Search

Cognitive engagement involves mental and physical involvement, with observable behaviors as indicators. Automatically measuring cognitive engagement can offer valuable insights for instructors. However, object occlusion, inter-class similarity, and intra-class variance make designing an effective detection method challenging. To deal with these problems, we propose the Object-Enhanced–You Only Look Once version 8 nano (OE-YOLOv8n) model. This model employs the YOLOv8n framework with an improved Inner Minimum Point Distance Intersection over Union (IMPDIoU) Loss to detect cognitive engagement. To evaluate the proposed methodology, we construct a real-world Students' Cognitive Engagement (SCE) dataset. Extensive experiments on the self-built dataset show the superior performance of the proposed model, which improves the detection performance of the five distinct classes with a precision of 92.5%. [ABSTRACT FROM AUTHOR]

Background: Video conferencing learning offers a more accessible and flexible learning mode for students who cannot attend face‐to‐face class in person. However, research about the video conferencing learning is still lacked, especially for students' engagement in this setting. Objectives: This study adopts the self‐system model of motivational development to uncover effects of the video conferencing learning environment (VCLE) to students' engagement. Methods: Two hundred and ninety‐eight primary school students involved in the VCLE were surveyed. The partial least square structural equation modelling was employed to explore the link between the VCLE and students' engagement with a mediating role of basic psychological needs satisfaction (BPNS). Results and Conclusions: Results showed that the BPNS played a mediating role in the processes from the VCLE to engagement. Both pedagogical and social affordance had a significant effect on the BPNS. However, their influences were different. Specifically, pedagogical affordance yielded a stronger effect on the BPNS for autonomy than social affordance. The predicting power of social affordance on BPNS for relatedness and competence was stronger than pedagogical affordance. Generally, social affordance yielded more contributions to behavioral and emotional engagement than pedagogical affordance. Moreover, BPNS for competence was the only significant mediator between the VCLE and behavioral engagement, and it owned the strongest mediating power on the relationship between the VCLE and emotional engagement, followed by autonomy, and the smallest of relatedness. This study suggests that practitioners and researchers should give more consideration to improve social affordance of the VCLE and to meet students' BPNS especially for competence in order to design and implement an engaged video conferencing course. Lay Description: What is already known about this topic: Video conferencing learning provides an accessible and flexible learning mode.Mechanism from learning environment to engagement is helpful to design an engaged video conferencing course.Less is known about how video conferencing learning affects students' engagement. What the paper adds: This study uncovered the functioning process of video conferencing learning to students' engagement.PLS‐SEM was employed to verify the mediating role of BPNS in this process.BPNS for autonomy, competence, and relatedness played mediating roles and their powers were different. Implications for practice and/or policy: More attention is needed to improve to social affordance to enhance students' emotional engagement.BPNS especially for competence deserves special consideration to improve students' behavioral and emotional engagement. [ABSTRACT FROM AUTHOR]

Detecting infrared (IR) small targets effectively and robustly is crucial for the tasks such as infrared searching and guarding. While methods based on the human vision system (HVS) have achieved great success in this field, detecting dim targets in complex backgrounds remains a challenge due to the multi-scale framework and over-simplified disparity calculations. In this paper, infrared small targets are detected with a novel local contrast measurement named double-layer patch-based contrast (DLPC). Firstly, we crafted an elaborated double-layer local contrast measure, to suppress the background, which can accurately measure the gray difference between the target and its surrounding complex background. Secondly, we calculated the absolute value of the grayscale difference between the target and the background in the diagonal directions as a weighting factor to further enhance the target. Then, an adaptive threshold on the DLPC was employed to extract the target from the IR image. The proposed method can detect small targets effectively with a fixed-scaled mask template while being computationally efficient. Experimental results in terms of background suppression factor (BSF), signal-to-clutter ratio gain (SCRG) and receiver operating characteristic (ROC) curve on five IR image datasets demonstrated that the proposed method has better detection performance compared to six state-of-the-art methods and is more robust in addressing complex backgrounds. [ABSTRACT FROM AUTHOR]

Learning through making enhances the learning experience and effectively develops learners' creativity. In this regard, maker education has been extensively incorporated into school education as a vital learning environment. However, less is known about how making advances students' creativity in maker education. Therefore, this study aimed to explore relationships among making, tinkering, and creating. A total of 372 middle school students enrolled in maker courses through the 2021–2022 school year were surveyed. They completed a questionnaire concerning learning engagement, individual creativity, collective creativity, and learning motivation. Structural equation modeling (SEM) was employed to test the hypotheses proposed in this study. The results revealed significant relationships among learning engagement, individual creativity, and collective creativity. Specifically, behavioral engagement positively predicted emotional and cognitive engagement. Moreover, both emotional and cognitive engagement positively related to collective creativity, while only the effect of cognitive engagement on individual creativity was observed. The direct effects of behavioral engagement on individual and collective creativity were not significant. Furthermore, emotional and cognitive engagement mediated the association between behavioral engagement and collective creativity, while only cognitive engagement played a mediating role between behavioral engagement and individual creativity. In addition, the moderating analysis showed that learning motivation moderated the association between behavioral engagement and emotional engagement in school-maker education. Some implications for integrating maker education into school education and cultivating students' creativity were discussed. [ABSTRACT FROM AUTHOR]

Hyperspectral image (HSI) classification is a hot topic in the remote sensing community. However, it is challenging to fully use spatial–spectral information for HSI classification due to the high dimensionality of the data, high intraclass variability, and the limited availability of training samples. To deal with these issues, we propose a novel feature extraction method called nonlocal correntropy matrix (NLCM) representation in this letter. NLCM can characterize the spectral correlation and effectively extract discriminative features for HSI classification. We verify the effectiveness of the proposed method on two widely used datasets. The results show that NLCM performs better than the state-of-the-art methods, especially when the training set size is small. Furthermore, the experimental results also demonstrate that the proposed method outperforms compared methods significantly when the land covers are complex and with irregular distributions. [ABSTRACT FROM AUTHOR]

This paper proposes a novel architecture, termed multiscale principle of relevant information (MPRI), to learn discriminative spectral-spatial features for hyperspectral image classification. MPRI inherits the merits of the principle of relevant information (PRI) to effectively extract multiscale information embedded in the given data, and also takes advantage of the multilayer structure to learn representations in a coarse-to-fine manner. Specifically, MPRI performs spectral-spatial pixel characterization (using PRI) and feature dimensionality reduction (using regularized linear discriminant analysis) iteratively and successively. Extensive experiments on three benchmark data sets demonstrate that MPRI outperforms existing state-of-the-art methods (including deep learning based ones) qualitatively and quantitatively, especially in the scenario of limited training samples. Code of MPRI is available at http://bit.ly/MPRI_HSI. [ABSTRACT FROM AUTHOR]

