Search

Neural Scene Flow Prior (NSFP) is of significant interest to the vision community due to its inherent robustness to out-of-distribution (OOD) effects and its ability to deal with dense lidar points. The approach utilizes a coordinate neural network to estimate scene flow at runtime, without any training. However, it is up to 100 times slower than current state-of-the-art learning methods. In other applications such as image, video, and radiance function reconstruction innovations in speeding up the runtime performance of coordinate networks have centered upon architectural changes. In this paper, we demonstrate that scene flow is different -- with the dominant computational bottleneck stemming from the loss function itself (i.e., Chamfer distance). Further, we rediscover the distance transform (DT) as an efficient, correspondence-free loss function that dramatically speeds up the runtime optimization. Our fast neural scene flow (FNSF) approach reports for the first time real-time performance comparable to learning methods, without any training or OOD bias on two of the largest open autonomous driving (AV) lidar datasets Waymo Open and Argoverse., 17 pages, 10 figures, 6 tables

BackgroundLong recovery time, large scar, postoperative swelling and pain are possible side effects of open reduction internal fixation (ORIF) for tarsal navicular fractures. Early exercise instruction is made possible by the use of an intraoperative robot-assisted percutaneous invasive closed reduction internal fixation. The goal of the trial was to determine whether percutaneous screw internal fixation with robot assistance might be used to treat navicular fractures.Methods27 patients with navicular fractures had surgical treatment between June 2019 and December 2021. Of those, 20 instances were treated with ORIF, while 7 cases had robot-assisted percutaneous screw internal fixation. At the final follow-up, the American Orthopaedic Foot & Ankle Society (AOFAS) hindfoot score and the visual analogue scale (VAS) score were compared to determine outcomes and function.ResultsFollow-up was obtained in all 27 patients after surgery, with a mean follow-up time of 21.81 months, ranging from 15 to 29 months . In the 7 instances of robot-assisted group, percutaneous guide wire insertion and screw placement only needed one attempt and the depth and position of the implant were both satisfactory. In the ORIF group, there were two patients who sustained cutaneous nerve injuries. The AOFAS score and the VAS score of the group receiving robot-assisted navigation percutaneous screw fixation were 92.25 ± 2.22 and 0.75 ± 0.25 respectively at the last follow-up, while 82.25 ± 7.15 and 0.50 ± 0.29 were the respective values for the ORIF group.ConclusionIntraoperative robot-assisted percutaneous closed reduction internal fixation for tarsal navicular fractures can accomplish exact localization of fracture site, reduce soft tissue damage and operative time. According to current view, this method offers fewer complications, a faster recovery after surgery, and more patient satisfaction.

Additional file 1. Supplementary figures and tables.

