Abstract: A new coronavirus, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has recently emerged to cause a human pandemic Although molecular diagnostic tests were rapidly developed, serologic assays are still lacking, yet urgently needed Validated serologic assays are needed for contact tracing, identifying the viral reservoir, and epidemiologic studies We developed serologic assays for detection of SARS-CoV-2 neutralizing, spike protein-specific, and nucleocapsid-specific antibodies Using serum samples from patients with PCR-confirmed SARS-CoV-2 infections, other coronaviruses, or other respiratory pathogenic infections, we validated and tested various antigens in different in-house and commercial ELISAs We demonstrated that most PCR-confirmed SARS-CoV-2-infected persons seroconverted by 2 weeks after disease onset We found that commercial S1 IgG or IgA ELISAs were of lower specificity, and sensitivity varied between the 2 assays;the IgA ELISA showed higher sensitivity Overall, the validated assays described can be instrumental for detection of SARS-CoV-2-specific antibodies for diagnostic, seroepidemiologic, and vaccine evaluation studies

Abstract: We introduce a novel technique for knowledge transfer, where knowledge from a pretrained deep neural network (DNN) is distilled and transferred to another DNN. As the DNN performs a mapping from the input space to the output space through many layers sequentially, we define the distilled knowledge to be transferred in terms of flow between layers, which is calculated by computing the inner product between features from two layers. When we compare the student DNN and the original network with the same size as the student DNN but trained without a teacher network, the proposed method of transferring the distilled knowledge as the flow between two layers exhibits three important phenomena: (1) the student DNN that learns the distilled knowledge is optimized much faster than the original model, (2) the student DNN outperforms the original DNN, and (3) the student DNN can learn the distilled knowledge from a teacher DNN that is trained at a different task, and the student DNN outperforms the original DNN that is trained from scratch.

Abstract: Deep neural networks have emerged as a widely used and effective means for tackling complex, real-world problems. However, a major obstacle in applying them to safety-critical systems is the great difficulty in providing formal guarantees about their behavior. We present a novel, scalable, and efficient technique for verifying properties of deep neural networks (or providing counter-examples). The technique is based on the simplex method, extended to handle the non-convex Rectified Linear Unit (ReLU) activation function, which is a crucial ingredient in many modern neural networks. The verification procedure tackles neural networks as a whole, without making any simplifying assumptions. We evaluated our technique on a prototype deep neural network implementation of the next-generation airborne collision avoidance system for unmanned aircraft (ACAS Xu). Results show that our technique can successfully prove properties of networks that are an order of magnitude larger than the largest networks verified using existing methods.

Abstract: Infertility affects an estimated 15% of couples globally, amounting to 485 million couples Males are found to be solely responsible for 20-30% of infertility cases and contribute to 50% of cases overall However, this number does not accurately represent all regions of the world Indeed, on a global level, there is a lack of accurate statistics on rates of male infertility Our report examines major regions of the world and reports rates of male infertility based on data on female infertility Our search consisted of systematic reviews, meta-analyses, and population-based studies by searching the terms “epidemiology, male infertility, and prevalence” We identified 16 articles for detailed study We typically used the assumption that 50% of all cases of infertility are due to female factors alone, 20-30% are due to male factors alone, and the remaining 20-30% are due to a combination of male and female factors Therefore, in regions of the world where male factor or rates of male infertility were not reported, we used this assumption to calculate general rates of male factor infertility Our calculated data showed that the distribution of infertility due to male factor ranged from 20% to 70% and that the percentage of infertile men ranged from 2·5% to 12% Infertility rates were highest in Africa and Central/Eastern Europe Additionally, according to a variety of sources, rates of male infertility in North America, Australia, and Central and Eastern Europe varied from 4 5-6%, 9%, and 8-12%, respectively This study demonstrates a novel and unique way to calculate the distribution of male infertility around the world According to our results, at least 30 million men worldwide are infertile with the highest rates in Africa and Eastern Europe Results indicate further research is needed regarding etiology and treatment, reduce stigma & cultural barriers, and establish a more precise calculation

Abstract: Scalable and sustainable solar hydrogen production through photocatalytic water splitting requires highly active and stable earth-abundant co-catalysts to replace expensive and rare platinum. Here we employ density functional theory calculations to direct atomic-level exploration, design and fabrication of a MXene material, Ti3C2 nanoparticles, as a highly efficient co-catalyst. Ti3C2 nanoparticles are rationally integrated with cadmium sulfide via a hydrothermal strategy to induce a super high visible-light photocatalytic hydrogen production activity of 14,342 μmol h−1 g−1 and an apparent quantum efficiency of 40.1% at 420 nm. This high performance arises from the favourable Fermi level position, electrical conductivity and hydrogen evolution capacity of Ti3C2 nanoparticles. Furthermore, Ti3C2 nanoparticles also serve as an efficient co-catalyst on ZnS or ZnxCd1−xS. This work demonstrates the potential of earth-abundant MXene family materials to construct numerous high performance and low-cost photocatalysts/photoelectrodes.

