RESUMEN
The unsupervised domain adaptation (UDA) based cross-scene remote sensing image classification has recently become an appealing research topic, since it is a valid solution to unsupervised scene classification by exploiting well-labeled data from another scene. Despite its good performance in reducing domain shifts, UDA in multisource data scenarios is hindered by several critical challenges. The first one is the heterogeneity inherent in multisource data complicates domain alignment. The second challenge is the incomplete representation of feature distribution caused by the neglect of the contribution from global information. The third challenge is the inaccuracies in alignment due to errors in establishing target domain conditional distributions. Since UDA does not guarantee the complete consistency of the distribution of the two domains, networks using simple classifiers are still affected by domain shifts, resulting in poor performance. In this paper, we propose a graph embedding interclass relation-aware adaptive network (GeIraA-Net) for unsupervised classification of multi-source remote sensing data, which facilitates knowledge transfer at the class level for two domains by leveraging aligned features to perceive inter-class relation. More specifically, a graph-based progressive hierarchical feature extraction network is constructed, capable of capturing both local and global features of multisource data, thereby consolidating comprehensive domain information within a unified feature space. To deal with the imprecise alignment of data distribution, a joint de-scrambling alignment strategy is designed to utilize the features obtained by a three-step pseudo-label generation module for more delicate domain calibration. Moreover, an adaptive inter-class topology based classifier is constructed to further improve the classification accuracy by making the classifier domain adaptive at the category level. The experimental results show that GeIraA-Net has significant advantages over the current state-of-the-art cross-scene classification methods.
RESUMEN
Hyperspectral target detection aims to locate targets of interest in the scene, and deep learning-based detection methods have achieved the best results. However, black box network architectures are usually designed to directly learn the mapping between the original image and the discriminative features in a single data-driven manner, a choice that lacks sufficient interpretability. On the contrary, this article proposes a novel deep spatial-spectral joint-sparse prior encoding network (JSPEN), which reasonably embeds the domain knowledge of hyperspectral target detection into the neural network, and has explicit interpretability. In JSPEN, the sparse encoded prior information with spatial-spectral constraints is learned end-to-end from hyperspectral images (HSIs). Specifically, an adaptive joint spatial-spectral sparse model (AS 2 JSM) is developed to mine the spatial-spectral correlation of HSIs and improves the accuracy of data representation. An optimization algorithm is designed for iteratively solving AS 2 JSM, and JSPEN is proposed to simulate the iterative optimization process in the algorithm. Each basic module of JSPEN one-to-one corresponds to the operation in the optimization algorithm so that each intermediate result in the network has a clear explanation, which is convenient for intuitive analysis of the operation of the network. With end-to-end training, JSPEN can automatically capture the general sparse properties of HSIs and faithfully characterize the features of background and target. Experimental results verify the effectiveness and accuracy of the proposed method. Code is available at https://github.com/Jiahuiqu/JSPEN.
RESUMEN
For hyperspectral image (HSI) and multispectral image (MSI) fusion, it is often overlooked that multisource images acquired under different imaging conditions are difficult to be perfectly registered. Although some works attempt to fuse unregistered images, two thorny challenges remain. One is that registration and fusion are usually modeled as two independent tasks, and there is no yet a unified physical model to tightly couple them. Another is that deep learning (DL)-based methods may lack sufficient interpretability and generalization. In response to the above challenges, we propose an unregistered HSI fusion framework energized by a unified model of registration and fusion. First, a novel registration-fusion consistency physical perception model (RFCM) is designed, which uniformly models the image registration and fusion problem to greatly reduce the sensitivity of fusion performance to registration accuracy. Then, an HSI fusion framework (MoE-PNP) is proposed to learn the knowledge reasoning process for solving RFCM. Each basic module of MoE-PNP one-to-one corresponds to the operation in the optimization algorithm of RFCM, which can ensure clear interpretability of the network. Moreover, MoE-PNP captures the general fusion principle for different unregistered images and therefore has good generalization. Extensive experiments demonstrate that MoE-PNP achieves state-of-the-art performance for unregistered HSI and MSI fusion. The code is available at https://github.com/Jiahuiqu/MoE-PNP.