Platelet transfusion refractoriness (PTR) is an intractable issue in hematological patients, which increases bleeding risks and hospitalization costs to a great extent. We reviewed 108 patients with hematological diseases including acute leukemia, myelodysplastic syndrome, aplastic anemia, and others who received allogeneic hematopoietic stem cell transplantation (HSCT) from January 2019 through December 2020. After multivariable logistic regression, we found that splenomegaly (odds ratio [OR] = 26.98, p < .001) and JAK mutation (OR = 17.32, p = .024) were independent risk factors for PTR. During the period of transplantation, patients in the PTR group had a significantly higher platelet transfusion demand, which was reflected in the increased number of platelet transfusions (10.23 ± 6.696 vs. 5.06 ± 1.904, p < .001). After multivariate adjustment, PTR turned out to be independently associated with worse overall survival (hazard ratio = 2.794, 95% confidence interval = 1.083–7.207, p = .034). In conclusion, we found that splenomegaly and JAK gene mutation were independent risk factors for PTR in patients with hematological diseases. A history of PTR prior to allo-HSCT indicates a poor prognosis. What is the context?Platelet transfusion refractoriness is a critical issue, and it greatly increases bleeding risks and hospitalization costs.Patients with hematological diseases tend to develop PTR.PTR results from immune and nonimmune factors and the latter account for 80–90%.At present, there are few studies focused on the inducing factors of PTR, and the specific mechanism is not clear. Platelet transfusion refractoriness is a critical issue, and it greatly increases bleeding risks and hospitalization costs. Patients with hematological diseases tend to develop PTR. PTR results from immune and nonimmune factors and the latter account for 80–90%. At present, there are few studies focused on the inducing factors of PTR, and the specific mechanism is not clear. What is new?In this study, we investigated 108 patients with hematological disorders who received allogeneic HSCT from January 2019 to December 2020.We found that splenomegaly and JAK gene mutation were independent risk factors for PTR in patients with hematological diseases.PTR had a passive effect on the prognosis of patients after HSCT, as indicated by worse OS and a trend toward lower platelets after transplantation.PTR might affect megakaryocyte reconstitution after transplantation. In this study, we investigated 108 patients with hematological disorders who received allogeneic HSCT from January 2019 to December 2020. We found that splenomegaly and JAK gene mutation were independent risk factors for PTR in patients with hematological diseases. PTR had a passive effect on the prognosis of patients after HSCT, as indicated by worse OS and a trend toward lower platelets after transplantation. PTR might affect megakaryocyte reconstitution after transplantation. What is the impact?This study provides evidence that hematological patients with splenomegaly should be alert to the occurrence of PTR, which often indicates a worse prognosis of transplantation.Spleen reduction and JAK inhibitors in the treatment of PTR are worth exploring. This study provides evidence that hematological patients with splenomegaly should be alert to the occurrence of PTR, which often indicates a worse prognosis of transplantation. Spleen reduction and JAK inhibitors in the treatment of PTR are worth exploring. Abbreviations PLT: platelets; PTR: platelet transfusion refractoriness; HSCT: hematopoietic stem cell transplantation; OR: odds ratio; HR: hazard ratio; CI: confidence interval; IQR: interquartile range; SD: standard deviation; HLA: human leukocyte antigen; HPA: human platelet antigen; OS: overall survival; RFS: relapse free survival; PI: post-transfusion increment; PPR: percentage platelet recovery; CCI: corrected count increment; ICU: intensive care unit; AA: aplastic anemia; MDS: myelodysplastic syndrome; AML: acute myeloid leukemia; ALL: acute lymphocytic leukemia; CML: chronic myeloid leukemia; CMML: chronic myelomonocytic leukemia; MPN: myeloproliferative neoplasm; SI: splenic irradiation; Abs: antibodies; CR: complete remission; DAC: decitabine; GVHD: graft-versus-host disease; BM: bone marrow; PB: peripheral blood

Prolonged isolated thrombocytopenia (PT) is a common complication affecting the outcome of stem cell transplantation. In this study, we undertook a real-world study of 303 myelodysplastic syndrome (MDS) patients who received allogeneic hematopoietic stem cell transplantation (HSCT) between December 2007 and June 2018. 28.4% of MDS patients suffered from PT after HSCT. Survival analysis indicated that PT was associated with worse overall survival (OS) in MDS patients. The 2-year and 5-year OS in MDS patients with PT after HSCT were 49% and 47%, significantly worse than that of 68% and 60% in patients without PT (P=0.005). For RFS, patients with PT did not have an increased risk of disease relapse (P=0.964). After multivariate adjustment, PT was proved to be the independent risk factor associated with the worse OS (HR 1.49, 95% CI 1.00-2.21, P =0.048). We further analyzed risk factors associated with the occurrence of PT in MDS patients. Multiple logistic regression identified grade II-IV aGVHD, extensive chronic GVHD, hemorrhagic cystitis, and CMV activation as significant risk factors for developing PT. Among these variables, the Odds Ratio (OR) of grade II-IV aGVHD was the highest (P =0.001, OR: 2.65, 95% CI: 1.51-4.64). These data indicated the prognostic value of PT in MDS after HSCT. The identification of risk factors for PT may help improve patient management and lead to the design of effective treatment strategies.

