Search

Abstract Although unsaturated and density‐dependent flow affect solute fate in groundwater, they are rarely both included in reactive transport models. Using the operator‐splitting method, a new reactive transport model (SUPHRE) was developed by combining a variable‐saturation and variable‐density multiple‐component solute transport model (SUTRA‐MS) with a geochemical reaction module (PhreeqcRM). In contrast to existing reactive transport models, SUPHRE accounts for both unsaturated and density‐dependent flow. Model setup for SUPHRE is convenient, as only one setup file is required in addition to the usual input files of SUTRA‐MS and PhreeqcRM. By further implementing a time‐variant boundary condition into SUTRA‐MS, SUPHRE can simulate multi‐component reactive transport in tidally influenced coastal unconfined aquifers where unsaturated and density‐dependent flow prevail. Two examples were used to validate the new reactive transport model, including single‐species decay and sorption within a one‐dimensional soil column and a four‐species decay chain in a two‐dimensional aquifer. Following validation, SUPHRE was adopted to reveal unsaturated flow effects on oxygen consumption and nitrate reduction in tidally influenced coastal unconfined aquifers. Whether for simulating oxygen consumption or nitrate reduction, there were visible deviations between numerical results without and with unsaturated flow, highlighting that unsaturated flow can affect reactive solute transport and transformation.

Abstract Background For patients with steroid-refractory acute graft-versus-host disease (SR-aGVHD), effective second-line regimens are urgently needed. Mesenchymal stromal cells (MSCs) have been used as salvage regimens for SR-aGVHD in the past. However, clinical trials and an overall understanding of the molecular mechanisms of MSCs combined with basiliximab for SR-aGVHD are limited, especially in haploidentical haemopoietic stem cell transplantation (HID HSCT). Methods The primary endpoint of this multicentre, randomized, controlled trial was the 4-week complete response (CR) rate of SR-aGVHD. A total of 130 patients with SR-aGVHD were assigned in a 1:1 randomization schedule to the MSC group (receiving basiliximab plus MSCs) or control group (receiving basiliximab alone) (NCT04738981). Results Most enrolled patients (96.2%) received HID HSCT. The 4-week CR rate of SR-aGVHD in the MSC group was obviously better than that in the control group (83.1% vs. 55.4%, P = 0.001). However, for the overall response rates at week 4, the two groups were comparable. More patients in the control group used ≥ 6 doses of basiliximab (4.6% vs. 20%, P = 0.008). We collected blood samples from 19 consecutive patients and evaluated MSC-derived immunosuppressive cytokines, including HO1, GAL1, GAL9, TNFIA6, PGE2, PDL1, TGF-β and HGF. Compared to the levels before MSC infusion, the HO1 (P = 0.0072) and TGF-β (P = 0.0243) levels increased significantly 1 day after MSC infusion. At 7 days after MSC infusion, the levels of HO1, GAL1, TNFIA6 and TGF-β tended to increase; however, the differences were not statistically significant. Although the 52-week cumulative incidence of cGVHD in the MSC group was comparable to that in the control group, fewer patients in the MSC group developed cGVHD involving ≥3 organs (14.3% vs. 43.6%, P = 0.006). MSCs were well tolerated, no infusion-related adverse events (AEs) occurred and other AEs were also comparable between the two groups. However, patients with malignant haematological diseases in the MSC group had a higher 52-week disease-free survival rate than those in the control group (84.8% vs. 65.9%, P = 0.031). Conclusions For SR-aGVHD after allo-HSCT, especially HID HSCT, the combination of MSCs and basiliximab as the second-line therapy led to significantly better 4-week CR rates than basiliximab alone. The addition of MSCs not only did not increase toxicity but also provided a survival benefit.

Abstract Background Pathological complete response (pCR) is associated with favorable prognosis in patients with triple-negative breast cancer (TNBC). However, only 30–40% of TNBC patients treated with neoadjuvant chemotherapy (NAC) show pCR, while the remaining 60–70% show residual disease (RD). The role of the tumor microenvironment in NAC response in patients with TNBC remains unclear. In this study, we developed a machine learning-based two-step pipeline to distinguish between various histological components in hematoxylin and eosin (H&E)-stained whole slide images (WSIs) of TNBC tissue biopsies and to identify histological features that can predict NAC response. Methods H&E-stained WSIs of treatment-naïve biopsies from 85 patients (51 with pCR and 34 with RD) of the model development cohort and 79 patients (41 with pCR and 38 with RD) of the validation cohort were separated through a stratified eightfold cross-validation strategy for the first step and leave-one-out cross-validation strategy for the second step. A tile-level histology label prediction pipeline and four machine-learning classifiers were used to analyze 468,043 tiles of WSIs. The best-trained classifier used 55 texture features from each tile to produce a probability profile during testing. The predicted histology classes were used to generate a histology classification map of the spatial distributions of different tissue regions. A patient-level NAC response prediction pipeline was trained with features derived from paired histology classification maps. The top graph-based features capturing the relevant spatial information across the different histological classes were provided to the radial basis function kernel support vector machine (rbfSVM) classifier for NAC treatment response prediction. Results The tile-level prediction pipeline achieved 86.72% accuracy for histology class classification, while the patient-level pipeline achieved 83.53% NAC response (pCR vs. RD) prediction accuracy of the model development cohort. The model was validated with an independent cohort with tile histology validation accuracy of 83.59% and NAC prediction accuracy of 81.01%. The histological class pairs with the strongest NAC response predictive ability were tumor and tumor tumor-infiltrating lymphocytes for pCR and microvessel density and polyploid giant cancer cells for RD. Conclusion Our machine learning pipeline can robustly identify clinically relevant histological classes that predict NAC response in TNBC patients and may help guide patient selection for NAC treatment.

Retinal vessel segmentation is very important for diagnosing and treating certain eye diseases. Recently, many deep learning-based retinal vessel segmentation methods have been proposed; however, there are still many shortcomings (e.g., they cannot obtain satisfactory results when dealing with cross-domain data or segmenting small blood vessels). To alleviate these problems and avoid overly complex models, we propose a novel network based on a multi-scale feature and style transfer (MSFST-NET) for retinal vessel segmentation. Specifically, we first construct a lightweight segmentation module named MSF-Net, which introduces the selective kernel (SK) module to increase the multi-scale feature extraction ability of the model to achieve improved small blood vessel segmentation. Then, to alleviate the problem of model performance degradation when segmenting cross-domain datasets, we propose a style transfer module and a pseudo-label learning strategy. The style transfer module is used to reduce the style difference between the source domain image and the target domain image to improve the segmentation performance for the target domain image. The pseudo-label learning strategy is designed to be combined with the style transfer module to further boost the generalization ability of the model. Moreover, we trained and tested our proposed MSFST-NET in experiments on the DRIVE and CHASE_DB1 datasets. The experimental results demonstrate that MSFST-NET can effectively improve the generalization ability of the model on cross-domain datasets and achieve improved retinal vessel segmentation results than other state-of-the-art methods.

This study examined the influence of preferential flow on pore water flows and marine nitrogen transport reaction in variable saturation and variable density coastal aquifers. The 2-D unconfined aquifer model established was based on the software COMSOL by coupling the dynamic and chemical processes together. The results showed that preferential flow affects groundwater flow and salinity distribution, leading to a more complicated mixing process. The preferential flow resulted in an increase in mixing zone area and the upper saline plume area of 10.33 and 2.62 m2, respectively, a decrease in saltwater wedge area of 7.22 m2, and an increase in nitrate (NO3-) removal efficiency from 7.9% to 8.97%. The NO3- removal efficiency increases progressively with the depth (h) and quantity (n) of preferential flows; however, it decreases after a certain quantity. Further quantitative analysis revealed an increase in the intensity of nitrification and dissolved oxygen inflow flux with preferential flow depth and quantity increase. This phenomenon usually occurs on coasts where biological caves are abundant. The results also offer significant implications for designing engineering measures to mitigate saltwater intrusion and are significant to prevent groundwater quality deterioration in coastal zones.

