Catalogue Search | MBRL
Are you sure you want to remove the book from the shelf?
{{itemTitle}}
13,660
result(s) for
"Deng, Wu"
Sort by:
Hyperspectral Image Classification Based on Fusing Ssup.3-PCA, 2D-SSA and Random Patch Network
Recently, the rapid development of deep learning has greatly improved the performance of image classification. However, a central problem in hyperspectral image (HSI) classification is spectral uncertainty, where spectral features alone cannot accurately and robustly identify a pixel point in a hyperspectral image. This paper presents a novel HSI classification network called MS-RPNet, i.e., multiscale superpixelwise RPNet, which combines superpixel-based S[sup.3] -PCA with two-dimensional singular spectrum analysis (2D-SSA) based on the Random Patches Network (RPNet). The proposed frame can not only take advantage of the data-driven method, but can also apply S[sup.3] -PCA to efficiently consider more global and local spectral knowledge at the super-pixel level. Meanwhile, 2D-SSA is used for noise removal and spatial feature extraction. Then, the final features are obtained by random patch convolution and other steps according to the cascade structure of RPNet. The layered extraction superimposes the different sparial information into multi-scale spatial features, which complements the features of various land covers. Finally, the final fusion features are classified by SVM to obtain the final classification results. The experimental results in several HSI datasets demonstrate the effectiveness and efficiency of MS-RPNet, which outperforms several current state-of-the-art methods.
Journal Article
An improved differential evolution algorithm and its application in optimization problem
The selection of the mutation strategy for differential evolution (DE) algorithm plays an important role in the optimization performance, such as exploration ability, convergence accuracy and convergence speed. To improve these performances, an improved differential evolution algorithm with neighborhood mutation operators and opposition-based learning, namely NBOLDE, is developed in this paper. In the proposed NBOLDE, the new evaluation parameters and weight factors are introduced into the neighborhood model to propose a new neighborhood strategy. On this basis, a new neighborhood mutation strategy based on DE/current-to-best/1, namely DE/neighbor-to-neighbor/1, is designed in order to replace large-scale global mutation by local neighborhood mutation with high search efficiency. Then, a generalized opposition-based learning is employed to optimize the initial population and select the better solution between the current solution and reverse solution in order to approximate global optimal solution, which can amend the convergence direction, accelerate convergence, improve efficiency, enhance the stability and avoid premature convergence. Finally, the proposed NBOLDE is compared with four state-of-the-art DE variants by 12 benchmark functions with low-dimension and high-dimension. The experiment results indicate that the proposed NBOLDE has a faster convergence speed, higher convergence accuracy, and better optimization capabilities in solving high-dimensional complex functions.
Journal Article
A single-cell atlas of adult Drosophila ovary identifies transcriptional programs and somatic cell lineage regulating oogenesis
Oogenesis is a complex developmental process that involves spatiotemporally regulated coordination between the germline and supporting, somatic cell populations. This process has been modeled extensively using the Drosophila ovary. Although different ovarian cell types have been identified through traditional means, the large-scale expression profiles underlying each cell type remain unknown. Using single-cell RNA sequencing technology, we have built a transcriptomic data set for the adult Drosophila ovary and connected tissues. Using this data set, we identified the transcriptional trajectory of the entire follicle-cell population over the course of their development from stem cells to the oogenesis-to-ovulation transition. We further identify expression patterns during essential developmental events that take place in somatic and germline cell types such as differentiation, cell-cycle switching, migration, symmetry breaking, nurse-cell engulfment, egg-shell formation, and corpus luteum signaling. Extensive experimental validation of unique expression patterns in both ovarian and nearby, nonovarian cells also led to the identification of many new cell type-and stage-specific markers. The inclusion of several nearby tissue types in this data set also led to our identification of functional convergence in expression between distantly related cell types such as the immune-related genes that were similarly expressed in immune cells (hemocytes) and ovarian somatic cells (stretched cells) during their brief phagocytic role in nurse-cell engulfment. Taken together, these findings provide new insight into the temporal regulation of genes in a cell-type specific manner during oogenesis and begin to reveal the relatedness in expression between cell and tissues types.