It is well noted that coordinate-based MLPs benefit -- in terms of preserving high-frequency information -- through the encoding of coordinate positions as an array of Fourier features. Hitherto, the rationale for the effectiveness of these positional encodings has been mainly studied through a Fourier lens. In this paper, we strive to broaden this understanding by showing that alternative non-Fourier embedding functions can indeed be used for positional encoding. Moreover, we show that their performance is entirely determined by a trade-off between the stable rank of the embedded matrix and the distance preservation between embedded coordinates. We further establish that the now ubiquitous Fourier feature mapping of position is a special case that fulfills these conditions. Consequently, we present a more general theory to analyze positional encoding in terms of shifted basis functions. In addition, we argue that employing a more complex positional encoding -- that scales exponentially with the number of modes -- requires only a linear (rather than deep) coordinate function to achieve comparable performance. Counter-intuitively, we demonstrate that trading positional embedding complexity for network deepness is orders of magnitude faster than current state-of-the-art; despite the additional embedding complexity. To this end, we develop the necessary theoretical formulae and empirically verify that our theoretical claims hold in practice., arXiv admin note: substantial text overlap with arXiv:2107.02561

The rapid development of China’s e-commerce industry has led to the rise of China’s second-hand car e-commerce. With the increasingly rich trust theory and the rapid development of e-commerce platforms, the issue of online consumer trust has attracted more and more scholars’ attention. This paper takes China’s B2C second-hand car e-commerce platforms as an example, combines the second-hand car research conclusions and consumer trust theory, and conducts a systematic study on the formation of consumer trust in second-hand car e-commerce platforms. Based on the trust of individual consumers, system environment, website/APP and platform companies, this paper explores the influencing factors of consumer trust and constructs the influencing factors model of trust formation. The empirical study was conducted by using the structural equation model and multiple regression to verify the degree of fitting of the theoretical hypothesis and the model. The research results have a certain reference value for the development of second-hand car e-commerce platforms.

Additional file 2: Figure S2. A Venn diagram shows the overlap of 22 optimal hub genes between GEO differential genes and WGCNA grey gene set.

Additional file 1: Figure S1. Correlation between gene significance and related module membership. A. The correlation between gene significance and module membership in the gray module. B. The correlation between gene significance and module membership in the blue module. *p < 0.05, **p < 0.01, ***p < 0.001.

Additional file 3: Figure S3. Identification of candidate hub genes to predict prognosis for AML patients. A. Partial likelihood deviance for different numbers of variables revealed by the LASSO regression model. Red dots represent the partial likelihood deviance values. Gray lines represent the partial likelihood deviance ± standard error. B. Fourteen candidate genes with minimum lambda values were obtained by LASSO regression with tenfold cross-validation. C. Candidate genes were filtrated from SVM-RFE algorithms. D. Five optimal hub genes overlapped between LASSO and SVM-RFE algorithm were shown by Venn diagram. Abbreviations: LASSO, least absolute shrinkage and selection operator; SVM-RFE, support vector machine recursive feature elimination.

Additional file 6: Table S1. Clinical characteristics of 48 AML patients. Table S2. Basic information for the included GEO databases.

Additional file 4: Figure S4. The expression levels of DLC1, NFIB, DENND5B, TANC2, and ELAVL4 in AML patients with different survival status. Scatter plots of the five gene’s expression level in AML patients with different survival status based on GEO dataset. *p < 0.05, **p < 0.01, ***p < 0.001. Abbreviations: GEO, Gene Expression Omnibus.

Additional file 5: Figure S5. Kaplan-Meier survival curve according to the DLC1 expression level for 48 AML patients in our center. Kaplan-Meier plots was used to visualize the overall survival probability of 48 patients based on the expression of DLC1 at diagnosis in our center.

