hippocampal temporal dynamics and spatial heterogeneity unveil vulnerability markers in the offspring of bipolar patients.

Bipolar disorder (BD) is a highly heritable disorder characterized by emotion dysregulation and recurrent oscillations between mood states. Despite the proven efficacy of early interventions, vulnerability markers in high-risk individuals are still lacking. BD patients present structural alterations of the hippocampus, a pivotal hub of emotion regulation networks composed of multiple subregions with different projections. However, the hippocampal dynamic functional connectivity (dFC) in BD remains unclear. We aim to investigate whether the dFC of hippocampal subdivisions differentiates BD patients, offspring of BD patients (BDoff), and healthy controls (HC); and whether it correlates with symptoms differently between these groups. We studied for the first time the dFC of the hippocampus through a cutting-edge micro-co-activation patterns (μCAPs) analysis of resting-state functional MRI data of 97 subjects (26 BD, 18 BDoff, 53 HC). μCAPs allow a data-driven differentiation within the seed region. dFC between the hippocampal body and a somatomotor-μCAP was lower both in BD patients (p-valueFDR:0.00015) and in BDoff (p-valueFDR:0.020) than in HC. Inversely, dFC between the hippocampal head and a limbic-μCAP was higher in BD patients than in HC (p-valueFDR: 0.005). Furthermore, the correlations between a frontoparietal-μCAP and both depression and emotion dysregulation symptoms were significantly higher in BD than HC (p-valueFDR <0.02). Overall, we observed alterations of large-scale functional brain networks associated with decreased cognitive control flexibility and disrupted somatomotor, saliency, and emotion processing in BD. Interestingly, BDoff presented an intermediate phenotype between BD and HC, suggesting that dFC of hippocampal subregions might represent a marker of vulnerability to BD.

Migratory movement of photo-identified humpback whales Megaptera novaeangliae along the southeastern coast of Africa

The global conservation of humpback whales Megaptera novaeangliae relies on continuous assessments of the populations’ post-whaling status, which in turn require information on population mixing and movements. One of the seven breeding stocks in the Southern Hemisphere, breeding stock C and its associated four substocks (C1–C4), utilises the western Indian Ocean in several discrete breeding grounds. However, information on the mixing of stocks across breeding grounds is limited. This study utilises photo-identification methods applied to the first multi-regional catalogue of fluke images available for substock C1S (the southern component of substock C1) which uses the southern portion of the breeding ground, developed as part of this study, to investigate the whales’ movements along the southeastern coast of southern Africa. The identification of 1 746 unique individuals, including 11 withinyear matches and 48 between-year matches of 45 individuals, provides new insight into the intra-regional migration patterns, fidelity and structure of substock C1S. The within-year match results confirmed that individual humpback whales are broadly seasonally present for extended periods and visit multiple subregions along this coast, while the between-year matches revealed long-term fidelity to this coastline. Moreover, five of the between-year matches between subregions in South Africa and Mozambique linked the South African migration corridor to the breeding ground in southern and central Mozambique. These findings provide valuable insights into the movement patterns of substock C1S and significantly address the challenges of effective population management, which includes difficulties associated with the whale-watching industry on the east coast of South Africa.

Region-specific activation in the accumbens nucleus by itch with modified scratch efficacy in mice – a model-free multivariate analysis