AbstractMutations in the gene encoding the transient receptor potential vanilloid member 4 (TRPV4), a Ca2+ permeable nonselective cation channel, cause TRPV4‐related disorders. TRPV4 is widely expressed in the brain; however, the pathogenesis underlying TRPV4‐mediated Ca2+ deregulation in neurodevelopment remains unresolved and an effective therapeutic strategy remains to be established. These issues were addressed by isolating mutant dental pulp stem cells from a tooth donated by a child diagnosed with metatropic dysplasia with neurodevelopmental comorbidities caused by a gain‐of‐function TRPV4 mutation, c.1855C > T (p.L619F). The mutation was repaired using CRISPR/Cas9 to generate corrected isogenic stem cells. These stem cells were differentiated into dopaminergic neurons and the pharmacological effects of folic acid were examined. In mutant neurons, constitutively elevated cytosolic Ca2+ augmented AKT‐mediated α‐synuclein (α‐syn) induction, resulting in mitochondrial Ca2+ accumulation and dysfunction. The TRPV4 antagonist, AKT inhibitor, or α‐syn knockdown, normalizes the mitochondrial Ca2+ levels in mutant neurons, suggesting the importance of mutant TRPV4/Ca2+/AKT‐induced α‐syn in mitochondrial Ca2+ accumulation. Folic acid was effective in normalizing mitochondrial Ca2+ levels via the transcriptional repression of α‐syn and improving mitochondrial reactive oxygen species levels, adenosine triphosphate synthesis, and neurite outgrowth of mutant neurons. This study provides new insights into the neuropathological mechanisms underlying TRPV4‐related disorders and related therapeutic strategies.

The increasing usage of interconnected devices within the Internet of Things (IoT) and Industrial IoT (IIoT) has significantly enhanced efficiency and utility in both personal and industrial settings but also heightened cybersecurity vulnerabilities, particularly through IoT malware. This paper explores the use of one-class classification, a method of unsupervised learning, which is especially suitable for unlabeled data, dynamic environments, and malware detection, which is a form of anomaly detection. We introduce the TF-IDF method for transforming nominal features into numerical formats that avoid information loss and manage dimensionality effectively, which is crucial for enhancing pattern recognition when combined with n-grams. Furthermore, we compare the performance of multi-class vs. one-class classification models, including Isolation Forest and deep autoencoder, that are trained with both benign and malicious NetFlow samples vs. trained exclusively on benign NetFlow samples. We achieve 100% recall with precision rates above 80% and 90% across various test datasets using one-class classification. These models show the adaptability of unsupervised learning, especially one-class classification, to the evolving malware threats in the IoT domain, offering insights into enhancing IoT security frameworks and suggesting directions for future research in this critical area.

The Banff Digital Pathology Working Group (DPWG) was established with the goal to establish a digital pathology repository; develop, validate, and share models for image analysis; and foster collaborations using regular videoconferencing. During the calls, a variety of artificial intelligence (AI)-based support systems for transplantation pathology were presented. Potential collaborations in a competition/trial on AI applied to kidney transplant specimens, including the DIAGGRAFT challenge (staining of biopsies at multiple institutions, pathologists’ visual assessment, and development and validation of new and pre-existing Banff scoring algorithms), were also discussed. To determine the next steps, a survey was conducted, primarily focusing on the feasibility of establishing a digital pathology repository and identifying potential hosts. Sixteen of the 35 respondents (46%) had access to a server hosting a digital pathology repository, with 2 respondents that could serve as a potential host at no cost to the DPWG. The 16 digital pathology repositories collected specimens from various organs, with the largest constituent being kidney (n = 12,870 specimens). A DPWG pilot digital pathology repository was established, and there are plans for a competition/trial with the DIAGGRAFT project. Utilizing existing resources and previously established models, the Banff DPWG is establishing new resources for the Banff community.

Nowadays, visualization is becoming an essential part of data analysis. Business Intelligence Visualization (BIV) is a powerful tool that helps modern business flows faster and smoother than ever before. However, research on BIV evaluation is severely lacking; most evaluation studies for BIV focus on usability, which have limited aspects covered for customers’ needs. The purpose of this research is to develop a UX framework that evaluates BIV (UXBIV), including decision-making experience and interactivity. First, we did a literature survey for good understanding of research progress in related fields, and established a conceptual framework. Second, we performed a case study that implemented this framework to demonstrate how our framework can be used in real business. Our analysis shows that this UXBIV framework is quite reasonable, and it can capture the differences among various BIV designs from the users’ standpoints. This UXBIV framework can help design BIV and promote better decision-making on business affairs.

Abstract Heterotrimeric G proteins, comprising Gα, Gβ, and Gγ subunits, are key regulators of eukaryotic intracellular signaling. Extra-large G (XLG) proteins are a subfamily of plant-specific Gα proteins interacting with plasma membrane-localized receptors to regulate multiple biological processes. The Nicotiana benthamiana genome encodes seven XLG proteins, NbXLG1–7, whose functions in disease resistance and underlying mechanisms are unknown. In this study, we silenced all the seven genes and found that disease susceptibility was enhanced when both NbXLG3 and NbXLG5 or NbXLG4 was silenced. Then, we generated N. benthamiana xlg3xlg5 double- and xlg4 single-mutant lines using the CRISPR-Cas9 approach. All the mutants showed reduced resistance to the bacterial pathogen Pseudomonas syringae pv. tomato DC3000, the fungal pathogen Sclerotinia sclerotiorum, and a series of oomycete pathogens, including Phytophthora capsici, Phytophthora infestans, and Phytophthora parasitica. We further demonstrated that NbXLG3/4/5 positively regulated microbial pattern-induced reactive oxygen species burst and defense gene expression by directly coupling to the tested plant immune receptors. In addition, we examined the role of NbXLG3/4/5 in abiotic stress tolerance and observed that NbXLG3 and NbXLG5 negatively regulated plant resistance to high-salt, mannitol, and PEG. Our study demonstrates the possible role of NbXLG3/4/5 in response to biotic and abiotic stresses and provides insights for the improvement of plant resistance to environmental changes.

Human behavior recognition has always been a hot spot for research in computer vision. In this paper, we propose a novel video behavior recognition method based on Actional-Structural Graph Convolution and a Temporal Extension Module under the framework of a Spatio-Temporal Graph Convolution Neural Network, which can optimize the spatial and temporal features simultaneously. The basic network framework of our method consists of three parts: spatial graph convolution module, temporal extension module and attention mechanism module. In the spatial dimension, the action graph convolution is utilized to obtain abundant spatial features by capturing the correlations of distant joint features, and the structural graph convolution expands the existing skeleton graph to acquire the spatial features of adjacent joints. In the time dimension, the sampling range of the temporal graph is expanded for extracting the same and adjacent joints of adjacent frames. Furthermore, attention mechanisms are introduced to improve the performance of our method. In order to verify the effectiveness and accuracy of our method, a large number of experiments were carried out on two standard behavior recognition datasets: NTU-RGB+D and Kinetics. Comparative experiment results show that our proposed method can achieve better performance.

Abstract Advances in microfluidic technologies rely on engineered 3D flow patterns to manipulate samples at the microscale. However, current methods for mapping flows only provide limited 3D and temporal resolutions or require highly specialized optical set-ups. Here, we present a simple defocusing approach based on brightfield microscopy and open-source software to map micro-flows in 3D at high spatial and temporal resolution. Our workflow is both integrated in ImageJ and modular. We track seed particles in 2D before classifying their Z-position using a reference library. We compare the performance of a traditional cross-correlation method and a deep learning model in performing the classification step. We validate our method on three highly relevant microfluidic examples: a channel step expansion and displacement structures as single-phase flow examples, and droplet microfluidics as a two-phase flow example. First, we elucidate how displacement structures efficiently shift large particles across streamlines. Second, we reveal novel recirculation structures and folding patterns in the internal flow of microfluidic droplets. Our simple and widely accessible brightfield technique generates high-resolution flow maps and it will address the increasing demand for controlling fluids at the microscale by supporting the efficient design of novel microfluidic structures.