RESUMEN
The ability to control the catalytic activity of enzymes in chemical transformations is essential for the design and development of artificial catalysts. Herein, we report the synthesis and characterization of functional ligands featuring two 1,4,7,10-tetraazacyclododecane units linked by an azobenzene group and their corresponding dinuclear Zn(II) complexes. We show that the configuration switching (E/Z) of the azobenzene spacer in the ligands and their dinuclear Zn(II) complexes is reversibly controlled by irradiation with UV and visible light. The Zn(II)-metal complexes are light-responsive catalysts for the hydrolytic cleavage of nerve agent simulants, i.e., p-nitrophenyl diphenyl phosphate and methyl paraoxon. The catalytic activity of the Z-isomers of the dinuclear Zn(II) complexes outperformed that of the E-counterparts. Moreover, combining the less active E-isomers with gold nanoparticles induced an enhancement in the hydrolysis rate of p-nitrophenyl diphenyl phosphate. Kinetic analysis has shown that the catalytic site appears to involve a single metal ion. We explain our results by considering the different desolvation effects occurring in the catalyst's configurations in the solution and the catalytic systems involving gold nanoparticles.
RESUMEN
Understanding the structure-function relationships encoded on chiral catalysts is important for investigating the fundamental principles of catalytic enantioselectivity. Herein, the synthesis and self-assembly of naphthalene substituted bis-l/d-histidine amphiphiles (bis-l/d-NapHis) in DMF/water solution mixture is reported. The resulting supramolecular assemblies featuring well-defined P/M nanoribbons (NRs). With combination of the (P/M)-NR and metal ion catalytic centers (Mn+ = Co2+, Cu2+, Fe3+), the (P)-NR-Mn+ as chiral supramolecular catalysts show catalytic preference to 3,4-dihydroxy-S-phenylalanine (S-DOPA) oxidation while the (M)-NR-Mn+ show enantioselective bias to R-DOPA oxidation. In contrast, their monomeric counterparts bis-l/d-NapHis-Mn+ display an inverse and dramatically lower catalytic selectivity in the R/S-DOPA oxidation. Among them, the Co2+-coordinated supramolecular nanostructures show the highest catalytic efficiency and enantioselectivity (select factor up to 2.70), while the Fe3+-coordinated monomeric ones show nearly racemic products. Analysis of the kinetic results suggests that the synergistic effect between metal ions and the chiral supramolecular NRs can significantly regulate the enantioselective catalytic activity, while the metal ion-mediated monomeric bis-l/d-NapHis were less active. The studies on association constants and activation energies reveal the difference in catalytic efficiency and enantioselectivity resulting from the different energy barriers and binding affinities existed between the chiral molecular/supramolecular structures and R/S-DOPA enantiomers. This work clarifies the correlation between chiral molecular/supramolecular structures and enantioselective catalytic activity, shedding new light on the rational design of chiral catalysts with outstanding enantioselectivity.
RESUMEN
Hyperspectral change detection, which provides abundant information on land cover changes in the Earth's surface, has become one of the most crucial tasks in remote sensing. Recently, deep-learning-based change detection methods have shown remarkable performance, but the acquirement of labeled data is extremely expensive and time-consuming. It is intuitive to learn changes from the scene with sufficient labeled data and adapting them into an unlabeled new scene. However, the nonnegligible domain shift between different scenes leads to inevitable performance degradation. In this article, a cycle-refined multidecision joint alignment network (CMJAN) is proposed for unsupervised domain adaptive hyperspectral change detection, which realizes progressive alignment of the data distributions between the source and target domains with cycle-refined high-confidence labeled samples. There are two key characteristics: 1) progressively mitigate the distribution discrepancy to learn domain-invariant difference feature representation and 2) update the high-confidence training samples of the target domain in a cycle manner. The benefit is that the domain shift between the source and target domains is progressively alleviated to promote change detection performance on the target domain in an unsupervised manner. Experimental results on different datasets demonstrate that the proposed method can achieve better performance than the state-of-the-art change detection methods.