Journal Article
A Novel K-Means Clustering Algorithm with a Noise Algorithm for Capturing Urban Hotspots
With the development of cities, urban congestion is nearly an unavoidable problem for almost every large-scale city. Road planning is an effective means to alleviate urban congestion, which is a classical non-deterministic polynomial time (NP) hard problem, and has become an important research hotspot in recent years. A K-means clustering algorithm is an iterative clustering analysis algorithm that has been regarded as an effective means to solve urban road planning problems by scholars for the past several decades; however, it is very difficult to determine the number of clusters and sensitively initialize the center cluster. In order to solve these problems, a novel K-means clustering algorithm based on a noise algorithm is developed to capture urban hotspots in this paper. The noise algorithm is employed to randomly enhance the attribution of data points and output results of clustering by adding noise judgment in order to automatically obtain the number of clusters for the given data and initialize the center cluster. Four unsupervised evaluation indexes, namely, DB, PBM, SC, and SSE, are directly used to evaluate and analyze the clustering results, and a nonparametric Wilcoxon statistical analysis method is employed to verify the distribution states and differences between clustering results. Finally, five taxi GPS datasets from Aracaju (Brazil), San Francisco (USA), Rome (Italy), Chongqing (China), and Beijing (China) are selected to test and verify the effectiveness of the proposed noise K-means clustering algorithm by comparing the algorithm with fuzzy C-means, K-means, and K-means plus approaches. The compared experiment results show that the noise algorithm can reasonably obtain the number of clusters and initialize the center cluster, and the proposed noise K-means clustering algorithm demonstrates better clustering performance and accurately obtains clustering results, as well as effectively capturing urban hotspots.
Journal Article
Hyperspectral Image Classification Based on Fusing S3-PCA, 2D-SSA and Random Patch Network
Recently, the rapid development of deep learning has greatly improved the performance of image classification. However, a central problem in hyperspectral image (HSI) classification is spectral uncertainty, where spectral features alone cannot accurately and robustly identify a pixel point in a hyperspectral image. This paper presents a novel HSI classification network called MS-RPNet, i.e., multiscale superpixelwise RPNet, which combines superpixel-based S3-PCA with two-dimensional singular spectrum analysis (2D-SSA) based on the Random Patches Network (RPNet). The proposed frame can not only take advantage of the data-driven method, but can also apply S3-PCA to efficiently consider more global and local spectral knowledge at the super-pixel level. Meanwhile, 2D-SSA is used for noise removal and spatial feature extraction. Then, the final features are obtained by random patch convolution and other steps according to the cascade structure of RPNet. The layered extraction superimposes the different sparial information into multi-scale spatial features, which complements the features of various land covers. Finally, the final fusion features are classified by SVM to obtain the final classification results. The experimental results in several HSI datasets demonstrate the effectiveness and efficiency of MS-RPNet, which outperforms several current state-of-the-art methods.
Journal Article
A novel collaborative optimization algorithm in solving complex optimization problems
To overcome the deficiencies of weak local search ability in genetic algorithms (GA) and slow global convergence speed in ant colony optimization (ACO) algorithm in solving complex optimization problems, the chaotic optimization method, multi-population collaborative strategy and adaptive control parameters are introduced into the GA and ACO algorithm to propose a genetic and ant colony adaptive collaborative optimization (MGACACO) algorithm for solving complex optimization problems. The proposed MGACACO algorithm makes use of the exploration capability of GA and stochastic capability of ACO algorithm. In the proposed MGACACO algorithm, the multi-population strategy is used to realize the information exchange and cooperation among the various populations. The chaotic optimization method is used to overcome long search time, avoid falling into the local extremum and improve the search accuracy. The adaptive control parameters is used to make relatively uniform pheromone distribution, effectively solve the contradiction between expanding search and finding optimal solution. The collaborative strategy is used to dynamically balance the global ability and local search ability, and improve the convergence speed. Finally, various scale TSP are selected to verify the effectiveness of the proposed MGACACO algorithm. The experiment results show that the proposed MGACACO algorithm can avoid falling into the local extremum, and takes on better search precision and faster convergence speed.