Due to the impact of the COVID-19 pandemic, many students are unable to attend face-to-face courses, Therefore, in this case, distance education should be promoted to replace face-to-face education. However, because of the imbalance of education in different regions, such as the imbalance of education resources between rural and urban areas, the quality of distance education may not be guaranteed. Therefore, in China and some regions, there have been efforts made to carry out blended synchronous classroom attempts. In hybrid synchronous classroom situations, teachers' workloads have increased, and it is difficult to fully understand students' learning efficiency and class participation. We use deep learning to identify the behaviors of teachers and students in a blended synchronous classroom-based situation, aiming to automate the analysis of classroom videos, which can help teachers in classroom reflection and summary in a blended synchronous classroom or face-to-face classroom. In the behavior recognition of students and teachers, we combine the head, hand, and body posture information of teachers and students and add the feature pyramid (FPN) and convolutional block attention module (CBAM) for comparative experiments. Finally, S–T (student–teacher) analysis and engagement analysis were carried out on the identification results. [ABSTRACT FROM AUTHOR]

Detecting small targets from infrared remote sensing images is still a challenging task. In this research, we propose a multidirectional local difference measure weighted by entropy (MDLDE) to detect small targets from infrared images with messy backgrounds. First, a new multidirectional local difference measure is proposed to suppress the clutter background. Then, the entropy, which captures the overall heterogeneity between the target and the background, is utilized to enhance the target. Lastly, an adaptive threshold was adopted to segment the target region from the background. The designed MDLDE could effectively enhance the target and simultaneously suppress the background clutter. Experimental results on six datasets indicate that the proposed method outperformed other state-of-the-art methods in terms of the signal-to-clutter ratio gain (SCRG), background suppression factor (BSF), and receiver operating characteristic (ROC) curves. [ABSTRACT FROM AUTHOR]

Background: Online learning is currently adopted by educational institutions worldwide to provide students with ongoing education during the COVID-19 pandemic. However, online learning has seen students lose interest and become anxious, which affects learning performance and leads to dropout. Thus, measuring students' engagement in online learning has become imperative. It is challenging to recognize online learning engagement due to the lack of effective recognition methods and publicly accessible datasets. Methods: This study gathered a large number of online learning videos of students at a normal university. Engagement cues were used to annotate the dataset, which was constructed with three levels of engagement: low engagement, engagement, and high engagement. Then, we introduced a bi-directional long-term recurrent convolutional network (BiLRCN) for online learning engagement recognition in video. Result: An online learning engagement dataset has been constructed. We evaluated six methods using precision and recall, where BiLRCN obtained the best performance. Conclusions: Both category balance and category similarity of the data affect the performance of the results; it is more appropriate to consider learning engagement as a process-based evaluation; learning engagement can provide intervention strategies for teachers from a variety of perspectives and is associated with learning performance. Dataset construction and deep learning methods need to be improved, and learning data management also deserves attention. [ABSTRACT FROM AUTHOR]

(1) Background: To evaluate the efficacy and safety of combined surgery (limited pars plana vitrectomy, anterior-chamber stabilized phacoemulsification, IOL implantation and posterior capsulotomy, LPPV + ACSP + IOL + PC) in complex nanophthalmos. (2) Methods: Patients with complex nanophthalmos were recruited to undergo LPPV + ACSP + IOL + PC from January 2017 to February 2021. Preoperative and post-operative intraocular pressure (IOP), best corrected visual acuity (BCVA), anterior chamber depth (ACD), and number of glaucoma medications were compared using the paired t-test or Wilcoxon signed rank sum tests. Surgical success rate was evaluated. Surgery-associated complications were documented. (3) Results: Forty-five eyes of 37 patients with complex nanophthalmos were enrolled. The mean follow-up period was 21.7 ± 10.6 months after surgery. Mean IOP decreased from 32.7 ± 8.7 mmHg before surgery to 16.9 ± 4.5 mmHg (p < 0.001) at the final follow-up visit, mean logMAR BCVA improved from 1.28 ± 0.64 to 0.96 ± 0.44 (p < 0.001), mean ACD significantly increased from 1.14 ± 0.51 mm to 3.07 ± 0.66 mm (p < 0.001), and the median number of glaucoma medications dropped from 3 (1, 4) to 2 (0, 4) (p < 0.001). The success rate was 88.9% (40 eyes) at the final follow-up visit. Two eyes had localized choroidal detachments which resolved with medical treatment. (4) Conclusions: LPPV + ACSP + IOL + PC is a safe and effective surgical procedure, which can decrease IOP, improve BCVA, deepen the anterior chamber, and reduce the number of glaucoma medications in patients with complex nanophthalmos. It can be considered as one of the first treatment in nanophthalmic eyes with complex conditions. [ABSTRACT FROM AUTHOR]

Engagement plays an essential role in the learning process. Recognition of learning engagement in the classroom helps us understand the student's learning state and optimize the teaching and study processes. Traditional recognition methods such as self-report and teacher observation are time-consuming and obtrusive to satisfy the needs of large-scale classrooms. With the development of big data analysis and artificial intelligence, applying intelligent methods such as deep learning to recognize learning engagement has become the research hotspot in education. In this paper, based on non-invasive classroom videos, first, a multi-cues classroom learning engagement database was constructed. Then, we introduced the power IoU loss function to You Only Look Once version 5 (YOLOv5) to detect the students and obtained a precision of 95.4%. Finally, we designed a bimodal learning engagement recognition method based on ResNet50 and CoAtNet. Our proposed bimodal learning engagement method obtained an accuracy of 93.94% using the KNN classifier. The experimental results confirmed that the proposed method outperforms most state-of-the-art techniques. [ABSTRACT FROM AUTHOR]

With the development of the hyperspectral imaging technique, hyperspectral image (HSI) classification is receiving more and more attention. However, due to high dimensionality, limited or unbalanced training samples, spectral variability, and mixing pixels, it is challenging to achieve satisfactory performance for HSI classification. In order to overcome these challenges, this paper proposes a feature extraction method called spatial perception correntropy matrix (SPCM), which makes use of spatial and spectral correlation simultaneously to improve the classification accuracy and robustness. Specifically, the dimension reduction is carried out firstly. Then, the spatial perception method is designed to select the local neighbour pixels. Thus, local spectral-spatial correlation is characterized by the correntropy matrix constructed using the selected neighbourhoods. Finally, SPCM representations are fed into the support vector machine for classification. The extensive experiments carried out on three widely used data sets have revealed that the proposed SPCM performs better than several state-of-the-art methods, especially when the training set is small. [ABSTRACT FROM AUTHOR]