Abstract: The isolation of graphene has triggered an avalanche of studies into the spin-dependent physical properties of this material, as well as graphene-based spintronic devices Here we review the experimental and theoretical state-of-art concerning spin injection and transport, defect-induced magnetic moments, spin-orbit coupling and spin relaxation in graphene Future research in graphene spintronics will need to address the development of applications such as spin transistors and spin logic devices, as well as exotic physical properties including topological states and proximity-induced phenomena in graphene and other 2D materials

Abstract: We study the problem of designing models for machine learning tasks defined on \emph{sets}. In contrast to traditional approach of operating on fixed dimensional vectors, we consider objective functions defined on sets that are invariant to permutations. Such problems are widespread, ranging from estimation of population statistics \cite{poczos13aistats}, to anomaly detection in piezometer data of embankment dams \cite{Jung15Exploration}, to cosmology \cite{Ntampaka16Dynamical,Ravanbakhsh16ICML1}. Our main theorem characterizes the permutation invariant functions and provides a family of functions to which any permutation invariant objective function must belong. This family of functions has a special structure which enables us to design a deep network architecture that can operate on sets and which can be deployed on a variety of scenarios including both unsupervised and supervised learning tasks. We also derive the necessary and sufficient conditions for permutation equivariance in deep models. We demonstrate the applicability of our method on population statistic estimation, point cloud classification, set expansion, and outlier detection.

Abstract: The Correlation Filter is an algorithm that trains a linear template to discriminate between images and their translations. It is well suited to object tracking because its formulation in the Fourier domain provides a fast solution, enabling the detector to be re-trained once per frame. Previous works that use the Correlation Filter, however, have adopted features that were either manually designed or trained for a different task. This work is the first to overcome this limitation by interpreting the Correlation Filter learner, which has a closed-form solution, as a differentiable layer in a deep neural network. This enables learning deep features that are tightly coupled to the Correlation Filter. Experiments illustrate that our method has the important practical benefit of allowing lightweight architectures to achieve state-of-the-art performance at high framerates.

Abstract: Deep Convolutional Neural Network (CNN) is a special type of Neural Networks, which has shown exemplary performance on several competitions related to Computer Vision and Image Processing. Some of the exciting application areas of CNN include Image Classification and Segmentation, Object Detection, Video Processing, Natural Language Processing, and Speech Recognition. The powerful learning ability of deep CNN is primarily due to the use of multiple feature extraction stages that can automatically learn representations from the data. The availability of a large amount of data and improvement in the hardware technology has accelerated the research in CNNs, and recently interesting deep CNN architectures have been reported. Several inspiring ideas to bring advancements in CNNs have been explored, such as the use of different activation and loss functions, parameter optimization, regularization, and architectural innovations. However, the significant improvement in the representational capacity of the deep CNN is achieved through architectural innovations. Notably, the ideas of exploiting spatial and channel information, depth and width of architecture, and multi-path information processing have gained substantial attention. Similarly, the idea of using a block of layers as a structural unit is also gaining popularity. This survey thus focuses on the intrinsic taxonomy present in the recently reported deep CNN architectures and, consequently, classifies the recent innovations in CNN architectures into seven different categories. These seven categories are based on spatial exploitation, depth, multi-path, width, feature-map exploitation, channel boosting, and attention. Additionally, the elementary understanding of CNN components, current challenges, and applications of CNN are also provided.

Abstract: The advent of multidrug resistance among pathogenic bacteria is imperiling the worth of antibiotics, which have previously transformed medical sciences. The crisis of antimicrobial resistance has been ascribed to the misuse of these agents and due to unavailability of newer drugs attributable to exigent regulatory requirements and reduced financial inducements. Comprehensive efforts are needed to minimize the pace of resistance by studying emergent microorganisms, resistance mechanisms, and antimicrobial agents. Multidisciplinary approaches are required across health care settings as well as environment and agriculture sectors. Progressive alternate approaches including probiotics, antibodies, and vaccines have shown promising results in trials that suggest the role of these alternatives as preventive or adjunct therapies in future.