Background: Neoadjuvant chemotherapy (NAC) is the standard treatment for early-stage triple negative breast cancer (TNBC). The primary endpoint of NAC is a pathological complete response (pCR). NAC results in pCR in only 30–40% of TNBC patients. Tumor-infiltrating lymphocytes (TILs), Ki67 and phosphohistone H3 (pH3) are a few known biomarkers to predict NAC response. Currently, systematic evaluation of the combined value of these biomarkers in predicting NAC response is lacking. In this study, the predictive value of markers derived from H&E and IHC stained biopsy tissue was comprehensively evaluated using a supervised machine learning (ML)-based approach. Identifying predictive biomarkers could help guide therapeutic decisions by enabling precise stratification of TNBC patients into responders and partial or non-responders. Methods: Serial sections from core needle biopsies (n = 76) were stained with H&E and immunohistochemically for the Ki67 and pH3 markers, followed by whole-slide image (WSI) generation. The serial section stains in H&E stain, Ki67 and pH3 markers formed WSI triplets for each patient. The resulting WSI triplets were co-registered with H&E WSIs serving as the reference. Separate mask region-based CNN (MRCNN) models were trained with annotated H&E, Ki67 and pH3 images for detecting tumor cells, stromal and intratumoral TILs (sTILs and tTILs), Ki67+, and pH3+ cells. Top image patches with a high density of cells of interest were identified as hotspots. Best classifiers for NAC response prediction were identified by training multiple ML models and evaluating their performance by accuracy, area under curve, and confusion matrix analyses. Results: Highest prediction accuracy was achieved when hotspot regions were identified by tTIL counts and each hotspot was represented by measures of tTILs, sTILs, tumor cells, Ki67+, and pH3+ features. Regardless of the hotspot selection metric, a complementary use of multiple histological features (tTILs, sTILs) and molecular biomarkers (Ki67 and pH3) resulted in top ranked performance at the patient level. Conclusions: Overall, our results emphasize that prediction models for NAC response should be based on biomarkers in combination rather than in isolation. Our study provides compelling evidence to support the use of ML-based models to predict NAC response in patients with TNBC.

This paper investigates the effects of the cutoff wall on the fate of nitrate (NO3-), the NO3- removal rate, and the salinity distribution in a coastal aquifer under tidal action. A numerical study was performed based on a coupled model with variable-saturation and variable-density flow and a convection-diffusion-reaction equation for solute transport in a coastal unconfined aquifer. The results showed that the cutoff wall led to a larger upper salinity plume (USP) and that the saltwater wedge (SW) further retreated seaward. The recirculation pathways of saltwater and groundwater were largely modified by the wall. The cutoff wall within the tidal range could increase the NO3- mass of denitrification and the NO3- removal efficiency and decrease the length of the SW and the freshwater flux. This modification of the saltwater and groundwater recirculation pathways was enhanced with increasing wall depth. A deeper cutoff wall led to a further retreated SW, lower freshwater flux, and greater improvements in the NO3- mass of denitrification and the NO3- removal efficiency. In addition, the cutoff wall significantly decreased the terrestrial dissolved organic carbon (TDOC) discharge into the sea. Dissolved organic carbon source (SDOC) promoted a higher NO3- removal efficiency. This study provides us with a better understanding of coastal physical-biogeochemical processes and dynamic mechanisms, as well as a guide for designing engineering measures to mitigate NO3- contamination and thus enhance groundwater quality management.

BackgroundAMKL without DS is a rare but aggressive hematological malignant disease in children, and it is associated with inferior outcomes. Several researchers have regarded pediatric AMKL without DS as high-risk or at least intermediate-risk AML and proposed that upfront allogenic hematopoietic stem cell transplantation (HSCT) in first complete remission might improve long-term survival.Patients and methodWe conducted a retrospective study with twenty-five pediatric (< 14 years old) AMKL patients without DS who underwent haploidentical HSCT in the Peking University Institute of Hematology, Peking University People’s Hospital from July 2016 to July 2021. The diagnostic criteria of AMKL without DS were adapted from the FAB and WHO: ≥ 20% blasts in the bone marrow, and those blasts expressed at least one or more of the platelet glycoproteins: CD41, CD61, or CD42. AMKL with DS and therapy related AML was excluded. Children without a suitable closely HLA-matched related or unrelated donor (donors with more than nine out of 10 matching HLA-A, HLA-B, HLA-C, HLA-DR, and HLA-DQ loci), were eligible to receive haploidentical HSCT. Definition was adapted from international cooperation group. All statistical tests were conducted with SPSS v.24 and R v.3.6.3.ResultsThe 2-year OS was 54.5 ± 10.3%, and the EFS was 50.9 ± 10.2% in pediatric AMKL without DS undergoing haplo-HSCT. Statistically significantly better EFS was observed in patients with trisomy 19 than in patients without trisomy 19 (80 ± 12.6% and 33.3 ± 12.2%, respectively, P = 0.045), and OS was better in patients with trisomy 19 but with no statistical significance (P = 0.114). MRD negative pre-HSCT patients showed a better OS and EFS than those who were positive (P < 0.001 and P = 0.003, respectively). Eleven patients relapsed post HSCT. The median time to relapse post HSCT was 2.1 months (range: 1.0–14.4 months). The 2-year cumulative incidence of relapse (CIR) was 46.1 ± 11.6%. One patient developed bronchiolitis obliterans and respiratory failure and died at d + 98 post HSCT.ConclusionAMKL without DS is a rare but aggressive hematological malignant disease in children, and it is associated with inferior outcomes. Trisomy 19 and MRD negative pre-HSCT might contribute to a better EFS and OS. Our TRM was low, haplo-HSCT might be an option for high-risk AMKL without DS.

For routine pathology diagnosis and imaging-based biomedical research, Whole-slide image (WSI) analyses have been largely limited to a 2D tissue image space. For a more definitive tissue representation to support fine-resolution spatial and integrative analyses, it is critical to extend such tissue-based investigations to a 3D tissue space with spatially aligned serial tissue WSIs in different stains, such as Hematoxylin and Eosin (H&E) and Immunohistochemistry (IHC) biomarkers. However, such WSI registration is technically challenged by the overwhelming image scale, the complex histology structure change, and the significant difference in tissue appearances in different stains. The goal of this study is to register serial sections from multi-stain histopathology whole-slide image blocks. We propose a novel translation-based deep learning registration network CGNReg that spatially aligns serial WSIs stained in H&E and by IHC biomarkers without prior deformation information for the model training. First, synthetic IHC images are produced from H&E slides through a robust image synthesis algorithm. Next, the synthetic and the real IHC images are registered through a Fully Convolutional Network with multi-scaled deformable vector fields and a joint loss optimization. We perform the registration at the full image resolution, retaining the tissue details in the results. Evaluated with a dataset of 76 breast cancer patients with 1 H&E and 2 IHC serial WSIs for each patient, CGNReg presents promising performance as compared with multiple state-of-the-art systems in our evaluation. Our results suggest that CGNReg can produce promising registration results with serial WSIs in different stains, enabling integrative 3D tissue-based biomedical investigations.

