Feature Pyramid U-Net with Attention for Semantic Segmentation of Forward-Looking Sonar Images.

Forward-looking sonar is a technique widely used for underwater detection. However, most sonar images have underwater noise and low resolution due to their acoustic properties. In recent years, the semantic segmentation model U-Net has shown excellent segmentation performance, and it has great potential in forward-looking sonar image segmentation. However, forward-looking sonar images are affected by noise, which prevents the existing U-Net model from segmenting small objects effectively. Therefore, this study presents a forward-looking sonar semantic segmentation model called Feature Pyramid U-Net with Attention (FPUA). This model uses residual blocks to improve the training depth of the network. To improve the segmentation accuracy of the network for small objects, a feature pyramid module combined with an attention structure is introduced. This improves the model's ability to learn deep semantic and shallow detail information. First, the proposed model is compared against other deep learning models and on two datasets, of which one was collected in a tank environment and the other was collected in a real marine environment. To further test the validity of the model, a real forward-looking sonar system was devised and employed in the lake trials. The results show that the proposed model performs better than the other models for small-object and few-sample classes and that it is competitive in semantic segmentation of forward-looking sonar images.

Fast single fringe-pattern processing with graphics processing unit.

Optical interferometric techniques provide noncontact, full-field, and high-precision measurements that are very attractive in various research and application fields. Single fringe-pattern processing (SFPP) is often required when measuring fast phenomena, which contain multiple steps including noise removal, phase demodulation, and unwrapping. However, several difficulties are encountered during SFPP, among which the processing time is of interest due to the increasing computational load brought by the large amount and high-resolution fringe patterns in recent years. In this paper, we propose a general and complete graphics processing unit (GPU)-based SFPP framework to perform a systematic discussion on SFPP acceleration. Typical methods from the spatial domain, the transform-based, and the path-related are chosen to have a variety of methods in the framework for better parallelization demonstration, namely, coherence-enhancing diffusion for denoising, spiral phase quadrature transform for demodulation, and quality-guided phase unwrapping. To the best of our knowledge, this is the first time a complete GPU-based framework has been proposed for SFPP. The advantages of performing the analysis and parallelization in framework level are demonstrated, where processing redundancy can be identified and reduced. The proposed framework can be used as an example to demonstrate the GPU-based parallelization in SFPP. Methods in the framework can be replaced but the framework level analysis, the parallel design, and the involved functions are always good references. Experiments are performed on simulated and experimental fringe patterns to demonstrate the effectiveness of the proposed work and achieve at most 29.8 times speedup compared with CPU-based sequential processing.

MiR-570 inhibited the cell proliferation and invasion through directly targeting B7-H1 in hepatocellular carcinoma.

A recent study reported that miR-570 was the most important microRNA in the microRNA gene networks of alcoholic liver disease that has the potential of progressing to hepatocellular carcinoma. However, litter is known regarding the expression and specific function of miR-570 in the progression of hepatocellular carcinoma, especially its molecular mechanisms by which miR-570 exerts its functions and modulates the malignant phenotypes of hepatocellular carcinoma cells. Here, we observed that miR-570 was highly expressed in hepatocellular carcinoma cell lines (Bel-7404, Huh-7, and HepG2), while B7-H1 was lowly expressed, compared to nonmalignant cell line (L-02 and HL-7702). Transfection of miR-570 mimics or knockdown of B-H1 suppressed the expression of B7-H1, which promotes cell apoptosis and inhibits the cell proliferation and invasion. Using a dual-luciferase reporter system, we verified that B7-H1 is a direct target of miR-570. The overexpression of B7-H1 reversed the inhibition of proliferation and invasion by miR-570. In addition, miR-570 suppressed tumorigenicity in vivo. Hence, our observation confirmed that miR-570 works as proliferation and metastatic suppressor in hepatocellular carcinoma cells through directly targeting B7-H1 in hepatocellular carcinoma cell and rationally presents that miR-570 has the potential to be a useful clinical noninvasive diagnostics or predictive marker in human hepatocellular carcinoma.

ZJUT-EIFD: A Synchronously Collected External and Internal Fingerprint Database.