Purpose: To identify biomarkers associated with CD8+ T cells in coronary artery disease (CAD) and initially explore their potential role in the tumor immune microenvironment. Materials and Methods: CAD-related datasets GSE12288, GSE34198, and GSE66360, were downloaded from the GEO database. First, GSVA was performed based on the GSE12288 dataset. Then WGCNA analysis was performed to identify the most relevant module and candidate hub gene for CD8+ T cells, followed by GO and KEGG analysis of this module. Secondly, the relationship between candidate hub genes and CD8+ T cells was verified using GSE34198 and GSE66360, which led to the identification of hub genes. The relationship of hub genes with CD8+ T cells in cancer was analyzed using the TIMER database. Methylation analysis of hub genes was performed using the DiseaseMeth database. CAD, pan-cancer, pan-cell lines, and pan-normal tissues, correlations between hub genes. In addition, potential drugs and TFs associated with hub genes were predicted, and the ceRNA network was constructed. Finally, GSEA was performed separately for hub genes. Results: CAD was shown to be associated with immune response by GSVA analysis. WGCNA identified the blue module as most related to CD8+ T cells and identified nine candidate hub genes. The relevance of CAD to immunity was further confirmed by GO and KEGG analysis of the module. Two additional datasets validated and identified three hub genes (FBXO7, RAD23A, and MKRN1) that significantly correlated with CD8+ T cells. In addition, we found that hub genes were positively associated with CD8+ T cells in TGCT, THCA, and KICH cancers by our analysis. Moreover, the hub gene was differentially methylated. We also analyzed the correlation between hub genes in CAD, different cancers, different cell lines, and different normal tissues. The results of all the analyses showed a positive correlation between them. Finally, we successfully constructed hub gene-associated TF-gene and ceRNA networks and predicted 11 drugs associated with hub genes. GSEA suggests that hub genes are related to multiple immune response processes. Conclusion: FBXO7, RAD23A, and MKRN1 are significantly associated with CD8+ T cells in CAD and multiple cancers and may act through immune responses in CAD and cancer. [ABSTRACT FROM AUTHOR]

The hyperspectral images (HSIs) classification technique has received widespread attention in the field of remote sensing. However, how to achieve satisfactory classification performance in the presence of a large amount of noise is still a problem worthy of consideration. In this article, a local correntropy matrix (LCEM)-based spatial–spectral feature representation method is proposed for HSI classification. Motivated by the successful application of information-theoretic learning (ITL), we propose to adopt correntropy matrix to represent the spatial–spectral features of HSI. Specifically, the dimension reduction is first performed on the original hyperspectral data. Then, for each pixel, we select its local neighbors within a sliding window using cosine distance for the construction of the LCEM. In this way, each pixel can be characterized as an LCEM. Finally, all the correntropy matrices are fed into a support vector machine (SVM) for final classification. In addition, we also propose a novel way to determine the size of the local window based on standard deviation. Because the LCEM as the feature descriptor can characterize discriminative spatial–spectral features, the proposed method has shown great interclass separability and intraclass compactness. Compared with other advanced approaches, the proposed LCEM method has achieved competitive performance in both evaluation indexes and visual effects, especially when the training size is very small. [ABSTRACT FROM AUTHOR]

Introduction: To introduce a modified technique for primary/secondary intraocular lens (IOL) fixation without corneal incision in vitrectomized eyes. Methods: Consecutive case series who had undergone previous or concomitant pars plana vitrectomy (PPV) to have primary/secondary IOL fixation were prospectively included. A self-sealing scleral incision was made underneath the superior scleral flap, through which the IOL was inserted into the anterior chamber. The suture tied with the IOL passed through the sclera to fix the IOL in the ciliary sulcus. Patients were followed up for at least 3 months. Main outcomes were best corrected visual acuity (BCVA), intraocular pressure (IOP), surgically induced astigmatism (SIA), and intraoperative and postoperative complications. Results: A total of 31 patients were included in the study. The mean follow-up time was 5.35 ± 4.14 months. The BCVA (log MAR unit) improved from 0.97 ± 0.58 preoperative to 0.42 ± 0.36 postoperative (P < 0.001). Mean IOP remained unchanged (preoperative IOP 14.03 ± 2.90 mmHg, postoperative IOP 13.26 ± 3.46 mmHg, P = 0.130). The mean SIA was 0.91 ± 0.76 diopters. No obvious intraoperative and postoperative complications were observed. Conclusion: This method has favorable postoperative visual recovery and IOP control. This modified method could be taken into account as an option by surgeons in vitreoretinal surgery. [ABSTRACT FROM AUTHOR]

Purpose: Because of the direct contact of intravitreal silicone oil (SO) with the subcapsular membrane, cataract is one of the main SO-related complications. In a group of patients, condense subcapsular opacification occurs, which adds difficulty and risk when having sequential treatment of it. The aim of the current study is to assess the long-term outcomes of pars plana subcapsulotomy to remove condense subcapsular opacification in combined surgery of SO removal and phacoemulsification. Methods: Retrospective cohort study. Consecutive patients who were scheduled to have combined surgery of SO removal and phacoemulsification, and with condense subcapsular opacification were included. After phacoemulsification and SO removal, circular subcapsulotomy (diameter = 3–5 mm) was performed with a 23-/25-gauge vitrectomy probe on each subject during the combined surgery. Main outcomes were pre- and postoperative best-corrected visual acuity (BCVA), intra- and postoperative complications. Results: One hundred and twenty patients (120 eyes) were included. Postoperative logMAR BCVA at day 1, week 1, month 1, and final follow-up examinations was 1.0 ± 0.5, 0.7 ± 0.4, 0.6 ± 0.4, and 0.6 ± 0.3, respectively. Statistically significant median differences of logMAR BCVA occurred between the preoperative examination and each postoperative follow-up examination (all p < 0.001). The sharpest median increase in logMAR BCVA occurred between the day 1 and week 1 postoperative examinations (p < 0.001). Conclusions: For condense subcapsular opacification caused by SO tamponade, pars plana subcapsulotomy with a 23-/25-gauge vitrectomy probe during combined surgery of SO removal and phacoemulsification is effective and safe to have surgical management of it. The systemic approach enables patients to experience rapid and long-lasting vision rehabilitation in a single procedure. [ABSTRACT FROM AUTHOR]