Before the deep learning revolution, many perception algorithms were based on runtime optimization in conjunction with a strong prior/regularization penalty. A prime example of this in computer vision is optical and scene flow. Supervised learning has largely displaced the need for explicit regularization. Instead, they rely on large amounts of labeled data to capture prior statistics, which are not always readily available for many problems. Although optimization is employed to learn the neural network, the weights of this network are frozen at runtime. As a result, these learning solutions are domain-specific and do not generalize well to other statistically different scenarios. This paper revisits the scene flow problem that relies predominantly on runtime optimization and strong regularization. A central innovation here is the inclusion of a neural scene flow prior, which uses the architecture of neural networks as a new type of implicit regularizer. Unlike learning-based scene flow methods, optimization occurs at runtime, and our approach needs no offline datasets -- making it ideal for deployment in new environments such as autonomous driving. We show that an architecture based exclusively on multilayer perceptrons (MLPs) can be used as a scene flow prior. Our method attains competitive -- if not better -- results on scene flow benchmarks. Also, our neural prior's implicit and continuous scene flow representation allows us to estimate dense long-term correspondences across a sequence of point clouds. The dense motion information is represented by scene flow fields where points can be propagated through time by integrating motion vectors. We demonstrate such a capability by accumulating a sequence of lidar point clouds., accepted by NeurIPS 2021 as "spotlight"

We report the light-induced modification of catalytic selectivity for photoelectrochemical CO₂ reduction in aqueous media using copper (Cu) nanoparticles dispersed onto p-type nickel oxide (p-NiO) photocathodes. Optical excitation of Cu nanoparticles generates hot electrons available for driving CO₂ reduction on the Cu surface, while charge separation is accomplished by hot-hole injection from the Cu nanoparticles into the underlying p-NiO support. Photoelectrochemical studies demonstrate that optical excitation of plasmonic Cu/p-NiO photocathodes imparts increased selectivity for CO₂ reduction over hydrogen evolution in aqueous electrolytes. Specifically, we observed that plasmon-driven CO₂ reduction increased the production of carbon monoxide and formate, while simultaneously reducing the evolution of hydrogen. Our results demonstrate an optical route toward steering the selectivity of artificial photosynthetic systems with plasmon-driven photocathodes for photoelectrochemical CO₂ reduction in aqueous media.

Objective To study the osteogenic potential of recombinant human bone morphogenetic protein 2 (rhBMP⁃2) combined with Bio ⁃ oss bone substitute in the implant restoration of bone defects in the anterior esthetic region. Methods Twelve patients who underwent the immediate placement of 20 implants with a bone augmentation procedure using rhBMP⁃2 and Bio⁃oss were included in this study. Changes in the height and thickness of the buccal bone over 6 months were measured, and the soft tissue was evaluated using the pink esthetic score (PES) after crown placement. Results All 20 implants were successfully osseointegrated, and the average increase in bone height was 1.9 mm; different degrees of bone height growth were observed for 17 (85%) implants sites. In one case, there was a severe bone fracture on the buccal side before the operation, resulting in bone plate resorption and decreased alveolar bone height. The bone height did not change significantly in 2 cases. The thickness of the buccal bone plate for all implants was greater than 1 mm. The average thickness was 1.9 mm, and the average PES was 9.8 points. Conclusion rhBMP ⁃ 2 combined with Bio⁃oss bone substitute has a preferable effect on the restoration of bone defects in the anterior esthetic area, and can achieve good aesthetic effect.