External fingerprints (EFs) based only on epidermal information are vulnerable to spoofing attacks and non-ideal skin conditions. To solve such shortcomings, internal fingerprints (IFs) collected using optical coherence tomography (OCT) have been proposed and widely researched. However, the development of IF is limited by the lack of in-depth researches on the IF and the EF-IF interoperability, which is partially caused by the lack of public OCT database. The obvious gap in the applications of EF and IF recognition motivated us to design and publish a comprehensive fingerprint database containing both traditional EFs and OCT IFs, denoted as ZJUT-EIFD. To the best of our knowledge, ZJUT-EIFD is the first public database that combines OCT and total internal reflection (TIR) via synchronous acquisition, with 399 different fingers from 60 subjects. In this article, the composition of the database, the quality of EFs and IFs, and the verification performance of different types of fingerprints were detailed. In addition, potential application directions of ZJUT-EIFD were demonstrated. ZJUT-EIFD can serve benchmarks and interoperability tests for EF-IF research, which will promote the research and development of EF and IF.

Motion-Aware Memory Network for Fast Video Salient Object Detection.

Previous methods based on 3DCNN, convLSTM, or optical flow have achieved great success in video salient object detection (VSOD). However, these methods still suffer from high computational costs or poor quality of the generated saliency maps. To address this, we design a space-time memory (STM)-based network that employs a standard encoder-decoder architecture. During the encoding stage, we extract high-level temporal features from the current frame and its adjacent frames, which is more efficient and practical than methods reliant on optical flow. During the decoding stage, we introduce an effective fusion strategy for both spatial and temporal branches. The semantic information of the high-level features is used to improve the object details in the low-level features. Subsequently, spatiotemporal features are methodically derived step by step to reconstruct the saliency maps. Moreover, inspired by the boundary supervision prevalent in image salient object detection (ISOD), we design a motion-aware loss that predicts object boundary motion, and simultaneously perform multitask learning for VSOD and object motion prediction. This can further enhance the model's capability to accurately extract spatiotemporal features while maintaining object integrity. Extensive experiments on several datasets demonstrate the effectiveness of our method and can achieve state-of-the-art metrics on some datasets. Our proposed model does not require optical flow or additional preprocessing, and can reach an impressive inference speed of nearly 100 FPS.

A Predictive Visual Analytics System for Studying Neurodegenerative Disease Based on DTI Fiber Tracts.

Diffusion tensor imaging (DTI) has been used to study the effects of neurodegenerative diseases on neural pathways, which may lead to more reliable and early diagnosis of these diseases as well as a better understanding of how they affect the brain. We introduce a predictive visual analytics system for studying patient groups based on their labeled DTI fiber tract data and corresponding statistics. The system's machine-learning-augmented interface guides the user through an organized and holistic analysis space, including the statistical feature space, the physical space, and the space of patients over different groups. We use a custom machine learning pipeline to help narrow down this large analysis space and then explore it pragmatically through a range of linked visualizations. We conduct several case studies using DTI and T1-weighted images from the research database of Parkinson's Progression Markers Initiative.

MUSE: Visual Analysis of Musical Semantic Sequence.

Visualization has the capacity of converting auditory perceptions of music into visual perceptions, which consequently opens the door to music visualization (e.g., exploring group style transitions and analyzing performance details). Current research either focuses on low-level analysis without constructing and comparing music group characteristics, or concentrates on high-level group analysis without analyzing and exploring detailed information. To fill this gap, integrating the high-level group analysis and low-level details exploration of music, we design a musical semantic sequence visualization analytics prototype system (MUSE) that mainly combines a distribution view and a semantic detail view, assisting analysts in obtaining the group characteristics and detailed interpretation. In the MUSE, we decompose the music into note sequences for modeling and abstracting music into three progressively fine-grained pieces of information (i.e., genres, instruments and notes). The distribution view integrates a new density contour, which considers sequence distance and semantic similarity, and helps analysts quickly identify the distribution features of the music group. The semantic detail view displays the music note sequences and combines the window moving to avoid visual clutter while ensuring the presentation of complete semantic details. To prove the usefulness and effectiveness of MUSE, we perform two case studies based on real-world music MIDI data. In addition, we conduct a quantitative user study and an expert evaluation.

Position Fusing and Refining for Clear Salient Object Detection.