Itch is a protective/defensive function with divalent motivational drives. Itch itself elicits an unpleasant experience, which triggers the urge to scratch, relieving the itchiness. Still, it can also result in dissatisfaction when the scratch is too intense and painful or unsatisfactory due to insufficient scratch effect. Therefore, it is likely that the balance between the unpleasantness/pleasure and satisfaction/unsatisfaction associated with itch sensation and scratching behavior is determined by complex brain mechanisms. The physiological/pathological mechanisms underlying this balance remain largely elusive. To address this issue, we targeted the "reward center" of the brain, the nucleus accumbens (NAc), in which itch-responsive neurons have been found in rodents. We examined how neurons in the NAc are activated or suppressed during histamine-induced scratching behaviors in mice. The mice received an intradermal injection of histamine or saline at the neck, and the scratching number was analyzed by recording the movement of the bilateral hind limbs for about 45 min after injection. To experimentally manipulate the scratch efficacy in these histamine models, we compared histamine's behavioral and neuronal effects between mice with intact and clipped nails on the hind paws. As expected, the clipping of the hind limb nail increased the number of scratches after the histamine injection. In the brains of mice exhibiting scratching behaviors, we analyzed the expression of the c-fos gene (Fos) as a readout of an immediate activation of neurons during itch/scratch and dopamine receptors (Drd1 and Drd2) using multiplex single-molecule fluorescence in situ hybridization (RNAscope) in the NAc and surrounding structures. We performed a model-free analysis of gene expression in geometrically divided NAc subregions without assuming the conventional core–shell divisions. The results indicated that even within the NAc, multiple subregions responded differentially to various itch/scratch conditions. We also found different clusters with neurons showing similar or opposite changes in Fos expression and the correlation between scratch number and Fos expression in different itch/scratch conditions. These regional differences and clusters would provide a basis for the complex role of the NAc and surrounding structures in encoding the outcomes of scratching behavior and itchy sensations.

Isotropic and orthotropic mechanical properties of in vivo mandible model: Assessment of computational modelling

This study assesses the various methodologies utilized in the literature for computational mandible modeling and their implications on research findings. Previous studies have explored diverse material properties for the mandible, whether modeled as a single entity, in pairs, or as multiple sub-regions. Within this framework, a finite element model of an intact mandible was created using sliced tomographic images to examine the dispersion of stress fields resulting from different sets of bone material properties sourced from specialized literature. This research introduces the methodology employed in mandible model development, strategies for achieving high precision, and the influence of mechanical properties. Through an evaluation of models proposed by different authors, significant dispersion was observed, leading to disparities ranging from 4.4% to 67.4% in stress field outcomes at the symphysis region. RMSE served as an estimator to assess the similarity between these models. The findings suggest that the Caraveo (2008) and Palka (2020) models exhibit notable convergence in predicting stress fields in the symphysis regions. By incorporating good modeling practices in the intact mandible model, a more reliable representation is ensured, laying the groundwork for future investigations involving nonlinear contact phenomena and topology optimization.

Correlation Between Prefrontal Functional Connectivity and the Degree of Cognitive Impairment in Alzheimer's Disease: A Functional Near-Infrared Spectroscopy Study.

The development of Alzheimer's disease (AD) can be divided into subjective cognitive decline (SCD), mild cognitive impairment (MCI), and dementia. Early recognition of pre-AD stages may slow the progression of dementia. This study aimed to explore functional connectivity (FC) changes of the brain prefrontal cortex (PFC) in AD continuum using functional near-infrared spectroscopy (fNIRS), and to analyze its correlation with cognitive function. All participants underwent 48-channel fNIRS at resting-state. Based on Brodmann partitioning, the PFC was divided into eight subregions. The NIRSIT Analysis Tool (v3.7.5) was used to analyze mean ΔHbO2 and FC. Spearman correlation analysis was used to examine associations between FC and cognitive function. Compared with HC group, the mean ΔHbO2 and FC were different between multiple subregions in the AD continuum. Both mean ΔHbO2 in the left dorsolateral PFC and average FC decreased sequentially from SCD to MCI to AD groups. Additionally, seven pairs of subregions differed in FC among the three groups: the differences between the MCI and SCD groups were in heterotopic connectivity; the differences between the AD and SCD groups were in left intrahemispheric and homotopic connectivity; whereas the MCI and AD groups differed only in homotopic connectivity. Spearman correlation results showed that FCs were positively correlated with cognitive function. These results suggest that the left dorsolateral PFC may be the key cortical impairment in AD. Furthermore, there are different resting-state prefrontal network patterns in AD continuum, and the degree of cognitive impairment is positively correlated with reduced FC strength.