Abstract Background The mixed-lineage leukemia (MLL) gene is located on chromosome 11q23. The MLL gene can be rearranged to generate partial tandem duplications (MLL-PTD), which occurs in about 5-10% of acute myeloid leukemia (AML) with a normal karyotype and in 5-6% of myelodysplastic syndrome (MDS) patients. Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is currently one of the curative therapies available for AML and MDS with excess blasts (MDS-EB). However, how the prognosis of patients with high levels of MLL-PTD after allo-HSCT, and whether MLL-PTD could be used as a reliable indicator for minimal residual disease (MRD) monitoring in transplant patients remains unknown. Our study purposed to analyze the dynamic changes of MLL-PTD peri-transplantation and the best threshold for predicting relapse after transplantation. Methods We retrospectively collected the clinical data of 48 patients with MLL-PTD AML or MDS-EB who underwent allo-HSCT in Peking University People’s Hospital. The MLL-PTD was examined by real-time quantitative polymerase chain reaction (RQ-PCR) at the diagnosis, before transplantation and the fixed time points after transplantation. Detectable MLL-PTD/ABL > 0.08% was defined as MLL-PTD positive in this study. Results The 48 patients included 33 AML patients and 15 MDS-EB patients. The median follow-up time was 26(0.7-56) months after HSCT. In AML patients, 7 patients (21.2%) died of treatment-related mortality (TRM), 6 patients (18.2%) underwent hematological relapse and died ultimately. Of the 15 patients with MDS-EB, 2 patients (13.3%) died of infection. The 3-year cumulative incidence of relapse (CIR), overall survival (OS), disease-free survival (DFS) and TRM were 13.7 ± 5.2, 67.8 ± 6.9, 68.1 ± 6.8 and 20.3% ± 6.1%, respectively. ROC curve showed that post-transplant MLL-PTD ≥ 1.0% was the optimal cut-off value for predicting hematological relapse after allo-HSCT. There was statistical difference between post-transplant MLL-PTD ≥ 1.0% and MLL-PTD

Abstract Recently, Siamese‐based trackers have drawn amounts of attention in visual tracking field because of their excellent performance on different tracking benchmarks. However, most Siamese‐based trackers encounter difficulties under circumstances such as similar objects interference and background clutters. Besides, there exists an extreme foreground–background data imbalance that weakens the performance during training but few loss functions pay attention to it. The authors intend to address the issues mentioned above by introducing a module named paralleled spatial and channel attention (PSCA) and adaptive focal loss (AFL). Firstly, paralleled spatial and channel attention is proposed to enhance the extracted features and eliminate the noise information from both spatial and channel aspects. Secondly, adaptive focal loss is proposed as the loss function to make the model focus on hard samples that contribute more to training process. Finally, paralleled spatial and channel attention and modified ResNet are combined for extracting more powerful features. Experimental results show that the authors' method achieves outstanding performance in multiple benchmarks while keeping a beyond‐real‐time frame rate.

BackgroundOncotype DX Recurrence Score (RS) has been widely used to predict chemotherapy benefits in patients with estrogen receptor-positive breast cancer. Studies showed that the features used in Magee equations correlate with RS. We aimed to examine whether deep learning (DL)-based histology image analyses can enhance such correlations.MethodsWe retrieved 382 cases with RS diagnosed between 2011 and 2015 from the Emory University and the Ohio State University. All patients received surgery. DL models were developed to detect nuclei of tumor cells and tumor-infiltrating lymphocytes (TILs) and segment tumor cell nuclei in hematoxylin and eosin (H&E) stained histopathology whole slide images (WSIs). Based on the DL-based analysis, we derived image features from WSIs, such as tumor cell number, TIL number variance, and nuclear grades. The entire patient cohorts were divided into one training set (125 cases) and two validation sets (82 and 175 cases) based on the data sources and WSI resolutions. The training set was used to train the linear regression models to predict RS. For prediction performance comparison, we used independent variables from Magee features alone or the combination of WSI-derived image and Magee features.ResultsThe Pearson’s correlation coefficients between the actual RS and predicted RS by DL-based analysis were 0.7058 (p-value = 1.32 × 10–13) and 0.5041 (p-value = 1.15 × 10–12) for the validation sets 1 and 2, respectively. The adjusted R2 values using Magee features alone are 0.3442 and 0.2167 in the two validation sets, respectively. In contrast, the adjusted R2 values were enhanced to 0.4431 and 0.2182 when WSI-derived imaging features were jointly used with Magee features.ConclusionOur results suggest that DL-based digital pathological features can enhance Magee feature correlation with RS.

BackgroundDeep learning methods have demonstrated remarkable performance in pathology image analysis, but they are computationally very demanding. The aim of our study is to reduce their computational cost to enable their use with large tissue image datasets.MethodsWe propose a method called Network Auto-Reduction (NAR) that simplifies a Convolutional Neural Network (CNN) by reducing the network to minimize the computational cost of doing a prediction. NAR performs a compound scaling in which the width, depth, and resolution dimensions of the network are reduced together to maintain a balance among them in the resulting simplified network. We compare our method with a state-of-the-art solution called ResRep. The evaluation is carried out with popular CNN architectures and a real-world application that identifies distributions of tumor-infiltrating lymphocytes in tissue images.ResultsThe experimental results show that both ResRep and NAR are able to generate simplified, more efficient versions of ResNet50 V2. The simplified versions by ResRep and NAR require 1.32× and 3.26× fewer floating-point operations (FLOPs), respectively, than the original network without a loss in classification power as measured by the Area under the Curve (AUC) metric. When applied to a deeper and more computationally expensive network, Inception V4, NAR is able to generate a version that requires 4× lower than the original version with the same AUC performance.ConclusionsNAR is able to achieve substantial reductions in the execution cost of two popular CNN architectures, while resulting in small or no loss in model accuracy. Such cost savings can significantly improve the use of deep learning methods in digital pathology. They can enable studies with larger tissue image datasets and facilitate the use of less expensive and more accessible graphics processing units (GPUs), thus reducing the computing costs of a study.

Abstract Liver cancer is one of the leading causes of cancer deaths in Asia and Africa. It is caused by the Hepatocellular carcinoma (HCC) in almost 90% of all cases. HCC is a malignant tumor and the most common histological type of the primary liver cancers. The detection and evaluation of viable tumor regions in HCC present an important clinical significance since it is a key step to assess response of chemoradiotherapy and tumor cell proportion in genetic tests. Recent advances in computer vision, digital pathology and microscopy imaging enable automatic histopathology image analysis for cancer diagnosis. In this paper, we present a multi-resolution deep learning model HistoCAE for viable tumor segmentation in whole-slide liver histopathology images. We propose convolutional autoencoder (CAE) based framework with a customized reconstruction loss function for image reconstruction, followed by a classification module to classify each image patch as tumor versus non-tumor. The resulting patch-based prediction results are spatially combined to generate the final segmentation result for each WSI. Additionally, the spatially organized encoded feature map derived from small image patches is used to compress the gigapixel whole-slide images. Our proposed model presents superior performance to other benchmark models with extensive experiments, suggesting its efficacy for viable tumor area segmentation with liver whole-slide images.

Abstract Cataract surgery is the most common intraocular procedure. To decrease postsurgical inflammation, prevent infection and reduce the incidence of secondary cataract, we built a temperature-sensitive drug delivery system carrying dexamethasone, moxifloxacin and genistein nanostructured lipid carrier (GenNLC) modified by mPEG-PLA based on F127/F68 as hydrogel. Characterizations and release profiles of the drug delivery system were studied. In vitro functions were detected by CCK-8 test, immunofluorescence, wound-healing assay, real time-PCR and western blotting. The size of GenNLCs was 39.47 ± 0.69 nm in average with surface charges of − 4.32 ± 0.84 mV. The hydrogel gelation temperature and time were 32 °C, 20 s with a viscosity, hardness, adhesiveness and stringiness of 6.135 Pa.s, 54.0 g, 22.0 g, and 3.24 mm, respectively. Transmittance of the gel-release medium was above 90% (93.44 ± 0.33% to 100%) at range of 430 nm to 800 nm. Moxifloxacin released completely within 10 days. Fifty percent of dexamethasone released at a constant rate in the first week, and then released sustainably with a tapering down rate until day 30. Genistein released slowly but persistently with a cumulative release of 63% at day 40. The thermoresponsive hydrogel inhibited the proliferation, migration and epithelial-mesenchymal transition of SRA 01/04 cells, which were confirmed by testing CCK-8, wound-healing assay, western blot, real time-PCR (RT-PCR) and immunofluorescence. These results support this intracameral thermoresponsive in situ multi-drug delivery system with programmed release amounts and release profiles to cut down the need of eye drops for preventing inflammation or infection and to reduce posterior capsular opacification following cataract surgery.