Multilevel feature fusion plays a pivotal role in salient object detection (SOD). High-level features present rich semantic information but lack object position information, whereas low-level features contain object position information but are mixed with noises such as backgrounds. Appropriately addressing the gap between low-and high-level features is important in SOD. We first propose a global position embedding attention (GPEA) module to minimize the discrepancy between multilevel features in this article. We extract the position information by utilizing the semantic information at high-level features to resist noises at low-level features. Object refine attention (ORA) module is introduced to refine features used to predict saliency maps further without any additional supervision and heighten discriminative regions near the salient object, such as boundaries. Moreover, we find that the saliency maps generated by the previous methods contain some blurry regions, and we design a pixel value (PV) loss to help the model generate saliency maps with improved clarity. Experimental results on five commonly used SOD datasets demonstrated that the proposed method is effective and outperforms the state-of-the-art approaches on multiple metrics.

Morphological and physiological changes in Artemisia selengensis under drought and after rehydration recovery.

Changes in global climate and precipitation patterns have exacerbated the existing uneven distribution of water, causing many plants to face the alternate situation of drought and water flooding. We studied the growth and physiological response of the wetland plant Artemisia selengensis to drought and rehydration. In this study, Artemisia selengensis seedlings were subjected to 32.89% (SD), 47.36 % (MD), 60.97% (MID), and 87.18 % (CK) field water holding capacity for 70 days, followed by 14 days of rehydration. The results showed that drought inhibited the increase of plant height, basal diameter, and biomass accumulation under SD and MD, but the root shoot ratio (R/S) increased. Drought stress also decreased the content of total chlorophyll (Chl), chlorophyll a (Chl-a), chlorophyll b (Chl-b), and carotenoid (Car). Soluble sugar (SS) and proline (Pro) were accumulated rapidly under drought, and the relative water content (RWC) of leaves was kept at a high level of 80%. After rehydration, the plant height, basal diameter, biomass, and R/S ratio could not be recovered under SD and MD, but these indicators were completely recovered under MID. The RWC, Chl, Chl-a, Chl-b, Car, and osmotic substances were partially or completely recovered. In conclusion, Artemisia selengensis not only can improve drought resistance by increasing the R/S ratio and osmotic substances but also adopt the compensatory mechanism during rehydration. It is predictable that A. selengensis may benefit from possible future aridification of wetlands and expand population distribution.

DT2VIS: A Focus+Context Answer Generation System to Facilitate Visual Exploration of Tabular Data.

The visual analysis dialog system utilizing natural language interface is emerging as a promising data analysis tool. However, previous work mostly focused on accurately understanding the query intention of a user but not on generating answers and inducing explorations. A focus+context answer generation approach, which allows users to obtain insight and contextual information simultaneously, is proposed in this work to address the incomplete user query (i.e., input query cannot reflect all possible intentions of the user). A query recommendation algorithm, which applies the historical query information of a user to recommend a follow-up query, is also designed and implemented to provide an in-depth exploration. These ideas are implemented in a system called DT2VIS. Specific cases of utilizing DT2VIS are also provided to analyze data. Finally, the results show that DT2VIS could help users easily and efficiently reach their analysis goals in a comparative study.

Scene Categorization by Deeply Learning Gaze Behavior in a Semisupervised Context.

Accurately recognizing different categories of sceneries with sophisticated spatial configurations is a useful technique in computer vision and intelligent systems, e.g., scene understanding and autonomous driving. Competitive accuracies have been observed by the deep recognition models recently. Nevertheless, these deep architectures cannot explicitly characterize human visual perception, that is, the sequence of gaze allocation and the subsequent cognitive processes when viewing each scenery. In this paper, a novel spatially aware aggregation network is proposed for scene categorization, where the human gaze behavior is discovered in a semisupervised setting. In particular, as semantically labeling a large quantity of scene images is labor-intensive, a semisupervised and structure-preserved non-negative matrix factorization (NMF) is proposed to detect a set of visually/semantically salient regions from each scenery. Afterward, the gaze shifting path (GSP) is engineered to characterize the process of humans perceiving each scene picture. To deeply describe each GSP, a novel spatially aware CNN termed SA-Net is developed. It accepts input regions with various shapes and statistically aggregates all the salient regions along each GSP. Finally, the learned deep GSP features from the entire scene images are fused into an image kernel, which is subsequently integrated into a kernel SVM to categorize different sceneries. Comparative experiments on six scene image sets have shown the advantage of our method.

Expressions of ECE-CYC2 clade genes relating to abortion of both dorsal and ventral stamens in Opithandra (Gesneriaceae).