Journal Article
Scleral hypoxia is a target for myopia control
Worldwide, myopia is the leading cause of visual impairment. It results from inappropriate extension of the ocular axis and concomitant declines in scleral strength and thickness caused by extracellular matrix (ECM) remodeling. However, the identities of the initiators and signaling pathways that induce scleral ECM remodeling in myopia are unknown. Here, we used single-cell RNA-sequencing to identify pathways activated in the sclera during myopia development. We found that the hypoxia-signaling, the eIF2-signaling, and mTOR-signaling pathways were activated in murine myopic sclera. Consistent with the role of hypoxic pathways in mouse model of myopia, nearly one third of human myopia risk genes from the genome-wide association study and linkage analyses interact with genes in the hypoxia-inducible factor-1α (HIF-1α)–signaling pathway. Furthermore, experimental myopia selectively induced HIF-1α up-regulation in the myopic sclera of both mice and guinea pigs. Additionally, hypoxia exposure (5% O₂) promoted myofibroblast transdifferentiation with down-regulation of type I collagen in human scleral fibroblasts. Importantly, the antihypoxia drugs salidroside and formononetin down-regulated HIF-1α expression as well as the phosphorylation levels of eIF2α and mTOR, slowing experimental myopia progression without affecting normal ocular growth in guinea pigs. Furthermore, eIF2α phosphorylation inhibition suppressed experimental myopia, whereas mTOR phosphorylation induced myopia in normal mice. Collectively, these findings defined an essential role of hypoxia in scleral ECM remodeling and myopia development, suggesting a therapeutic approach to control myopia by ameliorating hypoxia.
Journal Article
Screening and eradication of Helicobacter pylori for gastric cancer prevention: the Taipei global consensus
ObjectiveA global consensus meeting was held to review current evidence and knowledge gaps and propose collaborative studies on population-wide screening and eradication of Helicobacter pylori for prevention of gastric cancer (GC).Methods28 experts from 11 countries reviewed the evidence and modified the statements using the Delphi method, with consensus level predefined as ≥80% of agreement on each statement. The Grading of Recommendation Assessment, Development and Evaluation (GRADE) approach was followed.ResultsConsensus was reached in 26 statements. At an individual level, eradication of H. pylori reduces the risk of GC in asymptomatic subjects and is recommended unless there are competing considerations. In cohorts of vulnerable subjects (eg, first-degree relatives of patients with GC), a screen-and-treat strategy is also beneficial. H. pylori eradication in patients with early GC after curative endoscopic resection reduces the risk of metachronous cancer and calls for a re-examination on the hypothesis of ‘the point of no return’. At the general population level, the strategy of screen-and-treat for H. pylori infection is most cost-effective in young adults in regions with a high incidence of GC and is recommended preferably before the development of atrophic gastritis and intestinal metaplasia. However, such a strategy may still be effective in people aged over 50, and may be integrated or included into national healthcare priorities, such as colorectal cancer screening programmes, to optimise the resources. Reliable locally effective regimens based on the principles of antibiotic stewardship are recommended. Subjects at higher risk of GC, such as those with advanced gastric atrophy or intestinal metaplasia, should receive surveillance endoscopy after eradication of H. pylori.ConclusionEvidence supports the proposal that eradication therapy should be offered to all individuals infected with H. pylori. Vulnerable subjects should be tested, and treated if the test is positive. Mass screening and eradication of H. pylori should be considered in populations at higher risk of GC.
Journal Article
Neuronal cell cycle reentry events in the aging brain are more prevalent in neurodegeneration and lead to cellular senescence
Increasing evidence indicates that terminally differentiated neurons in the brain may recommit to a cell cycle-like process during neuronal aging and under disease conditions. Because of the rare existence and random localization of these cells in the brain, their molecular profiles and disease-specific heterogeneities remain unclear. Through a bioinformatics approach that allows integrated analyses of multiple single-nucleus transcriptome datasets from human brain samples, these rare cell populations were identified and selected for further characterization. Our analyses indicated that these cell cycle-related events occur predominantly in excitatory neurons and that cellular senescence is likely their immediate terminal fate. Quantitatively, the number of cell cycle re-engaging and senescent neurons decreased during the normal brain aging process, but in the context of late-onset Alzheimer’s disease (AD), these cells accumulate instead. Transcriptomic profiling of these cells suggested that disease-specific differences were predominantly tied to the early stage of the senescence process, revealing that these cells presented more proinflammatory, metabolically deregulated, and pathology-associated signatures in disease-affected brains. Similarly, these general features of cell cycle re-engaging neurons were also observed in a subpopulation of dopaminergic neurons identified in the Parkinson’s disease (PD)-Lewy body dementia (LBD) model. An extended analysis conducted in a mouse model of brain aging further validated the ability of this bioinformatics approach to determine the robust relationship between the cell cycle and senescence processes in neurons in this cross-species setting.
Journal Article