PointNet has recently emerged as a popular representation for unstructured point cloud data, allowing application of deep learning to tasks such as object detection, segmentation and shape completion. However, recent works in literature have shown the sensitivity of the PointNet representation to pose misalignment. This paper presents a novel framework that uses PointNet encoding to align point clouds and perform registration for applications such as 3D reconstruction, tracking and pose estimation. We develop a framework that compares PointNet features of template and source point clouds to find the transformation that aligns them accurately. In doing so, we avoid computationally expensive correspondence finding steps, that are central to popular registration methods such as ICP and its variants. Depending on the prior information about the shape of the object formed by the point clouds, our framework can produce approaches that are shape specific or general to unseen shapes. Our framework produces approaches that are robust to noise and initial misalignment in data and work robustly with sparse as well as partial point clouds. We perform extensive simulation and real-world experiments to validate the efficacy of our approach and compare the performance with state-of-art approaches. Code is available at https://github.com/vinits5/pointnet-registrationframework., 10 pages. arXiv admin note: substantial text overlap with arXiv:1908.07906

In order to evaluate the movements characteristics of aggregate particles in asphalt mixture under loading and temperature effects and to study the heterogeneity of asphalt mixture, rutting tests were conducted. Firstly, AC-20, AC-25, SMA-20 and SMA-25 specimens with the baseline labeled on sections were selected to conduct rutting tests. Then the section images at different loading times and different temperature were handled by digital image processing to study the aggregate particle movement behavior by means of aggregate particle movement parameters. Finally, the asphalt mixture heterogeneity was analyzed by the defined heterogeneity area and homogeneity area. The results show that the longitudinal displacement of aggregates is significantly greater than the lateral displacement of aggregates under the standard axle load. The aggregate particles irregular movement area for gap-graded asphalt mixture is greater than dense-graded asphalt mixture at the same temperature and loading. And the temperature or loading increasing can lead to the increase of the heterogeneity area.

The electrodeposition of graphene has drawn considerable attention due to its appealing applications for sensors, supercapacitors and lithium-ion batteries. However, there are still some limitations in the current electrodeposition methods for graphene. Here, we present a novel electrodeposition method for the direct deposition of reduced graphene oxide (rGO) with chitosan. In this method, a 2-hydroxypropyltrimethylammonium chloride-based chitosan-modified rGO material was prepared. This material disperses homogenously in the chitosan solution, forming a deposition solution with good dispersion stability. Subsequently, the modified rGO material was deposited on an electrode through codeposition with chitosan, based on the coordination deposition method. After electrodeposition, the homogeneous, deposited rGO/chitosan films can be generated on copper or silver electrodes or substrates. The electrodeposition method allows for the convenient and controlled creation of rGO/chitosan nanocomposite coatings and films of different shapes and thickness. It also introduces a new method of creating films, as they can be peeled completely from the electrodes. Moreover, this method allows for a rGO/chitosan film to be deposited directly onto an electrode, which can then be used for electrochemical detection.

Asphalt and filler interaction plays a key role on the performances of asphalt mastics and mixtures, while the asphalt and filler interaction ability is in turn affected by the material characteristics. In order to analyse the effects of material characteristics on asphalt and filler interaction, two kinds of matrix asphalt binders and three kinds of fillers were used to prepare asphalt mastics with different filler volume fractions. The dynamic shear rheological properties of these asphalt mastics were measured. The evaluation index K–B–G* was calculated subsequently to analyse the effects of material characteristics on the asphalt and filler interaction ability. The results show that filler volume fraction, filler particle size, SiO2 content and asphalt components had various levels of effects on the asphalt and filler interaction ability. Variance analysis was then conducted to rank these influencing factors from high to low as: (1) filler particle size, (2) asphaltene and resins content, (3) f...

A reliable UMAT subroutine of selected constitutive model of asphalt mixture is developed and applied to characterize the three-stage rutting behavior of asphalt pavement by using ABAQUS. It is indicated that the heavier traffic load not only induces the greater rutting depth, but also accelerates the rutting rate, causing the pavement earlier enter into rutting failure with shortened service life. The damage evolution and axle-load conversion are further discussed. Results show that the pavement subjected to heavier axle loads has the greater damage and faster damage evolution. The axle-load conversion index is 5.97 according to rut equivalence principle.