Pedestrian trajectory prediction is an important task in practical applications such as automatic driving and surveillance systems. It is challenging to effectively model social interactions among pedestrians and capture temporal dependencies. Previous methods typically emphasized social interactions among pedestrians but ignored the temporal consistency of predictions and suffered from superfluous interactions by dense undirected graphs, resulting in a considerable deviance from reality. In addition, autoregressive approaches predicted future locations conditioning on previous predictions one by one, which would lead to error accumulation and time consuming. To address these issues, we present Non-autoregressive Sparse Transformer (NaST) networks for pedestrian trajectory prediction. Specifically, NaST models sparse spatial interactions and sparse temporal dependency via a sparse spatial transformer and a sparse temporal transformer separately. Different from previous predictions such as RNN-based approaches, the transformer decoder works in non-autoregressive pattern and predicts all the future locations at one time from a query sequence, which could avoid the error accumulation and be less computationally intensive. We evaluate our proposed method on the ETH and UCY datasets, and the experimental results show our method outperforms comparative state-of-the-art methods.

Salt pans are commonly found in coastal marshes and play a vital role in the marsh plant zonation. However, the correlation between these hypersaline zones and the marsh hydrological conditions have barely been characterized. This study numerically investigates the effects of evaporation rate, tidal amplitude, and marsh platform slope on salt pan formation, and found that salt pans can hardly grow in the intertidal zone due to regular tidal flushing, while tend to form in the lower supratidal zone, where evaporation is sustained. The accumulated salts create an upward salinity gradient that trigger downward unstable flow. The decreases of potential evaporation rate, tidal amplitude and/or marsh platform slope strengthen the hydraulic connection between the marsh surface and the underlying watertable, the key to sustaining evaporation, and therefore result in thickener and wider salt pans. These findings offer a deeper insight into the marsh eco-hydrology and guidance for their degradation prevention.

Object detection has drawn the attention of many researchers due to its wide application in computer vision-related applications. In this paper, a novel model is proposed for object detection. Firstly, a new neck is designed for the proposed detection model, including an efficient SPPNet (Spatial Pyramid Pooling Network), a modified NLNet (Non Local Network) and a lightweight adaptive feature fusion module. Secondly, the detection head with double residual branch structure is presented to reduce the delay of a decoupled head and improve the detection ability. Finally, these improvements are embedded in YOLOX as plug-and-play modules for forming a high-performance detector, EYOLOX (EfficientYOLOX). Extensive experiments demonstrate that the EYOLOX achieves significant improvements, which increases YOLOX-s from 40.5% to 42.2% AP on the MS COCO dataset with a single GPU. Moreover, the performance of the detection of EYOLOX also outperforms YOLOv6 and some SOTA methods with the same number of parameters and GFLOPs. In particular, EYOLOX has only been trained on the COCO-2017 dataset without using any other datasets, and only the pre-training weights of the backbone part are loaded.

In order to improve the accuracy of bolt positioning for industrial robots, this paper studies the bolt positioning system of industrial robots combined with multieye vision technology. This paper introduces three reconstruction algorithms ART, SIRT, and SART from the perspective of theory and implementation. Moreover, the projection matrix calculated with 0, 1 weighting, length weighting, and linear interpolation weighting is used for three reconstruction algorithms to carry out reconstruction experiments. In addition, this paper combines the actual working conditions of bolt positioning to construct the system of this paper and conducts system simulation research combined with the working conditions of the robot. The research shows that the bolt positioning system based on the multieye vision industrial robot proposed in this paper has a good performance in bolt positioning.

Travel patterns reflect the regularity of residents’ mobility, and it is a crucial factor to evaluate the reasonability of urban spatial structure and connectivity of road networks. Therefore, exploring travel patterns is of practical significance for urban planning, traffic management, and improvement of the operational efficiency of the transportation system. In this study, we apply the tensor model to explore travel patterns under temporal and spatial dimensions based on the license plate recognition (LPR) data collected from the Changsha city, China. As travel patterns are influenced by many variables, a method framework based on the tensor model is proposed to explore the influence of variables on travel characteristics. Firstly, we apply clustering algorithms and the principal component analysis method to extract main feature variables, which can achieve the purpose of dimensionality reduction and eliminate the complex collinearity among variables. Then, the tensor decomposition and reconstruction algorithms are performed based on extracted feature variables to analyze their influence on travel patterns. The experiments demonstrate the advantages of the proposed method framework.

In recent years, MOOC has become one of the most popular trends in global education. However, the completion of courses is still a great challenge for MOOCs. Thus, the evaluation of MOOCs is of great importance to online education. The goal of this research is to evaluate the interfaces of localized MOOC on PC platforms, specifically, to explore the role of culture on the interface design and its user experience (UX). Three studies were conducted in our research. In study 1, through an interview, we selected two localized MOOCs (Chinese University MOOC and Coursera), typical user tasks, and main factors of UX. In study 2, a content analysis was conducted to compare the interface design of the MOOCs selected in study 1. The results showed significant cultural differences in the interface design between the two localized MOOCs. In study 3, the UX was evaluated on Chinese University MOOC (CUM) and Coursera. It is found that the task-based UX of CUM is better than Coursera on four typical tasks. And the overall UX (usability, aesthetics, and pleasure) is much higher on CUM. The findings reveal the cultural influence on the interface design of localized CUM and Course. Accordingly, the Chinese user experience is better on CUM than Course. This research indicates a better UX of localized MOOCs also depends on culture-preferred interface design. Meanwhile, from the perspective of UX, we also provide some theoretical and practical implications for the enhancement of interface designs of localized MOOCs. Hopefully, this research can contribute to the design of localized MOOCs in China.

Patients with prolonged isolated thrombocytopenia (PT) after allogeneic stem cell transplantation (allo-SCT) are known to have a poor prognosis. However, there is no standard treatment for PT. The present study aimed to investigate the potential effect of a novel recombinant human thrombopoietin (rhTPO) in a cohort of patients with PT following allo-SCT. A total of 24 patients with PT (including delayed platelet engraftment and secondary failure of platelet recovery) were treated with rhTPO from July 1, 2016 to May 31, 2017. rhTPO injections were administered at 300 IU/kg/d for 28 consecutive days or until platelet counts were ≥ 50 × 109/L, independent of platelet transfusion. Response was defined as platelet recovery to ≥ 20 × 109/L for seven consecutive days without transfusion support during or within 7 days of the end of rhTPO treatment. All patients completed the 28 days of treatment, and none were withdrawn due to adverse effects. The overall response was 45.8%, which was significantly higher than historical data (12.2%, p