A new spectral‐spatial hyperspectral image (HSI) classification method called hierarchical broad learning system (HBLS) has been proposed in this paper. Specifically, it combines wavelet, broad learning system (BLS) and Gabor filters into a hierarchical structure. First of all, wavelet is used to reduce the observation noise of HSIs. Then BLS is adopted to acquire a set of pixelwise probability maps from the input data, and Gabor filters are used to explore spatial information by refining these probability maps. These two operations (BLS and Gabor filtering) are alternated to form a hierarchical architecture. And the discriminative spectral‐spatial features can be extracted at each layer of the hierarchical architecture. Finally, the spectral‐spatial features are fed into the standard BLS for classification. Experimental results on three widely used HSIs reveal that HBLS outperforms some state‐of‐the‐art methods in terms of classification accuracy and sample complexity. [ABSTRACT FROM AUTHOR]

Considering that multiple features can improve the classification performance as they contain diversity information of images, a multiple-feature latent space learning-based method is proposed for hyperspectral image (HSI) classification in this letter. In the proposed method, a latent space that contains diversity information of multiple features and transformation matrices between the latent space and features are both learned. Moreover, spatial information is used for labeling unlabeled samples in the classification. Experimental results on the Indian Pines and University of Pavia data sets demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

In this paper, we propose a new method for hyperspectral images (HSI) classification, aiming to take advantage of both manifold learning-based feature extraction and neural networks by stacking layers applying locality sensitive discriminant analysis (LSDA) to broad learning system (BLS). BLS has been proven to be a successful model for various machine learning tasks due to its high feature representative capacity introduced by numerous randomly mapped features. However, it also produces redundancy, which is indiscriminate and finally lowers its performance and causes heavy computing demand, especially in cases of the input data bearing high dimensionality. In our work, a manifold learning method is integrated into the BLS by inserting two LSDA layers before the input layer and output layer separate, so the spectral-spatial HSI features are fully utilized to acquire the state-of-the-art classification accuracy. The extensive experiments have shown our method's superiority. [ABSTRACT FROM AUTHOR]

Uveal melanoma (UM) is the most frequent primary ocular tumour among adults. Here, we aimed to establish the immune cell-based signature to predict the overall survival (OS) of UM patients. The mRNA profile and matched clinical records of 80 UM patients were downloaded from The Cancer Genome Atlas (TCGA) database. CIBERSORT was used to verify the immune cell types of individuals. The univariate analysis found the CD8+ T cell, monocyte, CD4+ memory T cell (resting) and mast cell (resting) were significantly associated with the OS of UM patients. Subsequently, the LASSO Cox regression test was applied to establish the signature, by which the patients were separated into high- and low-risk subgroups. The Kaplan–Meier analyses found for these patients in the high-risk group had a poor survival rate than those in the low-risk group. The predictive value and stability were confirmed by the receiver operative characteristics curves. Pathway analyses found that the differentially expressed genes between the high- and low-risk subgroups were mainly centralised on immune response-related pathways. Further, the comparison of our signature with clinicopathological records confirmed its superiority and independence. In summary, we established an immune cell-based prognosis-predicting signature for UM patients, which will benefit the individual's treatment. [ABSTRACT FROM AUTHOR]

Teaching reflection based on videos is the main method in teacher education and professional development. However, it takes a long time to analyse videos, and teachers are easy to fall into the state of information overload. With the development of "AI + education," automatic recognition of teacher behavior to support teaching reflection has become an important research topic. In this paper, taking online open classroom teaching video as the data source, we collected and constructed a teacher behavior dataset. Using this dataset, we explored the behavior recognition methods based on RGB video and skeleton information, and the information fusion between them is carried out to improve the recognition accuracy. The experimental results show that the fusion of RGB information and skeleton information can improve the recognition accuracy, and the early-fusion effect is better than the late-fusion effect. This study helps to solve the problems of time-consumption and information overload in teaching reflection and then helps teachers to optimize the teaching strategies and improve the teaching efficiency. [ABSTRACT FROM AUTHOR]

This article proposes a functional data discriminant analysis (FDDA) method for hyperspectral image (HSI) classification. This method analyzes and processes the HSI data from a functional point of view, which is a novel perspective in HSI processing. The classical methods achieve dimensionality reduction by directly eliminating the redundancy of the HSI data. However, the proposed method extracts the functional features by utilizing the redundancy of the HSI data. Functional features can effectively reveal inherent characteristics of the HSI data with the change in the wavelengths. Based on this, a regularized weighted fitting model is first built for converting a spectral vector into a spectral curve. Second, an FDDA method defined in the function field is presented for extracting the functional features of the spectral curves. Finally, a novel spectral–spatial framework is designed for classification tasks of HSI data sets. Experimental results in three commonly used HSI data sets indicate that the proposed method is effective and leads to promising classification results compared with some benchmarking methods. More importantly, the work tries to diversify and develop the existing theory and methods of HSI classification from discrete (vector) data learning methods to continuous (functional) data learning methods. [ABSTRACT FROM AUTHOR]

Human activity recognition using depth videos remains a challenging problem while in some applications the available training samples is limited. In this article, we propose a new method for human activity recognition by crafting an integrated descriptor called multi-level fused features for depth sequences and devising a fast broad learning system based on matrix decomposition for classification. First, the surface normals are computed from original depth maps; the histogram of the surface normal orientations is obtained as a low-level feature by accumulating the contributions from normals, then a high-level feature is acquired by sparse coding and pooling on the aggregation of polynormals. After that, the principal component analysis is applied to the conjunction of the two-level features in order to obtain a low-dimensional and discriminative fused feature. At last, fast broad learning system based on matrix decomposition is proposed to accelerate the training process and enhance the classification results. The recognition results on three benchmark data sets show that our method outperforms the state-of-the-art methods in term of accuracy, especially when the number of training samples is small. [ABSTRACT FROM AUTHOR]

Purpose: Dickkopf 1 (DKK1) functions as a natural antagonist of the canonical Wnt/β-catenin pathway. The purpose of this study was to examine the expression of DKK1 in vitreous samples of patients with pathological myopia, in order to search for possible correlations between DKK1 and axial length. Materials and Methods: The expression of DKK1 and other cytokines in vitreous samples of 44 non-myopic eyes, 42 eyes with low-to-moderate myopia, and 51 eyes with pathological myopia were examined using multiplex cytokine detection technology. Ophthalmologic characteristics, including axial length and subfoveal choroidal thickness, were clinically measured for further analysis. Results: The intravitreous levels of DKK1 (P <.0001) were markedly higher in the pathological myopia group than in the control group. There were no differences of DKK1 levels in different vitreoretinal conditions. Additionally, we found that the DKK1 levels were positively correlated with HGF (β = 0.268, P =.032), and TIMP-3 (β = 0.209, P =.047) levels, as well as with axial length (β = 0.714, P <.0001) in the pathological myopia group. Conclusions: Elevated levels of DKK1 were found in the eyes with elongated axial length. [ABSTRACT FROM AUTHOR]

