Abstract Background Deep learning algorithms significantly improve the accuracy of pathological image classification, but breast cancer classification using only single-mode images still cannot meet needs clinical practice. Inspired by real scenario pathologists reading for diagnosis, we integrate and structured data extracted from electronic medical record (EMR) to further classification. Methods In this paper, propose a new richer fusion network benign malignant based on multimodal data....

Advances in snow-cover mapping techniques have resulted more accurate estimation of fractional snow cover (FSC) areas with no vegetation; however, vegetation interference limits the accuracy available information from satellite observations. The aim this study was to develop a robust and enhanced FSC-retrieval algorithm using Moderate Resolution Imaging Spectroradiometer (MODIS) surface reflectance data for vegetated areas. experiments were conducted North America, where is complex...

Modulation format identification (MFI) is one of the most critical functions embedded in digital coherent receivers elastic optical networks (EONs). In view inherent amplitude and phase characteristics received signals, different modulation formats exhibit a set notable features polar coordinate system, based on which an MFI scheme incorporating Gaussian weighted k-nearest neighbors (KNN) algorithm was proposed to identify polarization division multiplexed...

Ice crevasses are pervasive features across the Arctic and Antarctic ice sheets. These deep, open fractures in surface serve as critical conduits for transporting meltwater into englacial system, significantly impacting sheet hydrology stability. Accurate mapping of spatial temporal distribution is vital advancing our understanding dynamics their evolution. Remote sensing technology provides a robust platform to achieve this purpose, while rapid advancement machine learning algorithms offers...

Crevasses are formed by glacier movement and the stresses within ice. Knowledge of crevasses’ distribution is critical for understanding ice shelf stability. In this study, we propose an automated crevasse extraction framework based on Sentinel-1 SAR imagery improved U-Net network. The spatial crevasses Antarctic shelves in 2020 was mapped with a resolution ~40 m, characteristics Nickerson Ice Shelf, Jelbart Amery Thwaites Glacier, Shackleton Shelf were analyzed. results indicated accuracy...

Abstract The variation of polar sea ice is an indicator environmental change, which plays important role in the study regional and global climate change. In this paper, latest data sets from European Space Agency Climate Change Initiative were firstly combined to comprehensively analyze spatial temporal Arctic concentration (SIC) thickness (SIT) 2002 2017. results show that during period, SIC Kara Sea Barents decreases most significantly, reaching −1.11%·year −1 . annual average extent (SIE)...

The capability to grasp the target object is significant for manipulating robotic systems offer better services, and it still challenging under occlusion. This paper proposes a novel vision-based grasping method with SSD-based detector, an image inpainting recognition network (IRNet), deep guidance (DgGNet). Based on clustering of point cloud, IRNet combination three-stage MobileNet v2 introduced detect occluded that cannot be found by detector. Then, best grasped obtained DgGNet, which...

ABSTRACT Antarctica's response to climate change varies greatly both spatially and temporally. Surface melting impacts mass balance also lowers surface albedo. We use a 43-year record (from 1978 2020) of Antarctic snow melt seasons from space-borne microwave radiometers with machine-learning algorithm show that the onset end season are being delayed. Granger-causality analysis shows is delayed due increased heat flux ocean atmosphere at minimum sea-ice extent warming oceans. Melt...

Visual simultaneously localization and mapping (SLAM) is important for self-localization environment perception of service robots, where semantic SLAM can provide a more accurate result map with abundant information. In this article, we propose real-time PO-SLAM approach the combination both point object measurements. With point–point association in ORB-SLAM2, also consider point–object based on segmentation object–object association, employed by combining detection depth histogram. Also,...

Simultaneous localization and mapping is an indispensable yet challenging direction for mobile robots. Attracted by 3D LiDAR with accurate depth information robustness to illumination variations, many SLAM methods based on scan-to-map matching have been developed. However, there a critical issue of existing approaches, where large dense map generally required achieve satisfactory accuracy, leading low efficiency matching. To solve this problem, in letter, we propose novel directed geometry...