Abstract: Membranes with unprecedented solvent permeance and high retention of dissolved solutes are needed to reduce the energy consumed by separations in organic liquids We used controlled interfacial polymerization to form free-standing polyamide nanofilms less than 10 nanometers in thickness, and incorporated them as separating layers in composite membranes Manipulation of nanofilm morphology by control of interfacial reaction conditions enabled the creation of smooth or crumpled textures; the nanofilms were sufficiently rigid that the crumpled textures could withstand pressurized filtration, resulting in increased permeable area Composite membranes comprising crumpled nanofilms on alumina supports provided high retention of solutes, with acetonitrile permeances up to 112 liters per square meter per hour per bar This is more than two orders of magnitude higher than permeances of commercially available membranes with equivalent solute retention

Abstract: The bulk-boundary correspondence is among the central issues of non-Hermitian topological states. We show that a previously overlooked "non-Hermitian skin effect" necessitates redefinition of topological invariants in a generalized Brillouin zone. The resultant phase diagrams dramatically differ from the usual Bloch theory. Specifically, we obtain the phase diagram of the non-Hermitian Su-Schrieffer-Heeger model, whose topological zero modes are determined by the non-Bloch winding number instead of the Bloch-Hamiltonian-based topological number. Our work settles the issue of the breakdown of conventional bulk-boundary correspondence and introduces the non-Bloch bulk-boundary correspondence.

Abstract: This study uses National Health and Nutrition Examination Survey data to characterize trends in obesity prevalence among US youth and adults between 2007-2008 and 2015-2016.

Abstract: In this essay, I begin by identifying the reasons that automation has not wiped out a majority of jobs over the decades and centuries. Automation does indeed substitute for labor�as it is typically intended to do. However, automation also complements labor, raises output in ways that leads to higher demand for labor, and interacts with adjustments in labor supply. Journalists and even expert commentators tend to overstate the extent of machine substitution for human labor and ignore the strong complementarities between automation and labor that increase productivity, raise earnings, and augment demand for labor. Changes in technology do alter the types of jobs available and what those jobs pay. In the last few decades, one noticeable change has been a "polarization" of the labor market, in which wage gains went disproportionately to those at the top and at the bottom of the income and skill distribution, not to those in the middle; however, I also argue, this polarization is unlikely to continue very far into future. The final section of this paper reflects on how recent and future advances in artificial intelligence and robotics should shape our thinking about the likely trajectory of occupational change and employment growth. I argue that the interplay between machine and human comparative advantage allows computers to substitute for workers in performing routine, codifiable tasks while amplifying the comparative advantage of workers in supplying problem-solving skills, adaptability, and creativity.

Abstract: In the course of Parkinson's disease (PD), the enteric nervous system (ENS) and parasympathetic nerves are amongst the structures earliest and most frequently affected by alpha-synuclein pathology. Accordingly, gastrointestinal dysfunction, in particular constipation, is an important non-motor symptom in PD and often precedes the onset of motor symptoms by years. Recent research has shown that intestinal microbiota interact with the autonomic and central nervous system via diverse pathways including the ENS and vagal nerve. The gut microbiome in PD has not been previously investigated. We compared the fecal microbiomes of 72 PD patients and 72 control subjects by pyrosequencing the V1-V3 regions of the bacterial 16S ribosomal RNA gene. Associations between clinical parameters and microbiota were analyzed using generalized linear models, taking into account potential confounders. On average, the abundance of Prevotellaceae in feces of PD patients was reduced by 77.6% as compared with controls. Relative abundance of Prevotellaceae of 6.5% or less had 86.1% sensitivity and 38.9% specificity for PD. A logistic regression classifier based on the abundance of four bacterial families and the severity of constipation identified PD patients with 66.7% sensitivity and 90.3% specificity. The relative abundance of Enterobacteriaceae was positively associated with the severity of postural instability and gait difficulty. These findings suggest that the intestinal microbiome is altered in PD and is related to motor phenotype. Further studies are warranted to elucidate the temporal and causal relationships between gut microbiota and PD and the suitability of the microbiome as a biomarker.