RESUMEN
Hyperspectral (HS) pansharpening aims at fusing an observed HS image with a panchromatic (PAN) image, to produce an image with the high spectral resolution of the former and the high spatial resolution of the latter. Most of the existing convolutional neural networks (CNNs)-based pansharpening methods reconstruct the desired high-resolution image from the encoded low-resolution (LR) representation. However, the encoded LR representation captures semantic information of the image and is inadequate in reconstructing fine details. How to effectively extract high-resolution and LR representations for high-resolution image reconstruction is the main objective of this article. In this article, we propose a feature pyramid fusion network (FPFNet) for pansharpening, which permits the network to extract multiresolution representations from PAN and HS images in two branches. The PAN branch starts from the high-resolution stream that maintains the spatial resolution of the PAN image and gradually adds LR streams in parallel. The structure of the HS branch remains highly consistent with that of the PAN branch, but starts with the LR stream and gradually adds high-resolution streams. The representations with corresponding resolutions of PAN and HS branches are fused and gradually upsampled in a coarse to fine manner to reconstruct the high-resolution HS image. Experimental results on three datasets demonstrate the significant superiority of the proposed FPFNet over the state-of-the-art methods in terms of both qualitative and quantitative comparisons.
RESUMEN
Well-known deep learning (DL) is widely used in fusion based hyperspectral image super-resolution (HS-SR). However, DL-based HS-SR models have been designed mostly using off-the-shelf components from current deep learning toolkits, which lead to two inherent challenges: i) they have largely ignored the prior information contained in the observed images, which may cause the output of the network to deviate from the general prior configuration; ii) they are not specifically designed for HS-SR, making it hard to intuitively understand its implementation mechanism and therefore uninterpretable. In this paper, we propose a noise prior knowledge informed Bayesian inference network for HS-SR. Instead of designing a "black-box" deep model, our proposed network, termed as BayeSR, reasonably embeds the Bayesian inference with the Gaussian noise prior assumption to the deep neural network. In particular, we first construct a Bayesian inference model with the Gaussian noise prior assumption that can be solved iteratively by the proximal gradient algorithm, and then convert each operator involved in the iterative algorithm into a specific form of network connection to construct an unfolding network. In the process of network unfolding, based on the characteristics of the noise matrix, we ingeniously convert the diagonal noise matrix operation which represents the noise variance of each band into the channel attention. As a result, the proposed BayeSR explicitly encodes the prior knowledge possessed by the observed images and considers the intrinsic generation mechanism of HS-SR through the whole network flow. Qualitative and quantitative experimental results demonstrate the superiority of the proposed BayeSR against some state-of-the-art methods.
RESUMEN
Existing deep convolutional neural networks (CNNs) have recently achieved great success in pansharpening. However, most deep CNN-based pansharpening models are based on "black-box" architecture and require supervision, making these methods rely heavily on the ground-truth data and lose their interpretability for specific problems during network training. This study proposes a novel interpretable unsupervised end-to-end pansharpening network, called as IU2PNet, which explicitly encodes the well-studied pansharpening observation model into an unsupervised unrolling iterative adversarial network. Specifically, we first design a pansharpening model, whose iterative process can be computed by the half-quadratic splitting algorithm. Then, the iterative steps are unfolded into a deep interpretable iterative generative dual adversarial network (iGDANet). Generator in iGDANet is interwoven by multiple deep feature pyramid denoising modules and deep interpretable convolutional reconstruction modules. In each iteration, the generator establishes an adversarial game with the spatial and spectral discriminators to update both spectral and spatial information without ground-truth images. Extensive experiments show that, compared with the state-of-the-art methods, our proposed IU2PNet exhibits very competitive performance in terms of quantitative evaluation metrics and qualitative visual effects.