BACKGROUND: ECE-CYC2 clade genes known in patterning floral dorsoventral asymmetry (zygomorphy) in Antirrhinum majus are conserved in the dorsal identity function including arresting the dorsal stamen. However, it remains uncertain whether the same mechanism underlies abortion of the ventral stamens, an important morphological trait related to evolution and diversification of zygomorphy in Lamiales sensu lato, a major clade of predominantly zygomorphically flowered angiosperms. Opithandra (Gesneriaceae) is of particular interests in addressing this question as it is in the base of Lamiales s.l., an early representative of this type zygomorphy. RESULTS: We investigated the expression patterns of four ECE-CYC2 clade genes and two putative target cyclinD3 genes in Opithandra using RNA in situ hybridization and RT-PCR. OpdCYC gene expressions were correlated with abortion of both dorsal and ventral stamens in Opithandra, strengthened by the negatively correlated expression of their putative target OpdcyclinD3 genes. The complement of OpdcyclinD3 to OpdCYC expressions further indicated that OpdCYC expressions were related to the dorsal and ventral stamen abortion through negative effects on OpdcyclinD3 genes. CONCLUSION: These results suggest that ECE-CYC2 clade TCP genes are not only functionally conserved in the dorsal stamen repression, but also involved in arresting ventral stamens, a genetic mechanism underlying the establishment of zygomorphy with abortion of both the dorsal and ventral stamens evolved in angiosperms, especially within Lamiales s.l.

Referable diabetic retinopathy identification from eye fundus images with weighted path for convolutional neural network.

Diabetic retinopathy (DR) is the most common cause of blindness in middle-age subjects and low DR screening rates demonstrates the need for an automated image assessment system, which can benefit from the development of deep learning techniques. Therefore, the effective classification performance is significant in favor of the referable DR identification task. In this paper, we propose a new strategy, which applies multiple weighted paths into convolutional neural network, called the WP-CNN, motivated by the ensemble learning. In WP-CNN, multiple path weight coefficients are optimized by back propagation, and the output features are averaged for redundancy reduction and fast convergence. The experiment results show that with the efficient training convergence rate WP-CNN achieves an accuracy of 94.23% with sensitivity of 90.94%, specificity of 95.74%, an area under the receiver operating curve of 0.9823 and F1-score of 0.9087. By taking full advantage of the multipath mechanism, the proposed WP-CNN is shown to be accurate and effective for referable DR identification compared to the state-of-art algorithms.

Embedding Spatio-Temporal Information into Maps by Route-Zooming.

Analysis and exploration of spatio-temporal data such as traffic flow and vehicle trajectories have become important in urban planning and management. In this paper, we present a novel visualization technique called route-zooming that can embed spatio-temporal information into a map seamlessly for occlusion-free visualization of both spatial and temporal data. The proposed technique can broaden a selected route in a map by deforming the overall road network. We formulate the problem of route-zooming as a nonlinear least squares optimization problem by defining an energy function that ensures the route is broadened successfully on demand while the distortion caused to the road network is minimized. The spatio-temporal information can then be embedded into the route to reveal both spatial and temporal patterns without occluding the spatial context information. The route-zooming technique is applied in two instantiations including an interactive metro map for city tourism and illustrative maps to highlight information on the broadened roads to prove its applicability. We demonstrate the usability of our spatio-temporal visualization approach with case studies on real traffic flow data. We also study various design choices in our method, including the encoding of the time direction and choices of temporal display, and conduct a comprehensive user study to validate our embedded visualization design.

Sparse Learning with Stochastic Composite Optimization.

In this paper, we study Stochastic Composite Optimization (SCO) for sparse learning that aims to learn a sparse solution from a composite function. Most of the recent SCO algorithms have already reached the optimal expected convergence rate O(1/λT), but they often fail to deliver sparse solutions at the end either due to the limited sparsity regularization during stochastic optimization (SO) or due to the limitation in online-to-batch conversion. Even when the objective function is strongly convex, their high probability bounds can only attain O(√{log(1/δ)/T}) with δ is the failure probability, which is much worse than the expected convergence rate. To address these limitations, we propose a simple yet effective two-phase Stochastic Composite Optimization scheme by adding a novel powerful sparse online-to-batch conversion to the general Stochastic Optimization algorithms. We further develop three concrete algorithms, OptimalSL, LastSL and AverageSL, directly under our scheme to prove the effectiveness of the proposed scheme. Both the theoretical analysis and the experiment results show that our methods can really outperform the existing methods at the ability of sparse learning and at the meantime we can improve the high probability bound to approximately O(log(log(T)/δ)/λT).