Further remarks on de Sitter space, extremal surfaces, and time entanglement

We develop further the investigations by Narayan [de Sitter space, extremal surfaces, and time entanglement, ] on de Sitter space, extremal surfaces, and time entanglement. We discuss the no-boundary de Sitter extremal surface areas as certain analytic continuations from AdS while also amounting to space-time rotations. The structure of the extremal surfaces suggests a geometric picture of the time-entanglement or pseudo-entanglement wedge. We also study some entropy relations for multiple subregions. The analytic continuation suggests a heuristic Lewkowycz-Maldacena formulation of the extremal surface areas. In the bulk, this is now a replica formulation on the Wavefunction, which suggests interpretation as pseudo-entropy. Finally, we also discuss aspects of future-past entangled states and time evolution. Published by the American Physical Society 2024

Unsupervised classification of brain-wide axons reveals the presubiculum neuronal projection blueprint

We present a quantitative strategy to identify all projection neuron types from a given region with statistically different patterns of anatomical targeting. We first validate the technique with mouse primary motor cortex layer 6 data, yielding two clusters consistent with cortico-thalamic and intra-telencephalic neurons. We next analyze the presubiculum, a less-explored region, identifying five classes of projecting neurons with unique patterns of divergence, convergence, and specificity. We report several findings: individual classes target multiple subregions along defined functions; all hypothalamic regions are exclusively targeted by the same class also invading midbrain and agranular retrosplenial cortex; Cornu Ammonis receives input from a single class of presubicular axons also projecting to granular retrosplenial cortex; path distances from the presubiculum to the same targets differ significantly between classes, as do the path distances to distinct targets within most classes; the identified classes have highly non-uniform abundances; and presubicular somata are topographically segregated among classes. This study thus demonstrates that statistically distinct projections shed light on the functional organization of their circuit.

Atrophy patterns in hippocampus and amygdala subregions of depressed patients with Parkinson's disease

We aimed to explore the subregional atrophy patterns of the amygdala and hippocampus in Parkinson's disease (PD) with depression and their correlation with the severity of the depressive symptom. MRI scans were obtained for 34 depressed PD patients (DPD), 22 nondepressed PD patients (NDPD), and 28 healthy controls (HC). Amygdala and hippocampal subregions were automatically segmented, and the intergroup volume difference was compared. The relationships between the volumes of the subregions and depression severity were investigated. Logistic analysis and Receiver operator characteristic curve were used to find independent predictors of DPD. Compared with the HC group, atrophy of the bilateral lateral nucleus, left accessory basal nucleus, right cortical nucleus, right central nucleus, and right medial nucleus subregions of the amygdala were visible in the DPD group, while the right lateral nucleus subregion of the amygdala was smaller in the DPD group than in the NDPD group. The DPD group showed significant atrophy in the left molecular layer, left GC-DG, left CA3, and left CA4 subregions compared with the HC group for hippocampal subregion volumes. Also, the right lateral nuclei volume and disease duration were independent predictors of DPD. To sum up, DPD patients showed atrophy in multiple amygdala subregions and left asymmetric hippocampal subregions. The decreased amygdala and hippocampal subregion volumes were correlated with the severity of depressive symptoms. The volume of right lateral nuclei and disease duration could be used as a biomarker to detect DPD.

Research on defect detection of bottle cap interior based on low-angle and large divergence angle vision system.