Extensive intratumoral heterogeneity (ITH) is believed to contribute to therapeutic failure and tumor recurrence, as treatment-resistant cell clones can survive and expand. However, little is known about ITH in triple-negative breast cancer (TNBC) because of the limited number of single-cell sequencing studies on TNBC. In this study, we explored ITH in TNBC by evaluating gene expression-derived and imaging-derived multi-region differences within the same tumor. We obtained tissue specimens from 10 TNBC patients and conducted RNA sequencing analysis of 2–4 regions per tumor. We developed a novel analysis framework to dissect and characterize different types of variability: between-patients (inter-tumoral heterogeneity), between-patients across regions (inter-tumoral and region heterogeneity), and within-patient, between-regions (regional intratumoral heterogeneity). We performed a Bayesian changepoint analysis to assess and classify regional variability as low (convergent) versus high (divergent) within each patient feature (TNBC and PAM50 subtypes, immune, stroma, tumor counts and tumor infiltrating lymphocytes). Gene expression signatures were categorized into three types of variability: between-patients (108 genes), between-patients across regions (183 genes), and within-patients, between-regions (778 genes). Based on the between-patient gene signature, we identified two distinct patient clusters that differed in menopausal status. Significant intratumoral divergence was observed for PAM50 classification, tumor cell counts, and tumor-infiltrating T cell abundance. Other features examined showed a representation of both divergent and convergent results. Lymph node stage was significantly associated with divergent tumors. Our results show extensive intertumoral heterogeneity and regional ITH in gene expression and image-derived features in TNBC. Our findings also raise concerns regarding gene expression based TNBC subtyping. Future studies are warranted to elucidate the role of regional heterogeneity in TNBC as a driver of treatment resistance.

Nowadays, the facial expression recognition (FER) task has transitioned from a laboratory-controlled scenario to in-the-wild conditions. However, recognizing facial expressions in the wild is challenging due to factors such as variant backgrounds, low-quality facial images, and the subjectiveness of annotators. Therefore, deep neural networks have increasingly been leveraged to learn discriminative representations for FER. In this work, we propose the Multi-cues Fusion Net (MCF-Net), a novel deep learning model with a two-stream structure for FER. Our model first proposes a two-stream coding network to extract face and scene representations. Then, an adaptive fusion module is employed to fuse the two different representations for final recognition. In the face coding stream, a Sparse Mask Attention Learning (SMAL) module is utilized to adaptively generate the corresponding sparse face mask according to the input image. Meanwhile, we employ a Multi-scale Attention (MSA) module for extracting fine-grained feature subsets, which can obtain richer multi-scale interaction information. In the scene coding stream, a Relational Attention (RA) module is applied to construct the hidden relationship between the face and contextual features of non-facial regions by capturing the pairwise similarity. In order to verify the effectiveness and accuracy of our model, a large number of experiments are carried out on two public large-scale static facial expression image datasets, CAER-S and NCAER-S. By comparing the performance of our MCF-Net with other methods, the proposed model achieves superior results on two in-the-wild FER benchmarks: CAER-S with an accuracy of 81.82% and NCAER-S with an accuracy of 45.59%.

Gut acute graft-versus-host disease (aGVHD) is a serious complication after allogeneic hematopoietic stem cell transplantation (allo-HSCT) and is associated with high mortality. Mucosa-associated invariant T (MAIT) cells are a group of innate-like T cells enriched in the intestine that can be activated by riboflavin metabolites from various microorganisms. However, little is known about the function or mechanism of action of MAIT cells in the occurrence of gut aGVHD in humans. In our study, multiparameter flow cytometry (FCM) was used to evaluate the number of MAIT cells and functional cytokines. 16S V34 region amplicon sequencing analysis was used to analyze the intestinal flora of transplant patients. In vitro stimulation and coculture assays were used to study the activation and function of MAIT cells. The number and distribution of MAIT cells in intestinal tissues were analyzed by immunofluorescence technology. Our study showed that the number and frequency of MAIT cells in infused grafts in gut aGVHD patients were lower than those in no-gut aGVHD patients. Recipients with a high number of MAITs in infused grafts had a higher abundance of intestinal flora in the early posttransplantation period (+14 days). At the onset of gut aGVHD, the number of MAIT cells decreased in peripheral blood, and the activation marker CD69, chemokine receptors CXCR3 and CXCR4, and transcription factors Rorγt and T-bet tended to increase. Furthermore, when gut aGVHD occurred, the proportion of MAIT17 was higher than that of MAIT1. The abundance of intestinal flora with non-riboflavin metabolic pathways tended to increase in gut aGVHD patients. MAIT cells secreted more granzyme B, tumor necrosis factor (TNF)-α, and interferon (IFN)-γ under the interleukin (IL)-12/IL-18 stimulation [non-T-cell receptor (TCR) signal] and secreted most of the IL-17 under the cluster of differentiation (CD)3/CD28 stimulation (TCR signal). MAIT cells inhibited the proliferation of CD4+ T cells in vitro. In conclusion, the lower number of MAIT cells in infused grafts was related to the higher incidence of gut aGVHD, and the number of MAIT cells in grafts may affect the composition of the intestinal flora of recipients early after transplantation. The flora of the riboflavin metabolism pathway activated MAIT cells and promoted the expression of intestinal protective factors to affect the occurrence of gut aGVHD in humans.

Mitochondrial fission factor (MFF) is an adapter that targets dynamin-related protein 1 from the cytosol to the mitochondria for fission. Loss-of-function MFF mutations cause encephalopathy due to defective mitochondrial and peroxisomal fission 2 (EMPF2). To elucidate the molecular mechanisms that were involved, we analyzed the functional effects of MFF depletion in deciduous teeth-derived dental pulp stem cells differentiating into dopaminergic neurons (DNs). When treated with MFF-targeting small interfering RNA, DNs showed impaired neurite outgrowth and reduced mitochondrial signals in neurites harboring elongated mitochondria. MFF silencing also caused mitochondrial Ca2+ accumulation through accelerated Ca2+ influx from the endoplasmic reticulum (ER) via the inositol 1,4,5-trisphosphate receptor. Mitochondrial Ca2+ overload led DNs to produce excessive reactive oxygen species (ROS), and downregulated peroxisome proliferator-activated receptor-gamma co-activator-1 alpha (PGC-1α). MFF was co-immunoprecipitated with voltage-dependent anion channel 1, an essential component of the ER-mitochondrial Ca2+ transport system. Folic acid supplementation normalized ROS levels, PGC-1α mediated mitochondrial biogenesis, and neurite outgrowth in MFF depleted DNs, without affecting their mitochondrial morphology or Ca2+ levels. We propose that MFF negatively regulates the mitochondrial Ca2+ influx from the ER. MFF-insufficiency recapitulated the EMPF2 neuropathology with increased oxidative stress and suppressed mitochondrial biogenesis. ROS and mitochondrial biogenesis might be potential therapeutic targets for EMPF2.

Immunity to the severe diarrheal disease cholera is largely mediated by lipopolysaccharide (LPS)-specific antibodies. However, the properties and protective mechanism of functionally relevant antibodies have not been well defined.

With the development of visual sensor equipment (e.g., personal smart phones, vehicle cameras, surveillance videos and camcorders), scene recognition technology has attracted much attention due to its latent applications in visual surveillance, intelligent traffic and aerial remote sensing. Although some progress has been made in the field of scene recognition in recent years, the complexity of scene images and the inadequate numbers of labeled data pose challenges in this area. Hence, to effectively fuse the multiple features of each image and employ the information of both labeled and unlabeled images for scene recognition, we proposed a semi-supervised multi-feature regression (SSMFR) model in this paper. The SSMFR model possesses three advantages. First, the model propagates the labels of labeled data to unlabeled data by utilizing graph-based semi-supervised learning techniques so that both the information regarding unlabeled data and labeled data can be exploited to gain better performance. Second, SSMFR employs multiple graphs to characterize the structures of multiple feature spaces and adaptively assigns the weight to different graphs. Therefore, SSMFR can efficiently preserve the manifold structure of samples in each feature space and adequately exploit the complementary information of multiple features. Moreover, SSMFR adopts a $l_{2,1}$ -norm constraint to learn a sparse and robust classifier for scene recognition. To solve the SSMFR model, we proposed a simple and efficient iterative update optimization scheme. Finally, we also proved the convergence of SSMFR by theoretical analysis and experiments. Experiments were conducted on several benchmark scene datasets, and the experimental results demonstrated that the proposed SSMFR model can obtain better performance for scene recognition than some other state-of-the-art algorithms.