RESUMEN
The tendon-to-bone interface is a special structure connecting the tendon and bone and is crucial for mechanical load transfer between dissimilar tissues. After an injury, fibrous scar tissues replace the native tendon-to-bone interface, creating a weak spot that needs to endure extra loading, significantly decreasing the mechanical properties of the motor system. Macrophages play a critical role in tendon-bone healing and can be divided into various phenotypes, according to their inducing stimuli and function. During the early stages of tendon-bone healing, M1 macrophages are predominant, while during the later stages, M2 macrophages replace the M1 macrophages. The two macrophage phenotypes play a significant, yet distinct, role in tendon-bone healing. Growing evidence shows that regulating the macrophage phenotypes is able to promote tendon-bone healing. This review aims to summarize the impact of different macrophages on tendon-bone healing and the current immunomodulatory biomaterials for regulating macrophages, which are used to promote tendon-bone healing. Although macrophages are a promising target for tendon-bone healing, the challenges and limitations of macrophages in tendon-bone healing research are discussed, along with directions for further research.
RESUMEN
Spatio-spectral fusion of panchromatic (PAN) and hyperspectral (HS) images is of great importance in improving spatial resolution of images acquired by many commercial HS sensors. DenseNets have recently achieved great success for image super-resolution because they facilitate gradient flow by concatenating all the feature outputs in a feedforward manner. In this article, we propose a residual hyper-dense network (RHDN) that extends the DenseNet to solve the spatio-spectral fusion problem. The overall structure of the proposed RHDN method is a two-branch network, which allows the network to capture the features of HS images within and outside the visible range separately. At each branch of the network, a two-stream strategy of feature extraction is designed to process PAN and HS images individually. A convolutional neural network (CNN) with cascade residual hyper-dense blocks (RHDBs), which allows direct connections between the pairs of layers within the same stream and those across different streams, is proposed to learn more complex combinations between the HS and PAN images. The residual learning is adopted to make the network efficient. Extensive benchmark evaluations well demonstrate that the proposed RHDN fusion method yields significant improvements over many widely accepted state-of-the-art approaches.
RESUMEN
Hyperspectral (HS) pansharpening, which fuses the HS image with a high spatial resolution panchromatic (PAN) image, provides a good solution to overcome the limitation of HS imaging devices. However, most existing convolutional neural network (CNN)-based methods are hard to understand and lack interpretability due to the black-box design. In this Letter, we propose a multi-level spatial details cross-extraction and injection network (MSCIN) for HS pansharpening, which introduces the mature multi-resolution analysis (MRA) technology to the neural network. Following the general idea of MRA, the proposed MSCIN divides the pansharpening process into details extraction and details injection, in which the missing details and the injection gains are estimated by two specifically designed interpretable sub-networks. Experimental results on two widely used datasets demonstrate the superiority of the proposed method.
Asunto(s)
Redes Neurales de la ComputaciónRESUMEN
Hyperspectral (HS) pansharpening is of great importance in improving the spatial resolution of HS images for remote sensing tasks. HS image comprises abundant spectral contents, whereas panchromatic (PAN) image provides spatial information. HS pansharpening constitutes the possibility for providing the pansharpened image with both high spatial and spectral resolution. This article develops a specific pansharpening framework based on a generative dual-adversarial network (called PS-GDANet). Specifically, the pansharpening problem is formulated as a dual task that can be solved by a generative adversarial network (GAN) with two discriminators. The spatial discriminator forces the intensity component of the pansharpened image to be as consistent as possible with the PAN image, and the spectral discriminator helps to preserve spectral information of the original HS image. Instead of designing a deep network, PS-GDANet extends GANs to two discriminators and provides a high-resolution pansharpened image in a fraction of iterations. The experimental results demonstrate that PS-GDANet outperforms several widely accepted state-of-the-art pansharpening methods in terms of qualitative and quantitative assessment.