Mechanism of Bushen Jianpi decoction in preventing and treating osteoporosis caused by aromatase inhibitors in breast cancer treatment.

PURPOSE: Research on the mechanism of Bushen Jianpi decoction (BJD) for preventing and treating osteoporosis caused by aromatase inhibitors (AI) during treatment for breast cancer resection. METHODS: An ovariectomized mouse model was established using random division into 6 groups: a sham ovariectomized group, a blank control group, a control group, an alendronate group, a BJD group, and a drug combination group. Mice breast cancer cell lines (4T1) were cultured and seeded into the armpits of 6 groups of BALB/c mice. The mouse breast cancer postoperative model was built when resecting the tumor after 3 weeks following seeding tumor. After 1 weeks, the 6 groups of mice were given different drugs. Then the following analyses were made: estradiol (E2) levels and alkaline phosphatase (ALP) levels in the serum; detection of in vitro bone density and calcium and bone phosphorus content; tumor pathology and immunohistochemistry detection. RESULTS: The results suggested that BJD decreased levels of ALP in ovariectomized mice, and there was a trend for improved bone loss. BJD strengthened the trend of alendronate to improve bone loss, improved bone density, bone calcium and phosphorous, and reduced ALP. BJD had a certain role on the promotion of the expression of estrogen receptors (ERs) in the relapse of the tumor tissue. CONCLUSIONS: Combined therapy with BJD and alendronate can act synergistically against osteoporosis, which was possibly related to a reduced bone conversion rate through inhibiting bone resorption. BJD may block the MAPK signal pathway in breast cancer cells, increasing the expression of ERs and making cancer cells sensitive to endocrine treatment.

Coupled Dictionary Learning for the Detail-Enhanced Synthesis of 3-D Facial Expressions.

The desire to reconstruct 3-D face models with expressions from 2-D face images fosters increasing interest in addressing the problem of face modeling. This task is important and challenging in the field of computer animation. Facial contours and wrinkles are essential to generate a face with a certain expression; however, these details are generally ignored or are not seriously considered in previous studies on face model reconstruction. Thus, we employ coupled radius basis function networks to derive an intermediate 3-D face model from a single 2-D face image. To optimize the 3-D face model further through landmarks, a coupled dictionary that is related to 3-D face models and their corresponding 3-D landmarks is learned from the given training set through local coordinate coding. Another coupled dictionary is then constructed to bridge the 2-D and 3-D landmarks for the transfer of vertices on the face model. As a result, the final 3-D face can be generated with the appropriate expression. In the testing phase, the 2-D input faces are converted into 3-D models that display different expressions. Experimental results indicate that the proposed approach to facial expression synthesis can obtain model details more effectively than previous methods can.

Uncertainty-Aware Multidimensional Ensemble Data Visualization and Exploration.

This paper presents an efficient visualization and exploration approach for modeling and characterizing the relationships and uncertainties in the context of a multidimensional ensemble dataset. Its core is a novel dissimilarity-preserving projection technique that characterizes not only the relationships among the mean values of the ensemble data objects but also the relationships among the distributions of ensemble members. This uncertainty-aware projection scheme leads to an improved understanding of the intrinsic structure in an ensemble dataset. The analysis of the ensemble dataset is further augmented by a suite of visual encoding and exploration tools. Experimental results on both artificial and real-world datasets demonstrate the effectiveness of our approach.

EvoRiver: Visual Analysis of Topic Coopetition on Social Media.

Cooperation and competition (jointly called "coopetition") are two modes of interactions among a set of concurrent topics on social media. How do topics cooperate or compete with each other to gain public attention? Which topics tend to cooperate or compete with one another? Who plays the key role in coopetition-related interactions? We answer these intricate questions by proposing a visual analytics system that facilitates the in-depth analysis of topic coopetition on social media. We model the complex interactions among topics as a combination of carry-over, coopetition recruitment, and coopetition distraction effects. This model provides a close functional approximation of the coopetition process by depicting how different groups of influential users (i.e., "topic leaders") affect coopetition. We also design EvoRiver, a time-based visualization, that allows users to explore coopetition-related interactions and to detect dynamically evolving patterns, as well as their major causes. We test our model and demonstrate the usefulness of our system based on two Twitter data sets (social topics data and business topics data).