During the machine vision inspection of the inner section of bottle caps within pharmaceutical packaging, the unique conca bottom and convex side walls often create obstructions to the illumination. Consequently, this results in challenges such as irregular background and diminished feature contrast in the image, ultimately leading to the misidentification of defects. As a solution, a vision system characterized by a Low-Angle and Large Divergence Angle (LALDA) is presented in this paper. Using the large divergence angle of LED, combined with low-angle illumination, a uniform image of the side wall region with bright-field characteristics and a uniform image of inner circle region at the bottom with dark-field characteristics are obtained, thus solving the problems of light being obscured and brightness overexposure of the background. Based on the imaging characteristics of LALDA, a multi-channel segmentation (MCS) algorithm is designed. The HSV color space has been transformed, and the image is automatically segmented into multiple sub-regions by mutual calculation of different channels. Further, image homogenization and enhancement are used to eliminate fluctuations in the background and to enhance the contrast of defects. In addition, a variety of defect extraction methods are designed based on the imaging characteristics of different sub-regions, which can avoid the problem of over-segmentation in detection. In this paper, the LALDA is applied to the defect detection inside the cap of capsule medicine bottle, the detection speed is better than 400 pcs/min and the detection accuracy is better than 95%, which can meet the actual production line capacity and detection requirements.

Partition-A-Medical-Image: Extracting Multiple Representative Subregions for Few-Shot Medical Image Segmentation

Partition-A-Medical-Image: Extracting Multiple Representative Subregions for Few-Shot Medical Image Segmentation

High-Throughput Microelectrode Arrays for Precise Functional Localization of the Globus Pallidus Internus.

The globus pallidus internus (GPi) was considered a common target for stimulation in Parkinson's disease (PD). Located deep in the brain and of small size, pinpointing it during surgery is challenging. Multi-channel microelectrode arrays (MEAs) can provide micrometer-level precision functional localization, which can maximize the surgical outcome. In this paper, a 64-channel MEA modified by platinum nanoparticles with a detection site impedance of 61.1 kΩ was designed and prepared, and multiple channels could be synchronized to cover the target brain region and its neighboring regions so that the GPi could be identified quickly and accurately. The results of the implant trajectory indicate that, compared to the control side, there is a reduction in local field potential (LFP) power in multiple subregions of the upper central thalamus on the PD-induced side, while the remaining brain regions exhibit an increasing trend. When the MEA tip was positioned at 8,700 μm deep in the brain, the various characterizations of the spike signals, combined with the electrophysiological characteristics of the β-segmental oscillations in PD, enabled MEAs to localize the GPi at the single-cell level. More precise localization could be achieved by utilizing the distinct characteristics of the internal capsule (ic), the thalamic reticular nucleus (Rt), and the peduncular part of the lateral hypothalamus (PLH) brain regions, as well as the relative positions of these brain structures. The MEAs designed in this study provide a new detection method and tool for functional localization of PD targets and PD pathogenesis at the cellular level.

Development of A Radiomic Model for MGMT Promoter Methylation Detection in Glioblastoma Using Conventional MRI.

The methylation of the O6-methylguanine-DNA methyltransferase (MGMT) promoter is a molecular marker associated with a better response to chemotherapy in patients with glioblastoma (GB). Standard pre-operative magnetic resonance imaging (MRI) analysis is not adequate to detect MGMT promoter methylation. This study aims to evaluate whether the radiomic features extracted from multiple tumor subregions using multiparametric MRI can predict MGMT promoter methylation status in GB patients. This retrospective single-institution study included a cohort of 277 GB patients whose 3D post-contrast T1-weighted images and 3D fluid-attenuated inversion recovery (FLAIR) images were acquired using two MRI scanners. Three separate regions of interest (ROIs) showing tumor enhancement, necrosis, and FLAIR hyperintensities were manually segmented for each patient. Two machine learning algorithms (support vector machine (SVM) and random forest) were built for MGMT promoter methylation prediction from a training cohort (196 patients) and tested on a separate validation cohort (81 patients), based on a set of automatically selected radiomic features, with and without demographic variables (i.e., patients' age and sex). In the training set, SVM based on the selected radiomic features of the three separate ROIs achieved the best performances, with an average of 83.0% (standard deviation: 5.7%) for accuracy and 0.894 (0.056) for the area under the curve (AUC) computed through cross-validation. In the test set, all classification performances dropped: the best was obtained by SVM based on the selected features extracted from the whole tumor lesion constructed by merging the three ROIs, with 64.2% (95% confidence interval: 52.8-74.6%) accuracy and 0.572 (0.439-0.705) for AUC. The performances did not change when the patients' age and sex were included with the radiomic features into the models. Our study confirms the presence of a subtle association between imaging characteristics and MGMT promoter methylation status. However, further verification of the strength of this association is needed, as the low diagnostic performance obtained in this validation cohort is not sufficiently robust to allow clinically meaningful predictions.