RESUMEN
AIMS: Non-small cell lung cancer (NSCLC) is considered a highly fatal tumor. Importantly, angiogenesis is critical for tumor progression. Long non-coding RNAs (lncRNAs), which are untranslatable, control cell functions through different pathways. lncRNA EPIC1 has been reported to promote cell viability, cell cycle progression, and invasion. However, the relationship between EPIC1 and tumor angiogenesis remains an enigma. We explored the role of EPIC1 in tumor angiogenesis in NSCLC. MATERIALS AND METHODS: First, EPIC1 expression was analyzed using the GEPIA database and was further verified using qPCR in tumor tissues from patients with NSCLC and NSCLC cell lines. Next, EPIC1 function was detected using loss-of-function and gain-of-function assays. Moreover, EdU staining, flow cytometry, and channel formation assays were performed to assess HUVEC proliferation and channel the formation in the NSCLC-HUVEC transwell co-culture system. KEY FINDINGS: EPIC1 expression was significantly upregulated in NSCLC tissues and cell lines. Furthermore, the overexpression of EPIC1 in NSCLC cells stimulated HUVEC channel formation and proliferation by activating Ang2/Tie2 signaling, and the opposite results were obtained when EPIC1 was silenced in NSCLC cells. The density of new blood vessels was simultaneously increased by EPIC1 overexpression in vivo, using CAM angiogenesis model and a nude mouse tumor model. Finally, all these experimental findings could be established in the samples from patients with NSCLC. We postulate that EPIC1 promotes tumor angiogenesis by activating the Ang2/Tie2 axis in NSCLC. SIGNIFICANCE: Elucidating the molecular and cellular mechanisms of EPIC1 in tumor angiogenesis provides a novel perspective on NSCLC clinical therapy.
Asunto(s)
Angiopoyetina 2/metabolismo , Carcinoma de Pulmón de Células no Pequeñas/genética , Neoplasias Pulmonares/genética , ARN Largo no Codificante/genética , Receptor TIE-2/metabolismo , Angiopoyetina 2/genética , Animales , Carcinoma de Pulmón de Células no Pequeñas/irrigación sanguínea , Carcinoma de Pulmón de Células no Pequeñas/metabolismo , Carcinoma de Pulmón de Células no Pequeñas/patología , Línea Celular Tumoral , Proliferación Celular/fisiología , Supervivencia Celular/fisiología , Embrión de Pollo , Bases de Datos Genéticas , Modelos Animales de Enfermedad , Xenoinjertos , Células Endoteliales de la Vena Umbilical Humana , Humanos , Neoplasias Pulmonares/irrigación sanguínea , Neoplasias Pulmonares/metabolismo , Neoplasias Pulmonares/patología , Masculino , Ratones , Ratones Endogámicos BALB C , Ratones Desnudos , Neovascularización Patológica/genética , Neovascularización Patológica/patología , ARN Largo no Codificante/metabolismo , Receptor TIE-2/genética , Transducción de SeñalRESUMEN
Injection model-based algorithms have been proved to be effective techniques to solve the pansharpening problem. However, the existing injection model-based algorithms often face an imbalance between over-sharpening and image blurring in the fused image. This paper proposes a local model-based pansharpening method to solve this problem from two aspects. First, an optimization constraint equation formed by the quality index is proposed to reduce the difference between the details of hyperspectral (HS) images and panchromatic (PAN) images. Second, the sliding-window-based fusion scheme is proposed for the first time, to the best of our knowledge, to adaptively fuse the details of HS and PAN images to reduce redundancy. Simulation experiments show that the proposed algorithm has excellent fusion performance from the aspects of subjectivity and objectivity.