The present study assessed changes in macular capillary vessel density and thickness associated with ocular silicone oil tamponade (SOT) following complex vitreoretinal surgery. A total of 23 patients who underwent pars plana vitrectomy combined with intravitreal SOT were retrospectively enrolled, and a single eye was assessed in each participant. Furthermore, 20 patients who required silicone oil removal (SOR) were included. The macular capillary vessel density and the foveal avascular zone (FAZ) area were analyzed using optical coherence tomography angiography (OCTA) and retinal thickness was assessed by OCT. The results demonstrated that the macular capillary vessel density of the superficial capillary plexus (SCP) and deep capillary plexus (DCP), the FAZ area and full retinal thickness were retained at a stable level following SOT (P>0.05), and also following SOR (P>0.05). Furthermore, the parafoveal (P=0.008), superior-hemi (P=0.007), temporal (P=0.015), superior (P=0.028) and nasal (P=0.002) inner retinal thickness decreased in the SOT group, whereas the inner retinal thickness was unaltered after SOR vs. baseline (P>0.05). In conclusion, silicone oil has no significant effect on macular capillary vessel density within a 6-month period but may compress and reduce the thickness of the inner retina. [ABSTRACT FROM AUTHOR]

Statement: The current article has not been published elsewhere and has not been submitted simultaneously for publication elsewhere. Purpose: To investigate the preliminary use of three-dimensional (3D) heads-up display (HUD) viewing system for vitreoretinal surgery under various status. Materials and Methods: Nonrandomized case-control study. Consecutive cases to have vitreoretinal surgery under various status were prospectively recruited. Twenty-five-gauge vitrectomy platform and 3D viewing system were used. Main outcomes included: luminous emittance (lux) of endoillumination pipe, surgical duration, the surgeon and residents' preference and ergonomics. Consecutive patients to have vitreoretinal surgery with the conventional viewing system were recruited as control group following the same inclusion and exclusion criteria and underwent surgeries by the same surgeon with the same microscope and vitrectomy platform. Results: Thirty-one patients (31 eyes; Group Study) and twenty-eight patients (28 eyes; Group Control) were included; without significantly statistical difference in terms of age, gender, main diagnosis, surgical duration, and difficulty rating between both groups (all P > 0.05). Lower endoillumination intensity was needed in Group Study than that in Group Control (10% vs. 35%; 598.7 ± 5.4 vs. 1913.0 ± 12.9 lux, P < 0.001). The surgeon and residents expressed overwhelming preference with the 3D system in both groups. Improved ergonomic was rated in Group Study (4.4 ± 0.8 vs. 3.2 ± 1.0, P < 0.001). Some intraoperative difficulties and discomforts appeared to the surgeon and assistants when using the 3D viewing system. Conclusion: Vitreoretinal surgery under various status can be well finished with the HUD platform by novice at the system. Main benefits included lower endoillumination intensity, enhanced users' preference, and improved ergonomics. Some further refinements of the system are expected. [ABSTRACT FROM AUTHOR]

Observing student body gesture has been widely used to assess teaching effectiveness over the past few decades. However, manual observation is not suitable for the automatic data analysis in the field of learning analytics. Consequently, a student body gesture recognition method based on Fisher Broad Learning System (FBLS) and Local Log-Euclidean Multivariate Gaussian (L2EMG) is proposed in this paper. FBLS is designed by introducing the discriminative information into the hidden layer of Broad Learning System (BLS) and reducing the dimensionality of hidden-layer representations. FBLS has superiorities in accuracy and speed. In addition, L2EMG, which is a highly distinctive descriptor, characterizes the local image with a multivariate Gaussian distribution. So L2EMG features are fed into the FBLS for recognition in this paper. Extensive experimental results on self-built dataset show that the proposed student body gesture recognition method obtains better results than other benchmarking methods. [ABSTRACT FROM AUTHOR]

Introduction: Rhegmatogenous retinal detachment (RRD) with extensive choroidal detachment (CD), suprachoroidal fluid (SCF), and lower intraocular pressure (IOP) markedly increase the difficulty of pars plana vitrectomy (PPV) because it makes the insertion and placement of the trocar cannulas for PPV time-consuming and technically challenging. In this study, we developed a systemic approach that avoids the risks of sclerotomy or using a longer trocar to help in the safe placement of the trocar cannulas in order to prevent inadvertent fluid/air perfusion during PPV in RRD eyes with extensive CD, SCF, and hypotony.Methods: This is a prospective case series study. Consecutive patients diagnosed with RRD with extensive CD, SCF, and hypotony were recruited. The main steps included injecting transconjunctival intravitreal fluid to increase IOP and enable the passive egress of SCF into the subconjunctival space, creating and broadening the inner opening for the trocar cannulas, transillumination with an endoillumination light pipe, and indentation of the infusion cannula to create maximum visual exposure of the inner segment and help peripheral shaving around the infusion tip.Main Outcomes: the success rate of the infusion placement into the vitreous cavity, surgical duration, intra- and postoperative complications, and functional and anatomical outcomes.Results: Thirty-eight patients (38 eyes) were included. The first-time success rate of cannula placement into the vitreous cavity was 100%, with a mean surgical duration of 8.5 ± 1.8 min. No intra- or postoperative complications related to the approach were observed. Significant vision improvement was noted in 68.4% (26/38) of the patients at the final follow-up. The primary and final retinal reattachment rates were 94.7% (36/38) and 100% (38/38), respectively.Conclusion: This systemic approach is a feasible option to ensure the safe placement of trocar cannulas and prevent inadvertent perfusion during small-gauge PPV in RRD eyes with extensive CD, SCF, and hypotony, without the need for a draining sclerotomy, a longer trocar, or a sharp clearing of the cannula tip. [ABSTRACT FROM AUTHOR]