LiDAR odometry (LO) has gained popularity in recent years due to accurate depth measurement and robustness illumination. Typically, the solutions based on scan-to-map matching mainly optimize current pose. To further reduce accumulated error of pose estimation, fixed-lag smoothing that optimizes fixed-size poses simultaneously by corresponding point features multiple frames becomes necessary. The integration with LO still needs exploration. In this paper, a general module is proposed, which...

Abstract Background Rhizosphere fungi play an important role in plant community dynamics and biogeochemical cycling. While the drivers of fungal assembly have been studied varied ecosystems, it is still unclear how these processes function for rhizosphere soil temperate forests. Furthermore, unknown whether relative contributions determinants remain consistent or vary across ecological guilds. This study used high-throughput next-generation sequencing to characterize communities 247 samples...

Terrain traversability analysis is a challenging problem for mobile robots to adapt complex environments, including the detection of cluttered obstacles, potholes, or even slopes. With accurate distance information, using sensors such as three-dimensional light and ranging (LiDAR) terrain description becomes preferred choice. In this article, method based on elevation grid map presented. After generated, ground segmented with aid height difference kernel non-ground grids in are then...

Abstract The optimal use of hemispheric‐scale snow depth (SD) products for various hydrometeorological applications requires a comprehensive assessment their quality. Most previous validation studies SD adopted in situ observations as the ground truth, which may cause representativeness errors due to spatial scale mismatch between point‐based measurements and grid‐based products. extended triple collocation (ETC) technique is powerful tool estimate uncertainty three independent data sets...

The start date of vegetation growing season (SOS) is generally considered as an essential indicator to reflect growth condition. To date, relatively little research has explored the combined effects temperature and precipitation on responses forest spring phenology snow cover in detail. investigate this, we applied developed plant index (PPI) derived from Moderate Resolution Image Spectroradiometer (MODIS) estimate SOS over Northeast China 2004 2018, relationship between climate variables,...

Localization is a fundamental prerequisite, no matter whether single robot or multirobot system, where light detection and ranging (LiDAR) odometry has attracted great interest with accurate depth information robustness to illumination variations. In this article, novel 3-D LiDAR approach based on sparse geometric proposed. Different from map-based methods point features, we concern significant line plane features eigenvalues of neighboring points. Furthermore, line-to-line plane-to-plane...

A decade-long pronounced increase in temperatures the Arctic resulted a global warming hotspot over Greenland ice sheet (GrIS). Associated changes cryosphere were consequence and led to demand for monitoring glacier changes, which are one of major parameters analyze responses GrIS climate change. Long-term altimetry data (e.g. ICESat, CryoSat-2, ICESat-2) can provide elevation different periods, many methods have been developed alone obtain changes. In this work, we provided long-term change...

To develop and validate a midwifery-led task list in the task-shifting context.An extensive literature review followed by two-iterative Delphi survey.Twenty university hospitals, three non-university hospitals four colleges from nine provincial regions of China.Purposive non-probability sampling national panel experts maternal healthcare, obstetrics, nursing midwifery. Experts were asked to rate each midwifery service item regarding importance, feasibility advancement on 5-point scale, order...

Introduction Bladder cancer (BLCA) is the ninth most common malignancy worldwide and fourth in men. Copper levels are significantly altered patients with thyroid, breast, lung, cervical, ovarian, pancreatic, oral, gastric, bladder, prostate cancers. Outcomes can be predicted by constructing signatures using lncRNA-related genes associated outcomes. Methods We identified lncRNAs related to outcomes, those differentially expressed bladder cancer, cuproptosis-related from TCGA. intersection...

High-efficiency high-temperature superconducting (HTS) induction heating equipment has been regarded as a promising application for thermal processing of metal material, in which an HTS magnet generates high magnetic field to form the background iron core and rotating billet is heated by eddy current process temperature. Efficiency analysis traditional new structures are studied this paper. A type structure proposed. Through parallel connection multiple circuits, it can make more effective...