Objective: Haploidentical stem cell transplantation (haplo-SCT) has demonstrated encouraging results in younger patients. There is also an increasing need for haplo-SCT in older patients. However, the high risk of treatment-related mortality (TRM) in older patients is still a major concern. We aimed to investigate a novel conditioning regimen (Bu/Flu/Cy/ATG) followed by haplo-SCT in older patients.Method: This prospective, single-arm clinical trial was performed at Peking University Institute of Hematology, China. Patients were enrolled if they were (1) diagnosed with acute leukemia or MDS; (2) without MSD and MUD, and with HID available; and (3) age ≥55 years. The Bu/Flu/Cy/ATG regimen consisted of the following agents: Ara-C (2 g/m2/day, injected i.v.) on days-10 and−9; BU (9.6 mg/kg, injected i.v. in 12 doses) on days-8,−7, and−6; Flu (30 mg/m2/day, injected i.v.) from day−6 to day−2; Cy (1 g/m2/day, injected i.v.) on days−5 and−4; semustine (250 mg/m2, orally) on day-3 and antithymocyte globulin (ATG) [2.5 mg/kg/day, rabbit, SangStat (Lyon, France)] on days−5,−4,−3, and−2. The primary endpoint was 1-year TRM.Results: From April 1, 2018 to April 10, 2020, a total of 50 patients were enrolled. All patients achieved neutrophil engraftment with complete donor chimerism. The cumulative incidence of grade 2-4 aGVHD at day-100 was 22.0%. The cumulative incidences of CMV viremia and EBV viremia on day 100 were 68.0 and 20.0%, respectively. The cumulative incidence of TRM at 1-year was 23.3%. and the cumulative incidence of relapse (CIR) at 1 year after transplantation was 16.5%. The overall survival (OS) and leukemia-free survival (LFS) at 1 year were 63.5 and 60.2%, respectively. The outcomes were also comparable with patients who received Bu/Cy/ATG regimen using a propensity score matching method.Conclusions: In conclusion, this study suggested that a novel conditioning regimen followed by haploidentical HSCT might be a promising option for older patients. The study was registered as a clinical trial.Clinical Trial Registration:www.ClinicalTrials.gov, identifier: NCT03412409.

Understanding cellular remodeling in response to mechanical stimuli is a critical step in elucidating mechanical activation of biochemical signaling pathways. Experimental evidence indicates that external stress-induced subcellular adaptation is accomplished through dynamic cytoskeletal reorganization. To study the interactions between subcellular structures involved in transducing mechanical signals, we combined experimental data and computational simulations to evaluate real-time mechanical adaptation of the actin cytoskeletal network. Actin cytoskeleton was imaged at the same time as an external tensile force was applied to live vascular smooth muscle cells using a fibronectin-functionalized atomic force microscope probe. Moreover, we performed computational simulations of active cytoskeletal networks under an external tensile force. The experimental data and simulation results suggest that mechanical structural adaptation occurs before chemical adaptation during filament bundle formation: actin filaments first align in the direction of the external force by initializing anisotropic filament orientations, then the chemical evolution of the network follows the anisotropic structures to further develop the bundle-like geometry. Our findings present an alternative two-step explanation for the formation of actin bundles due to mechanical stimulation and provide new insights into the mechanism of mechanotransduction.

Spatial cross-matching operation over geospatial polygonal datasets is a highly compute-intensive yet an essential task to a wide array of real-world applications. At the same time, modern computing systems are typically equipped with multiple processing units capable of task parallelization and optimization at various levels. This mandates for the exploration of novel strategies in the geospatial domain focusing on efficient utilization of computing resources, such as CPUs and GPUs. In this paper, we present a CPU-GPU hybrid platform to accelerate the cross-matching operation of geospatial datasets. We propose a pipeline of geospatial subtasks that are dynamically scheduled to be executed on either CPU or GPU. To accommodate geospatial datasets processing on GPU using pixelization approach, we convert the floating point-valued vertices into integer-valued vertices with an adaptive scaling factor as a function of the area of minimum bounding box. We present a comparative analysis of GPU enabled cross-matching algorithm implementation in CUDA and OpenACC accelerated C++. We test our implementations over Natural Earth Data and our results indicate that although CUDA based implementations provide better performance, OpenACC accelerated implementations are more portable and extendable while still providing considerable performance gain as compared to CPU. We also investigate the effects of input data size on the IO / computation ratio and note that a larger dataset compensates for IO overheads associated with GPU computations. Finally, we demonstrate that an efficient cross-matching comparison can be achieved with a cost-effective GPU.

Isolated cerebral mucormycosis is a clinical type of mucormycosis that is estimated to account for 8% of all mucormycosis cases. The clinical symptoms of isolated cerebral mucormycosis are elusive, and thus conventional techniques often lake sensitivity and specificity. Moreover, cultures are often negative, even when direct microscopy examination is positive. Although histopathology will probably remain the gold standard for the diagnosis of mucormycosis, obtaining a biopsy specimen is not always feasible in most vulnerable populations. Thus, molecular approaches are currently used as an advantageous assistant examination method to improve the early identification of the causative agent and subsequently guide therapy to improve the prognosis of patients. Here, we report a case of isolated cerebral mucormycosis caused by Rhizopus microspores in a healthy young adult that was identified using next-generation sequencing technology.

Technologies for digitizing tissues provide important quantitative data for liver histopathology investigation. We aimed to assess liver fibrosis degree with quantitative morphometric measurements of histopathological sections utilizing digital image analysis (DIA) and to further investigate if a correlation with histopathologic scoring (Scheuer staging) exists. A retrospective study of patients with at least two post-liver transplant biopsies having a Scheuer stage of ≤ 2 at baseline were gathered. Portal tract fibrotic percentage (%) and size (μm2) were measured by DIA, while clinical fibrosis score was measured by the Scheuer system. Correlations between DIA measurements and Scheuer scores were computed by Spearman correlation analysis. Differences between mean levels of fibrosis (score, size, and percentage) at baseline versus second visit were computed by Student's t-test. P values < 0.05 were considered significant. Of 22 patients who met the study criteria, 54 biopsies were included for analysis. Average levels ±standard error [S.E.] of portal tract fibrotic percentage (%) and size (μm2) progressed from 46.5 ± 3.6% at baseline to 61.8 ± 3.8% at the second visit (P = 0.005 by Student's t-test), and from 28,075 ± 3,232 μm2 at base line to 67,146 ± 10,639 μm2 at the second visit (P = 0.002 by Student's t-test), respectively. Average levels of Scheuer fibrosis scores progressed from 0.55±0.19 at baseline to 1.14±0.26 at the second visit (P = 0.02 by Student's t-test). Portal tract fibrotic percentage (%) and portal tract fibrotic size were directly correlated with clinical Scheuer fibrosis stage, with Spearman correlation coefficient and P value computed as r = 0.70, P < 0.0001 and r = 0.41, P = 0.002, respectively. Digital quantitative assessment of portal triad size and fibrosis percentage demonstrates a strong correlation with visually assessed histologic stage of liver fibrosis and complements the standard assessment for allograft monitoring, suggesting the utility of future WSI analysis.