A PAD-Based Unmanned Aerial Vehichle Route Planning Scheme for Remote Sensing in Huge Regions.

Unmanned aerial vehicles (UAVs) have been employed extensively for remote-sensing missions. However, due to their energy limitations, UAVs have a restricted flight operating time and spatial coverage, which makes remote sensing over huge regions that are out of UAV flight endurance and range challenging. PAD is an autonomous wireless charging station that might significantly increase the flying time of UAVs by recharging them in the air. In this work, we introduce PADs to simplify UAV-based remote sensing over a huge region, and then we explore the UAV route planning problem once PADs have been predeployed throughout a huge remote sensing region. A route planning scheme, named PAD-based remote sensing (PBRS), is proposed to solve the problem. The PBRS scheme first plans the UAV's round-trip routes based on the location of the PADs and divides the whole target region into multiple PAD-based subregions. Between adjacent subregions, the UAV flight subroute is planned by determining piggyback points to minimize the total time for remote sensing. We demonstrate the effectiveness of the proposed scheme by conducting several sets of simulation experiments based on the digital orthophoto model of Hutou Village in Beibei District, Chongqing, China. The results show that the PBRS scheme can achieve excellent performance in three metrics of remote sensing duration, the number of trips to charging stations, and the data-storage rate in UAV remote-sensing missions over huge regions with predeployed PADs through effective planning of UAVs.

DepressionGraph: A Two-Channel Graph Neural Network for the Diagnosis of Major Depressive Disorders Using rs-fMRI

Major depressive disorder (MDD) is a prevalent psychiatric condition with a complex and unknown pathological mechanism. Resting-state functional magnetic resonance imaging (rs-fMRI) has emerged as a valuable non-invasive technology for MDD diagnosis. By utilizing rs-fMRI data, a dynamic brain functional connection network (FCN) can be constructed to represent the complex interacting relationships of multiple brain sub-regions. Graph neural network (GNN) models have been widely employed to extract disease-associated information. The simple averaging or summation graph readout functions of GNNs may lead to a loss of critical information. This study introduces a two-channel graph neural network (DepressionGraph) that effectively aggregates more comprehensive graph information from the two channels based on the node feature number and node number. Our proposed DepressionGraph model leverages the transformer–encoder architecture to extract the relevant information from the time-series FCN. The rs-fMRI data were obtained from a cohort of 533 subjects, and the experimental data show that DepressionGraph outperforms both traditional GNNs and simple graph readout functions for the MDD diagnosis task. The introduced DepressionGraph framework demonstrates efficacy in extracting complex patterns from rs-fMRI data and exhibits promising capabilities for the precise diagnosis of complex neurological disorders. The current study acknowledges a potential gender bias due to an imbalanced gender distribution in the dataset. Future research should prioritize the development and utilization of gender-balanced datasets to mitigate this limitation and enhance the generalizability of the findings.

Intra-Inter Region Adaptive Graph Convolutional Networks for skeleton-based action recognition

To effectively capture the spatio-temporal dependencies of the skeletal data, graph convolution has been widely applied. However, it tends to emphasize the dependence relationship between adjacent joints and does not consider long-distance dependence relationships among joints. Another problem is single-structure temporal convolution, which is difficult to extract global temporal features. To address the above issues, we propose Intra-Inter Region Adaptive Graph Convolutional Networks (IIR-AGCN), which models the long-distance relationships of skeleton sequences at temporal and spatial scales. Our contributions are three-fold: First, to enhance global topological learning capabilities of graph convolution, we propose a regional-coupled attention module, which divides joint features into multiple sub-regions, and then constructs coupled relationships between intra-inter regions by self-attention mechanism, which realizes global joint information interaction. Second, to address the issue of difficulty in extracting global temporal features, we propose a pyramid multi-scale temporal module that extracts deep multi-scale temporal features and implements adaptive cross-scale feature fusion. Third, IIR-AGCN adopts a two-stream architecture, evaluating performances on two large-scale human skeleton datasets, NTU-RGB+D 60 and NTU-RGB+D 120, respectively.