Abstract: In real-world face detection, large visual variations, such as those due to pose, expression, and lighting, demand an advanced discriminative model to accurately differentiate faces from the backgrounds. Consequently, effective models for the problem tend to be computationally prohibitive. To address these two conflicting challenges, we propose a cascade architecture built on convolutional neural networks (CNNs) with very powerful discriminative capability, while maintaining high performance. The proposed CNN cascade operates at multiple resolutions, quickly rejects the background regions in the fast low resolution stages, and carefully evaluates a small number of challenging candidates in the last high resolution stage. To improve localization effectiveness, and reduce the number of candidates at later stages, we introduce a CNN-based calibration stage after each of the detection stages in the cascade. The output of each calibration stage is used to adjust the detection window position for input to the subsequent stage. The proposed method runs at 14 FPS on a single CPU core for VGA-resolution images and 100 FPS using a GPU, and achieves state-of-the-art detection performance on two public face detection benchmarks.

Abstract: Cancer immunotherapy has transformed the treatment of cancer. However, increasing use of immune-based therapies, including the widely used class of agents known as immune checkpoint inhibitors, has exposed a discrete group of immune-related adverse events (irAEs). Many of these are driven by the same immunologic mechanisms responsible for the drugs' therapeutic effects, namely blockade of inhibitory mechanisms that suppress the immune system and protect body tissues from an unconstrained acute or chronic immune response. Skin, gut, endocrine, lung and musculoskeletal irAEs are relatively common, whereas cardiovascular, hematologic, renal, neurologic and ophthalmologic irAEs occur much less frequently. The majority of irAEs are mild to moderate in severity; however, serious and occasionally life-threatening irAEs are reported in the literature, and treatment-related deaths occur in up to 2% of patients, varying by ICI. Immunotherapy-related irAEs typically have a delayed onset and prolonged duration compared to adverse events from chemotherapy, and effective management depends on early recognition and prompt intervention with immune suppression and/or immunomodulatory strategies. There is an urgent need for multidisciplinary guidance reflecting broad-based perspectives on how to recognize, report and manage organ-specific toxicities until evidence-based data are available to inform clinical decision-making. The Society for Immunotherapy of Cancer (SITC) established a multidisciplinary Toxicity Management Working Group, which met for a full-day workshop to develop recommendations to standardize management of irAEs. Here we present their consensus recommendations on managing toxicities associated with immune checkpoint inhibitor therapy.

Abstract: Discriminative Correlation Filters (DCF) have demonstrated excellent performance for visual object tracking. The key to their success is the ability to efficiently exploit available negative data by including all shifted versions of a training sample. However, the underlying DCF formulation is restricted to single-resolution feature maps, significantly limiting its potential. In this paper, we go beyond the conventional DCF framework and introduce a novel formulation for training continuous convolution filters. We employ an implicit interpolation model to pose the learning problem in the continuous spatial domain. Our proposed formulation enables efficient integration of multi-resolution deep feature maps, leading to superior results on three object tracking benchmarks: OTB-2015 (+5.1% in mean OP), Temple-Color (+4.6% in mean OP), and VOT2015 (20% relative reduction in failure rate). Additionally, our approach is capable of sub-pixel localization, crucial for the task of accurate feature point tracking. We also demonstrate the effectiveness of our learning formulation in extensive feature point tracking experiments. Code and supplementary material are available at this http URL.

Abstract: Background Among patients with stable coronary disease and moderate or severe ischemia, whether clinical outcomes are better in those who receive an invasive intervention plus medical ther...