PointNet has recently emerged as a popular representation for unstructured point cloud data, allowing application of deep learning to tasks such as object detection, segmentation and shape completion. However, recent works in literature have shown the sensitivity of the PointNet representation to pose misalignment. This paper presents a novel framework that uses the PointNet representation to align point clouds and perform registration for applications such as tracking, 3D reconstruction and pose estimation. We develop a framework that compares PointNet features of template and source point clouds to find the transformation that aligns them accurately. Depending on the prior information about the shape of the object formed by the point clouds, our framework can produce approaches that are shape specific or general to unseen shapes. The shape specific approach uses a Siamese architecture with fully connected (FC) layers and is robust to noise and initial misalignment in data. We perform extensive simulation and real-world experiments to validate the efficacy of our approach and compare the performance with state-of-art approaches.

Jinping Neutrino Experiment (Jinping) is proposed to significantly improve measurements on solar neutrinos and geoneutrinos in China Jinping Laboratory - a lab with a number of unparalleled features, thickest overburden, lowest reactor neutrino background, etc., which identify it as the world-best low-energy neutrino laboratory. The proposed experiment will have target mass of 4 kilotons of liquid scintillator or water-based liquid scintillator, with a fiducial mass of 2 kilotons for neutrino-electron scattering events and 3 kilotons for inverse-beta interaction events. A number of initial sensitivities studies have been carried out, including on the transition phase for the solar neutrinos oscillation from the vacuum to the matter effect, the discovery of solar neutrinos from the carbon-nitrogen-oxygen (CNO) cycle, the resolution of the high and low metallicity hypotheses, and the unambiguous separation on U and Th cascade decays from the dominant crustal anti-electron neutrinos in China.

A re-analysis of a heavy charged particle production event observed at the cloudy chamber of the Yunnan Cosmic Ray Station (YCRS) in 1972 indicates that the mysterious heavy particle may be identified as a supersymmetric (SUSY) particle produced by a bombard of a neutral SUSY cosmic ray particle on proton. Based on the assumption, following the literatures that the neutral SUSY particle which constitutes the main fraction of the cold dark matter is scalar neutrino (sneutrino) or neutralino (photino), we evaluate the flux of such SUSY particles which gain sufficient energies via elastic scattering with charged cosmic particles on the way to an Earth detector and the capture rates in both the sneutrino and photino cases respectively. The errors appearing in the study are briefly discussed and this work may provide a basis of designing cosmic ray detectors to search for SUSY particles from the heaven.

Sex therapy in China has developed over the past three years, despite being largely unknown by the public and many clinical practitioners. This paper describes a short-term, comprehensive sex therapy, employing cognitive-behavioural techniques which were used to treat 21 couples with erectile dysfunction, where the remission rate was 61.8%. The authors believe that the technique of couple evaluation proved to be an objective measure of change. There was a close relationship between the outcome of therapy and the following factors: the course of therapy, the wife's attitude to sex, the level of the patients' psychosocial problems, and the state of the marriage itself.

We report the light-induced modification of catalytic selectivity for photoelectrochemical CO2 reduction in aqueous media using copper (Cu) nanoparticles dispersed onto p-type nickel oxide (p-NiO) photocathodes. Optical excitation of Cu nanoparticles generates hot electrons available for driving CO2 reduction on the Cu surface, while charge separation is accomplished by hot-hole injection from the Cu nanoparticles into the underlying p-NiO support. Photoelectrochemical studies demonstrate that optical excitation of plasmonic Cu/p-NiO photocathodes imparts increased selectivity for CO2 reduction over hydrogen evolution in aqueous electrolytes. Specifically, we observed that plasmon-driven CO2 reduction increased the production of carbon monoxide and formate, while simultaneously reducing the evolution of hydrogen. Our results demonstrate an optical route toward steering the selectivity of artificial photosynthetic systems with plasmon-driven photocathodes for photoelectrochemical CO2 reduction in aqueous media.