Small-gauge vitrectomy has become popular due to its notable advantages, including less trauma, shortened convalescence and improved manoeuvrability. The aim of the present study was to compare the surgical outcomes of 27-gauge (27-G) vitrectomy with those of 25-gauge (25-G) vitrectomy in the management of proliferative diabetic retinopathy (PDR) with preoperative intravitreal injection of conbercept. The data of 48 consecutive patients with PDR (48 eyes) were retrospectively collected. The patients underwent conbercept intravitreal injection and pars plana vitrectomy with a 27-G group (23 eyes) or 25-G group (25 eyes) vitrectomy system. The operating time, suturing rate, endodiathermy rate, postoperative best-corrected visual acuity (BCVA), intraocular pressure (IOP) and complications were recorded. The mean postoperative BCVA at final follow-up was significantly improved compared with that at the baseline in both groups (P<0.001 for both). The differences in the mean BCVA changes between the two groups were not significant (P>0.99), and no differences were observed in the final central foveal thickness (P=0.51) between the two groups. The final IOP remained stable compared with that at the baseline in the 27-G group (P=0.36) and the 25-G group (P=0.05). The suturing rate was significantly decreased in the 27-G group compared with the 25-G group (P=0.04). There were no significant differences between the two groups in terms of the operating time (P=0.18), rate of endodiathermy use (P>0.99), iatrogenic retinal breaks (P=0.42) or postoperative recurrent vitreous haemorrhage (P>0.99). In addition, no case of ocular hypotony was observed in either group. In conclusion, 27-G vitrectomy was as efficient and safe as 25-G vitrectomy in the management of PDR in terms of operating time and complications. With reference to the literature, preoperative conbercept injection appears to assist in decreasing the incidence of intraoperative and postoperative complications. [ABSTRACT FROM AUTHOR]

Purpose: To investigate the use of 27-gauge pars plana vitrectomy (PPV) with short-term tamponade of perfluorocarbon liquid (PFCL) for repair of giant retinal tears (GRT).Methods: Retrospective case series study. Consecutive patients with GRT were treated with 27-gauge PPV and short-term tamponade of PFCL for 7-10 days. PFCL was completely removed with a secondary surgery.Results: Twenty-three eyes of 23 patients were included. All the patients achieved primary anatomic success. Preoperative logarithm of the minimum angle of resolution (logMAR) BCVA was 1.59 ± 0.58 (Snellen 20/778). Five (21.7%), nine (39.1%), eleven (47.8%) and fifteen (65.2%) eyes experienced vision improvement at the Month-1, Month-3, Month-6 and final follow-ups, respectively. The final logMAR BCVA was 0.84 ± 0.51 (Snellen 20/138), being statistically better than the preoperative one (P < 0.001). Surgical complications included foreign body response (n = 7), transient elevated intraocular pressure (n = 5), cataract formation/deterioration (n = 11) and posterior capsule opacity (n = 16).Conclusions: 27-Gauge PPV with short-term tamponade of PFCL is safe and effective for the repair of GRT. Side effects of the surgery mainly included foreign body response, transient elevated intraocular pressure, cataract formation/deterioration and posterior capsule opacity. [ABSTRACT FROM AUTHOR]

Background: A detailed analysis of the tomographic thickness of intraretinal layers may provide more information on neurodegeneration in patients with mild cognitive impairment (MCI) and Alzheimer's disease (AD).Objective: The goal was to analyze tomographic thickness patterns of intraretinal layers in patients with AD andMCI.Method: Forty-nine patients (25 AD and 24 MCI) and 21 cognitively normal (CN) controls were imaged using ultra-high-resolution optical coherence tomography to obtain volumetric data centered on the fovea. The segmented intraretinal layers were retinal nerve fiber layer (RNFL), ganglion cell- inner plexiform layer (GCIPL), inner nuclear layer (INL), outer nuclear layer (ONL), outer plexiform layer (OPL), and retinal photoreceptor (PR), in addition to the total retinal thickness(TRT).Results: The thickness differences were negative (thinning) mainly in TRT, RNFL, and GCIPL in both AD and MCI groups in comparison to CN, while the thickness differences were positive (thickening) mainly in ONL and PR in AD. GCIPL of AD and MCI was thinner in superior, nasal superior, and temporal superior quadrants, compared to CN (p < 0.05). GCIPL of the inner superior, inner nasal superior, inner temporal superior, and outer nasal superior sectors was significantly thinner in AD than CN (p < 0.05). GCIPL of the outer superior, inner temporal superior, outer nasal, and temporal superior sectors was significantly thinner in MCI than CN (p < 0.05).Conclusion: Focal thinning of the GCIPL was visualized and quantified by detailed partitions in AD and MCI, which provides specific information about neurodegeneration in MCI and AD. [ABSTRACT FROM AUTHOR]

Purpose. To determine the prevalence of peripheral retinopathy and its associated risk factors among a sample of Guangzhou office computer workers. Methods. A cross-sectional study of Guangzhou Chinese computer workstations and operators in different departments and units of the Guangzhou Power Supply Bureau, China, in 2016. Peripheral retinopathy was recorded and analyzed using a scanning laser ophthalmoscope (SLO; Optos, Daytona, United Kingdom) and slit-lamp microscopy combined with a three-mirror contact lens. Results. The 1934 eyes of 967 subjects (513 females and 454 males) were included in this study. In total, 79.1% of the eyes were myopic in workers aged 20–29 years, 72.9% in workers aged 30–39 years, 62.2% in workers aged 40–49 years, and 43.4% in workers aged 50–59 years (p<0.001). Most eyes had optic nerve crescents (81.3%). Various peripheral degenerations were found: 7 eyes (0.4%) had microcystoid degeneration, 40 (2.1%) had peripheral pigmentary degeneration, 87 (4.5%) had lattice degeneration, and 4 (0.2%) had snail-track degeneration. Nineteen (1.0%) eyes had paving-stone degeneration, 11 (0.6%) eyes had a retinal hole or tear, and 16 (0.8%) eyes had chorioretinal degeneration. Multivariate regression confirmed that greater axial length (OR: 1.18 (1.03, 1.35), p=0.012) and more serious spherical equivalent (OR: 0.82 (0.77, 0.88), p<0.001) were significant risk factors for peripheral retinal changes. Conclusion. Peripheral retinal degenerative changes were found in a larger proportion of younger computer workers than older ones. Myopia is occurring in younger and younger people, accompanied by peripheral retinal degeneration. [ABSTRACT FROM AUTHOR]