RESUMEN
Cellular immunotherapeutics aim to employ immune cells as anticancer agents. Ex vivo engineering of dendritic cells (DCs), the initial role of an immune response, benefits tumor elimination by boosting specific antitumor responses. However, directly activating DCs in vivo is less efficient and therefore quite challenging. Here, we designed a nanoactivator that manufactures DCs through autophagy upregulating in vivo directly, which lead to a high-efficiency antigen presention of DCs and antigen-specific T cells generation. The nanoactivator significantly enhances tumor antigen cross-presentation and subsequent T cell priming. Consequently, in vivo experiments show that the nanoactivators successfully reduce tumor growth and prolong murine survival. Taken together, these results indicate in situ DCs manipulation by autophagy induction is a promising strategy for antigen presentation enhancement and tumor elimination.
Asunto(s)
Autofagia/inmunología , Células Dendríticas/inmunología , Inmunoterapia , Melanoma Experimental/terapia , Nanopartículas/química , Animales , Presentación de Antígeno/inmunología , Línea Celular Tumoral , Femenino , Melanoma Experimental/inmunología , Melanoma Experimental/patología , Ratones , Ratones Endogámicos C57BL , Tamaño de la Partícula , Propiedades de Superficie , Linfocitos T/inmunologíaRESUMEN
Despite the frequency of lung metastasis and its associated mortality, the mechanisms behind metastatic tumor cell survival and colonization in the lungs remain elusive. Here, we show that tumor cell-released microparticles (T-MPs) from the primary tumor site play a critical role in the metastatic process. The T-MPs remodeled the lung parenchyma via a macrophage-dependent pathway to create an altered inflammatory and mechanical response to tumor cell invasion. Mechanistically, we show that circulating T-MPs readily enter the lung parenchyma where they are taken up by local macrophages and induce CCL2 production. CCL2 recruits CD11b+Ly6Chigh inflammatory monocytes to the lungs where they mature into F4/80+CD11b+Ly6C- macrophages that not only produce IL6 but also trigger fibrin deposition. IL6 and the deposited fibrin facilitate the survival and growth of tumor-repopulating cells in the lungs by providing chemical and mechanical signals, respectively, thus setting the stage for lung metastasis. These data illustrate that T-MPs reprogram the lung microenvironment promoting metastasis. Cancer Immunol Res; 6(9); 1046-56. ©2018 AACR.
Asunto(s)
Micropartículas Derivadas de Células/inmunología , Inflamación , Neoplasias Pulmonares/patología , Macrófagos/inmunología , Metástasis de la Neoplasia/inmunología , Animales , Micropartículas Derivadas de Células/patología , Femenino , Pulmón/citología , Pulmón/inmunología , Neoplasias Pulmonares/inmunología , Ratones , Ratones Endogámicos BALB C , Ratones Endogámicos C57BL , Microambiente Tumoral/inmunologíaRESUMEN
Dormancy is recognized as a critical biological event for tumorigenic cells surviving in an extremely harsh environment. Understanding the molecular process of dormancy can unlock novel approaches to tackle cancers. We recently reported that stem-like tumor-repopulating cells (TRC) sense mechanical signals and rapidly proliferate in a 90 Pa soft fibrin matrix. Here, we show that a stiff mechanical environment induces TRC dormancy via an epigenetic program initiated by translocation of Cdc42, a cytosolic regulator of mechanotransduction, into the nucleus, where it promotes transcription of hydroxymethylating enzyme Tet2. Tet2 epigenetically activated cell-cycle-inhibiting genes p21 and p27 to induce dormancy, but also caused downregulation of integrin ß3 to maintain dormancy. This stiffness-mediated dormancy was recapitulated in mouse models for both murine and primary human melanoma TRCs. These data identify an epigenetic program directed by mechanics, which drives highly tumorigenic TRCs to enter dormancy in a stiff mechanical environment.Significance: A mechanics-directed epigenetic program enables tumor-repopulating cells to enter dormancy in a stiff mechanical environment. Cancer Res; 78(14); 3926-37. ©2018 AACR.