Background. Diabetic nephropathy (DN) is an important cause of end-stage renal disease and is recognized as a public health problem worldwide. However, there have been no nationwide surveys of DN prevalence in China. This study is aimed at estimating the pooled prevalence of DN among patients with type 2 diabetes in China. Methods. Published studies on the prevalence of DN among patients with type 2 diabetes published from January 1980 to October 2019 were systematically reviewed using PubMed, Embase, Google Scholar, Chinese Wanfang databases, and Chinese National Knowledge Infrastructure. The pooled prevalence of DN was estimated with the random effects model using R software. Prevalence estimates were also stratified by study design, methodological approach, and study population characteristics. Results. Thirty studies with a total of 79,364 participants were included in our study. The overall pooled prevalence of DN was 21.8% [95% confidence interval (CI): 18.5-25.4%]. Subgroup analysis found that the prevalence of DN varied significantly according to different DM and DN diagnostic criteria (P

Summary: Cancer stem cells promote neoplastic growth, in part by deregulating asymmetric cell division and enhancing self-renewal. To uncover mechanisms and potential therapeutic targets in glioma stem cell (GSC) self-renewal, we performed a genetic suppressor screen for kinases to reverse the tumor phenotype of our Drosophila brain tumor model and identified dCdk5 as a critical regulator. CDK5, the human ortholog of dCdk5 (79% identity), is aberrantly activated in GBMs and tightly aligned with both chromosome 7 gains and stem cell markers affecting tumor-propagation. Our investigation revealed that pharmaceutical inhibition of CDK5 prevents GSC self-renewal in vitro and in xenografted tumors, at least partially by suppressing CREB1 activation independently of PKA/cAMP. Finally, our TCGA GBM data analysis revealed that CDK5, stem cell, and asymmetric cell division markers segregate within non-mesenchymal patient clusters, which may indicate preferential dependence on CDK5 signaling and sensitivity to its inhibition in this group. : Glioblastoma is the most common and deadliest form of brain tumor and can withstand current therapies due to the resilience of glioma stem cells (GSCs). Mukherjee et al. examine Cdk5 and its role in promoting stemness in asymmetric division of brain tumor stem cells in Drosophila and mice. Keywords: glioma stem cells, CDK5, asymmetric cell division, self-renewal, CREB1, GBM non-mesenchymal subtypes

Scene recognition is a significant and challenging problem in the field of computer vision. One of the principal bottlenecks in applying machine learning techniques to scene recognition tasks is the requirement of a large number of labeled training data. However, labeling massive training data manually (especially labeling images and videos) is very expensive in terms of human time and effort. In this paper, we present a novel multicriteria-based active discriminative dictionary learning (M-ADDL) algorithm to reduce the human annotation effort and create a robust scene recognition model. The M-ADDL algorithm possesses three advantages. First, M-ADDL introduces an active learning strategy into the discriminative dictionary learning model so that the performance of discriminative dictionary learning can be improved when the number of labeled samples is small. Second, different from most existing active learning methods that measure either the informativeness or representativeness of unlabeled samples to select useful samples for expanding the training dataset, M-ADDL employs both informativeness and representativeness to query useful unlabeled samples and utilizes the manifold-preserving ability of unlabeled samples as an additional sample selection criterion. Finally, a more effective representativeness criterion is presented based on the reconstruction coefficients of the samples. The experimental results of four standard scene recognition databases demonstrate the feasibility and validity of the proposed M-ADDL algorithm.

Face recognition has a great potential to play an important role in computer vision field. However, the majority of face recognition methods are based on the low-level features, which may not yield good results. Inspired by a simple deep learning model principal component analysis network (PCANet), we propose a novel deep learning network called circular symmetrical Gabor filter (2D)2PCA neural networks [CSGF(2D)2PCANet]. Previous models used in face recognition have three major issues of data redundancy, computation time, and no rotation invariance. We introduce the CSGF to address these issues. Two-directional 2-D PCA [(2D)2PCA] is used in feature extraction stage. Binary hashing, blockwise histograms, and linear SVM are used for the output stage. The proposed CSGF (2D)2PCANet learns highlevel features and provides more recognition information during the training phase, which may result in a higher recognition rate when testing the sample. We tested the proposed method on XM2VTS, ORL, AR, Extend Yale B, and LFW databases. Test results show that the CSGF (2D)2PCANet is more robust to the variation of occlusion, illumination, pose, noise, and expression, which is a promising method in face recognition.

In recent years, graph-based dimensionality reduction methods became increasingly more significant since they have been successfully applied in various computer vision and machine learning problems. The key point in graph-based dimensionality reduction methods is how to construct an appropriate graph to reflect the underlying distribution of data. However, most existing methods usually consider graph construction and dimensionality reduction as two separate processes. To overcome this limitation, a multiple locality-constrained graph optimization for dimensionality reduction (MLGODR) algorithm is proposed in this paper. The proposed MLGODR possesses two characteristics. First, MLGODR integrates graph optimization and dimensionality reduction into a unified framework. Thus, a graph that characterizes the distribution of input data and a matrix that projects the input data into a low-dimensional subspace can be learned simultaneously. Second, to better exploit the local structure of input data, a locality constraint that adaptively combines multiple distance measurements is introduced into our objective function. Moreover, an effective updating algorithm is also designed to solve the proposed MLGODR. Extensive experiments are performed on four image databases and four UCI data sets. The experimental results demonstrate that our method outperforms the compared approaches in both classification and cluster tasks.

Graph Convolutional Networks (GCNs) have attracted a lot of attention and shown remarkable performance for action recognition in recent years. For improving the recognition accuracy, how to build graph structure adaptively, select key frames and extract discriminative features are the key problems of this kind of method. In this work, we propose a novel Adaptive Attention Memory Graph Convolutional Networks (AAM-GCN) for human action recognition using skeleton data. We adopt GCN to adaptively model the spatial configuration of skeletons and employ Gated Recurrent Unit (GRU) to construct an attention-enhanced memory for capturing the temporal feature. With the memory module, our model can not only remember what happened in the past but also employ the information in the future using multi-bidirectional GRU layers. Furthermore, in order to extract discriminative temporal features, the attention mechanism is also employed to select key frames from the skeleton sequence. Extensive experiments on Kinetics, NTU RGB+D and HDM05 datasets show that the proposed network achieves better performance than some state-of-the-art methods.

An online community is one of the important ways for people with mental disorders to receive assistance and obtain support. This study aims to help users with mental disorders to obtain more support and communication through online communities, and to provide community managers with the possible influence mechanisms based on the information adoption model. We obtained a total of 49,047 posts of an online mental health communities in China, over a 40-day period. Then we used a combination of text mining and empirical analysis. Topic and sentiment analysis were used to derive the key variables—the topic of posts that the users care about most, and the emotion scores contained in posts. We then constructed a theoretical model based on the information adoption model. As core independent variables of information quality, on online mental health communities, the topic of social experience in posts (0.368 ***), the topic of emotional expression (0.353 ***), and the sentiment contained in the text (0.002 *) all had significant positive relationships with the number of likes and reposts. This study found that the users of online mental health communities are more attentive to the topics of social experience and emotional expressions, while they also care about the non-linguistic information. This study highlights the importance of helping community users to post on community-related topics, and gives administrators possible ways to help users gain the communication and support they need.

Recently, deep learning-based image deblurring and deraining have been well developed. However, most of these methods fail to distill the useful features. What is more, exploiting the detailed image features in a deep learning framework always requires a mass of parameters, which inevitably makes the network suffer from a high computational burden. We propose a lightweight fusion distillation network (LFDN) for image deblurring and deraining to solve the above problems. The proposed LFDN is designed as an encoder–decoder architecture. In the encoding stage, the image feature is reduced to various small-scale spaces for multi-scale information extraction and fusion without much information loss. Then, a feature distillation normalization block is designed at the beginning of the decoding stage, which enables the network to distill and screen valuable channel information of feature maps continuously. Besides, an information fusion strategy between distillation modules and feature channels is also carried out by the attention mechanism. By fusing different information in the proposed approach, our network can achieve state-of-the-art image deblurring and deraining results with a smaller number of parameters and outperform the existing methods in model complexity.