The goal of this study was to determine the retinal blood flow rate (BFR) and blood flow velocity (BFV) of pre-capillary arterioles and post-capillary venules in patients with mild cognitive impairment (MCI) and Alzheimer’s disease (AD). Forty patients (20 AD and 20 MCI) and 21 cognitively normal (CN) controls with a similar age range (± 5 yrs) were recruited. A retinal function imager (RFI) was used to measure BFRs and BFVs of arterioles and venules in the macular region. The thickness of the ganglion cell-inner plexiform layer (GCIPL) was measured using Zeiss Cirrus optical coherence tomography. Macular BFRs in AD group were 2.64 ± 0.20 nl/s (mean ± standard deviation) in arterioles and 2.23 ± 0.19 nl/s in venules, which were significantly lower than in MCI and CN groups (P < 0.05). In addition, BFRs in MCI were lower than in CN in both arterioles and venules (P < 0.05). The BFV of the arterioles was 3.20 ± 1.07 mm/s in AD patients, which was significantly lower than in CN controls (3.91 ± 0.77 mm/s, P = 0.01). The thicknesses of GCIPL in patients with AD and MCI were significantly lower than in CN controls (P < 0.05). Neither BFV nor BFR in arterioles and venules was related to age, GCIPL thickness, mini mental state examination (MMSE) score and disease duration in patients with AD and MCI (P > 0.05). The lower BFR in both arterioles and venules in AD and MCI patients together with the loss of GCIPL were evident, indicating the impairment of the two components in the neurovascular-hemodynamic system, which may play a role in disease progression. [ABSTRACT FROM AUTHOR]

Robust and efficient infrared (IR) small target detection plays an important role in image processing for IR remote sensing. In order to detect the IR small target with high detection rate, low false alarm rate (FAR), and high detection speed, a novel method called high-boost-based multiscale local contrast measure (HB-MLCM) is proposed in this letter. First, improved high boost filter is proposed to enhance the high frequency signal where the target may appear and suppress the low frequency signal. Then, a simple MLCM is proposed for further enhancing the target and suppressing the background. Finally, a simple and adaptive thresholding method is used to segment targets from the contrast map. Experimental results on three real image sequences with various typical complex backgrounds demonstrate that the proposed method can effectively detect the target with faster speed, higher detection rate, and lower FAR compared with the state-of-the-art methods. [ABSTRACT FROM PUBLISHER]

Background: To investigate the anatomic and functional outcomes of pars plana vitrectomy (PPV) with partial tamponade of filtered air for rhegmatogenous retinal detachment (RRD) caused by superior retinal breaks.Methods: Retrospective, comparative, consecutive case series study. Patients with RRD caused by superior retinal breaks undergone PPV with partial tamponade (Group A) and whole tamponade (Group B) of filtered air were included. The main outcomes were primary and final success rates, best corrected visual acuity (BCVA), and rate of postoperative cataract surgery.Results: Forty-one patients (41 eyes) were included in Group A and 36 patients (36 eyes) were included in Group B. There were no significant differences in primary or final success rates between Groups A and B (P = 0.618 and P = 1.000, respectively). The patients in Group A experienced quicker postoperative vision improvement (from the Week 1 follow-up) than the patients in Group B (from the Month 3 follow-up). The postoperative cataract surgery rate of Group A (7/31) was lower than that of Group B (13/26) (P = 0.031).Conclusions: PPV with partial tamponade of air is effective in achieving a high anatomic success rate, quicker postoperative vision improvement, and lower rate of postoperative cataract surgery in RRD caused by superior retinal breaks. [ABSTRACT FROM AUTHOR]

This paper proposes a spectral–spatial feature learning (SSFL) method to obtain robust features of hyperspectral images (HSIs). It combines the spectral feature learning and spatial feature learning in a hierarchical fashion. Stacking a set of SSFL units, a deep hierarchical model called the spectral–spatial networks (SSN) is further proposed for HSI classification. SSN can exploit both discriminative spectral and spatial information simultaneously. Specifically, SSN learns useful high-level features by alternating between spectral and spatial feature learning operations. Then, kernel-based extreme learning machine (KELM), a shallow neural network, is embedded in SSN to classify image pixels. Extensive experiments are performed on two benchmark HSI datasets to verify the effectiveness of SSN. Compared with state-of-the-art methods, SSN with a deep hierarchical architecture obtains higher classification accuracy in terms of the overall accuracy, average accuracy, and kappa ( \kappa ) coefficient of agreement, especially when the number of the training samples is small. [ABSTRACT FROM PUBLISHER]

Age-related macular degeneration (AMD) is the major cause of loss of sight globally. There is currently no effective treatment available. Retinal pigment epithelial (RPE) cells are an important part of the outer blood-retina barrier and their death is a determinant of AMD. Propofol, a common clinically used intravenous anesthetic agent, has been shown to act as an efficacious neuroprotective agent with antioxidative and anti-inflammatory properties in vivo and in vitro. However, little is known about its effects on RPE cells. The purpose of our research was to investigate whether propofol could protect RPE cells from apoptosis through endoplasmic reticulum (ER) stress–dependent pathways. To this end, prior to stimulation with thapsigargin (TG), ARPE-19 cells were pretreated with varying concentrations of propofol. A protective effect of propofol in TG-treated ARPE-9 was apparent, TUNEL and flow cytometric assays showed decreased apoptosis. We further demonstrated that propofol pretreatment attenuated or inhibited the effects caused by TG, such as upregulation of Bax, BiP, C/EBP homologous protein (CHOP), active caspase 12, and cleaved caspase 3, and downregulation of Bcl2. It also decreased the TG-induced levels of ER stress–related molecules such as p-PERK, p-eIF2α, and ATF4. Furthermore, it downregulated the expression of nuclear factor κB (NF-κB). This study elucidated novel propofol-induced cellular mechanisms for antiapoptotic activities in RPE cells undergoing ER stress and demonstrated the potential value of using propofol in the treatment of AMD. [ABSTRACT FROM AUTHOR]

Small infrared (IR) target detection is an important and challenging issue in IR target tracking system. In this paper, a patch ordering-based method is proposed for small IR target detection in an image with complicated background. Inspired by the contrast mechanism of human vision system, a patch ordering-based contrast measure (POCM) is designed to deal with the input image, which cannot only suppress the background noise and clutter but also enhance the targets significantly. In this way, POCM can increase the contrast ratio between target and background. This leads to higher Signal Clutter Ratio (SCR). Then the true target can be detected by applying simple adaptive thresholding method. The experimental results on two sequences show that the proposed method can efficiently detect small IR target from heavy clutter. [ABSTRACT FROM AUTHOR]