Investigation of clear cell renal cell carcinoma grades using diffusion-relaxation correlation spectroscopic imaging with optimized spatial-spectrum analysis.

To differentiate high-grade from low-grade clear cell renal cell carcinoma (ccRCC) using diffusion-relaxation correlation spectroscopic imaging (DR-CSI) spectra in an equal separating analysis. Eighty patients with 86 pathologically confirmed ccRCCs who underwent DR-CSI were enrolled. Two radiologists delineated the region of interest. The spectrum was derived based on DR-CSI and was further segmented into multiple equal subregions from 2*2 to 9*9. The agreement between the 2 radiologists was assessed by the intraclass correlation coefficient (ICC). Logistic regression was used to establish the regression model for differentiation, and 5-fold cross-validation was used to evaluate its accuracy. McNemar's test was used to compare the diagnostic performance between equipartition models and the traditional parameters, including the apparent diffusion coefficient (ADC) and T2 value. The inter-reader agreement decreased as the divisions in the equipartition model increased (overall ICC ranged from 0.859 to 0.920). The accuracy increased from the 2*2 to 9*9 equipartition model (0.68 for 2*2, 0.69 for 3*3 and 4*4, 0.70 for 5*5, 0.71 for 6*6, 0.78 for 7*7, and 0.75 for 8*8 and 9*9). The equipartition models with divisions >7*7 were significantly better than ADC and T2 (vs ADC: P = .002-.008; vs T2: P = .001-.004). The equipartition method has the potential to analyse the DR-CSI spectrum and discriminate between low-grade and high-grade ccRCC. The evaluation of DR-CSI relies on prior knowledge, and how to assess the spectrum derived from DR-CSI without prior knowledge has not been well studied.

Design of the Light Source Layout Optimization Strategy Based on Region Partition and Pre-Bias Compensation for Indoor Visible Light Communication Systems

Optimization of the light source layout is an important issue for indoor visible light communication systems, as it affects the received optical power distribution and user perception. In this paper, we propose a local optimization strategy for the light source layout that balances optimization effectiveness with optimization efficiency. First, we divide the optimization region into multiple sub-regions with different sizes and optimization priorities, where the sizes and optimization priorities of the individual sub-regions are determined based on the effect minimization principle among the sub-regions. We then calculate the pre-bias factor based on the equivalent mapping, which can compensate for the effect of the light sources in the latter optimized sub-region on the source layout optimization in the current sub-region. Finally, we search for the coordinate of a single light source for each sub-region using the variance of the squared distance between the projection of the light source on the receiving plane and the received point as a fitness function. Simulation results show that the proposed optimization strategy performs well when the vertical distance between the ceiling and the receiving plane is not less than 2.85 m.

Convolutional Neural Network–Bidirectional Gated Recurrent Unit Facial Expression Recognition Method Fused with Attention Mechanism

The relationships among different subregions in facial images and their varying contributions to facial expression recognition indicate that using a fixed subregion weighting scheme would result in a substantial loss of valuable information. To address this issue, we propose a facial expression recognition network called BGA-Net, which combines bidirectional gated recurrent units (BiGRUs) with an attention mechanism. Firstly, a convolutional neural network (CNN) is employed to extract feature maps from facial images. Then, a sliding window cropping strategy is applied to divide the feature maps into multiple subregions. The BiGRUs are utilized to capture the dependencies among these subregions. Finally, an attention mechanism is employed to adaptively focus on the most discriminative regions. When evaluated on CK+, FER2013, and JAFFE datasets, our proposed method achieves promising results.