Background: Glaucoma combined with an extremely shallow anterior chamber and cataracts remains as a complex condition to deal with. And the emergence of microincision vitrectomy surgery (MIVS) system may provide an ideal option for the treatment of that. We report a clinical study of surgical outcomes of 23-gauge transconjunctival pars plana vitrectomy (PPV) combined with lensectomy in the treatment of glaucomatous eyes with extremely shallow anterior chamber and cataract.Methods: Prospective, nonrandomized and noncomparative case series study. Consecutive patients with secondary glaucoma, extremely shallow anterior chamber and cataract were recruited to have combined surgeries of 23-gauge transconjunctival pars plana vitrectomy and lensectomy. The main outcomes were best corrected visual acuity (BCVA), intraocular pressure (IOP), anterior chamber depth (ACD), number of anti-glaucoma medications and surgery-associated complications.Results: Seventeen consecutive patients with secondary glaucoma, extremely shallow anterior chamber and cataract were recruited. The mean follow-up was 21.2 ± 8.8 months. Postoperatively, there was no significant improvement of BCVA (P = 0.25). The mean intraocular (IOP) decreased significantly from 43.14 ± 6.53 mmHg to 17.29 ± 1.80 mmHg (P < 0.001), and the mean depth of anterior chamber increased significantly from 0.507 ± 0.212 mm to 3.080 ± 0.313 mm (P < 0.001). The mean number of anti-glaucoma medications decreased from 4.1 ± 0.8 to 0.6 ± 0.8 (P < 0.001). No severe vision-threatening intra- or post-operative complications occurred.Conclusions: Glaucoma with an extremely shallow anterior chamber and cataract can be managed well with the combined surgeries of 23-gauge pars plana vitrectomy and lensectomy. The surgical procedure is an effective and safe method to resolve the pupillary block and deepen the anterior chamber. [ABSTRACT FROM AUTHOR]

Background: Suture exposure remains to be a potential problem of transscleral fixated posterior chamber intraocular lens (PCIOL). We report a modified technique to minimize the risk of suture exposure for the transscleral fixation of PCIOL.Methods: The modified surgical technique is as following: at first, two 3 mm × 4 mm square scleral pockets were created from groove incisions at opposite positions. A straight needle attached to a 10-0 polypropylene suture was passed through one incision groove. Then, a 27-Gauge hollow needle passed through the opposite sclera incision bed was used to retrieve the straight fine needle via its barrel. The sutures were tied to themselves after one more bite on the scleral bed. At last, the suture ends were left long (about 4 mm) and laid flat into corresponding laminar scleral pockets. This modified technique of PCIOL was performed in 48 post-traumatic aphakic vitrectomized eyes from 48 patients (47 male, one female) with mean age of 34.8 ± 14.8 years. Main outcome measures included best corrective visual acuity (BCVA), IOL decentration, IOL tilt, and postoperative complications.Results: The mean follow-up was 32.3 ± 10.8 months (3-67 months). The LogMAR BCVA remained stable, from a preoperative value of 0.46 ± 0.34 to postoperative 0.44 ± 0.34 (p = 0.69). Mild IOL tilt (5-10°) was observed in five eyes, and slight IOL decentration (0.5-1.0 mm) was seen in three cases. No case of suture exposure, suture breakage, IOL dislocation, or endophthalmitis was observed during the follow up period.Conclusion: The modified technique allowed stable placement of PCIOLs in post-traumatic aphakic eyes with a wide range of follow-up. Our procedure might have the potential benefit to avoid suture exposure in scleral-fixated IOL implantation. [ABSTRACT FROM AUTHOR]

Background: To evaluate the efficacy and safety of the Ahmed glaucoma valve (AGV) and the risk factors associated with AGV implantation failure in a population of Chinese patients with refractory glaucoma. Method: In total, 79 eyes with refractory glaucoma from 79 patients treated in our institution from November 2007 to November 2010 were enrolled in this retrospective study. The demographic data, preoperative and postoperative intraocular pressures (IOPs), best corrected visual acuity (BCVA), number of anti-glaucoma medications used, completed and qualified surgery success rates and postoperative complications were recorded to evaluate the outcomes of AGV implantation. Factors that were associated with implant failure were determined using Cox proportional hazard regression model analysis and multiple linear regression analysis. Principle Findings: The average follow-up time was 12.7±5.8 months (mean±SD). We observed a significant reduction in the mean IOP from 39.9±12.6 mm Hg before surgery to 19.3±9.6 mm Hg at the final follow-up. The complete success rate was 59.5%, and the qualified success rate was 83.5%. The number of previous surgeries was negatively correlated with qualified success rate (P<0.05, OR=0.736, 95% CI 0.547-0.99). Patients with previous trabeculectomy were more likely to use multiple anti-glaucoma drugs to control IOP (P<0.01). The primary complication was determined to be a flat anterior chamber (AC). Conclusion: AGV implantation was safe and effective for the management of refractory glaucoma. Patients with a greater number of previous surgeries were more likely to experience surgical failure, and patients with previous trabeculectomy were more likely to use multiple anti-glaucoma drugs to control postoperative IOP. [ABSTRACT FROM AUTHOR]

Background suppression is vitally important for the small target detection, which aims to enhance targets and improve the signal‐to‐noise ratio of small target images. Consequently, the study proposes a background suppression approach based on the fast local reverse entropy operator, which is designed according to the fact that the appearance of a small target could result in the great change of the value of local reverse entropy in the local region. The operator is adopted to suppress complex backgrounds of small target images in order to enhance small targets, and then bring about high probabilities of detection and low probabilities of false alarm in the small target detection. Both quantitative and qualitative analyses contribute to confirm the validity and efficiency of the proposed approach. [ABSTRACT FROM AUTHOR]

There have been different approaches to compute the analytic instantaneous frequency with a variety of background reasoning and applicability in practice, as well as restrictions. This paper presents a new non-analytic instantaneous frequency definition for a large class of so-called simple waves. Simple wave contains a wide range of signals for which the concept instantaneous frequency has a perfect physical sense. There are a number of automatic sleep spindle detection methods in the literature, however, due to the complexity of the sleep spindle characteristics from different people and absence of generally accepted sleep spindle definition in the literature, researchers are still exploring more practically acceptable and workable methods. A novel instantaneous frequency definition based semi-automatic sleep spindle detection approach is also proposed in this paper. A set of experiments on publicly available database clearly indicate that the proposed approaches are very promising. Copyright © 2012 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

Deep learning is now receiving widespread attention in hyperspectral image (HSI) classification. However, due to the imbalance between a huge number of weights and limited training samples, many problems and difficulties have arisen from the use of deep learning methods in HSI classification. To handle this issue, an efficient deep learning-based HSI classification method, namely, spatial-aware network (SANet) has been proposed in this paper. The main idea of SANet is to exploit discriminative spectral-spatial features by incorporating prior domain knowledge into the deep architecture, where edge-preserving side window filters are used as the convolution kernels. Thus, SANet has a small number of parameters to optimize. This makes it fit for small sample sizes. Furthermore, SANet is able not only to aware local spatial structures using side window filtering framework, but also to learn discriminative features making use of the hierarchical architecture and limited label information. The experimental results on four widely used HSI data sets demonstrate that our proposed SANet significantly outperforms many state-of-the-art approaches when only a small number of training samples are available. [ABSTRACT FROM AUTHOR]