Abstract: for which we are very grateful. All the authors and investigators of dementia studies who provided us more specific data from their work. foreword Today, over 46 million people live with dementia worldwide, more than the population of Spain. This number is estimated to increase to 131.5 million by 2050. Dementia also has a huge economic impact. Today, the total estimated worldwide cost of dementia is US $818 billion, and it will become a trillion dollar disease by 2018. This means that if dementia care were a country, it would be the world's 18th largest economy, more than the market values of companies such as Apple (US$ 742 billion), Google (US$ 368 billion) and Exxon (US$ 357 billion). In many parts of the world, there is a growing awareness of dementia, but across the globe it remains the case that a diagnosis of dementia can bring with it stigma and social isolation. Today, we estimate that 94% of people living with dementia in low and middle income countries are cared for at home. These are regions where health and care systems often provide limited or no support to people living with dementia or to their families. The 2015 World Alzheimer Report updates data on the prevalence, incidence, cost and trends of dementia worldwide. It also estimates how these numbers will increase in the future, leaving us with no doubt that dementia, including Alzheimer's disease and other causes, is one of the biggest global public health and social care challenges facing people today and in the future. The two organisations we lead are ADI, the only worldwide federation of Alzheimer associations and global voice on dementia, and Bupa, a purpose-driven global health and care company that is the leading international provider of specialist dementia care, caring for around 60,000 people living with dementia each year. Together, we are committed to ensuring that dementia becomes an international health priority. We believe national dementia plans are the first step towards ensuring all countries are equipped to enable people to live well with dementia, and help to reduce the risk of dementia for future generations. There is now a growing list of countries which have such provision in place or which are developing national dementia plans, but it's not enough. Given the epidemic scale of dementia, with no known cure on the horizon, and with a global ageing population, we're calling on governments and …

Abstract: MicroRNAs (miRNAs) are small noncoding RNAs that act as master regulators in many biological processes. miRNAs function mainly by downregulating the expression of their gene targets. Thus, accurate prediction of miRNA targets is critical for characterization of miRNA functions. To this end, we have developed an online database, miRDB, for miRNA target prediction and functional annotations. Recently, we have performed major updates for miRDB. Specifically, by employing an improved algorithm for miRNA target prediction, we now present updated transcriptome-wide target prediction data in miRDB, including 3.5 million predicted targets regulated by 7000 miRNAs in five species. Further, we have implemented the new prediction algorithm into a web server, allowing custom target prediction with user-provided sequences. Another new database feature is the prediction of cell-specific miRNA targets. miRDB now hosts the expression profiles of over 1000 cell lines and presents target prediction data that are tailored for specific cell models. At last, a new web query interface has been added to miRDB for prediction of miRNA functions by integrative analysis of target prediction and Gene Ontology data. All data in miRDB are freely accessible at http://mirdb.org.

Abstract: Model selection using likelihood-based criteria (e.g., AIC) is one of the first steps in phylogenetic analysis. One must select both a substitution matrix and a model for rates across sites. A simple method is to test all combinations and select the best one. We describe heuristics to avoid these extensive calculations. Runtime is divided by $2 with results remaining nearly the same, and the method performs well compared with ProtTest and jModelTest2. Our software, "Smart Model Selection" (SMS), is implemented in the PhyML environment and available using two interfaces: command-line (to be integrated in pipelines) and a web server (http://www.atgc-montpellier.fr/phyml-sms/).

Abstract: Importance The retinal vasculature is involved in many ocular diseases that cause visual loss. Although fluorescein angiography is the criterion standard for evaluating the retina vasculature, it has risks of adverse effects and known defects in imaging all the layers of the retinal vasculature. Optical coherence tomography (OCT) angiography can image vessels based on flow characteristics and may provide improved information. Objective To investigate the ability of OCT angiography to image the vascular layers within the retina compared with conventional fluorescein angiography. Design, Setting, and Participants In this study, performed from March 14, 2014, through June 24, 2014, a total of 5 consecutive, overlapping B-scan OCT angiography images composed of 216 A-scans were obtained at 216 discrete positions within a region of interest, typically a 2 × 2-mm area of the retina. The flow imaging was based on split-spectrum amplitude decorrelation angiography (SSADA), which can dissect layers of vessels in the retina. These distinct layers were compared with the fluorescein angiograms in 12 healthy eyes from patients at a private practice retina clinic to evaluate the ability to visualize the radial peripapillary capillary network. The proportion of the inner vs outer retinal vascular layers was estimated by 3 masked readers and compared with conventional fluorescein angiograms of the same eyes. Main Outcomes and Measures Outcome measures were visualization of the radial peripapillary capillary network in the fluorescein and SSADA scans and the proportion of the inner retinal vascular plexus vs the outer retinal capillary plexus as seen in SSADA scans that would match the fluorescein angiogram. Results In none of the 12 eyes could the radial peripapillary capillary network be visualized completely around the nerve head by fluorescein angiography, whereas the network was readily visualized in the SSADA scans. The fluorescein angiograms were matched, with a mean proportion of the inner vascular plexus being 95.3% (95% CI, 92.2%-97.8%) vs 4.7% (95% CI, 2.6%-5.7%) for the outer capillary plexus from the SSADA scans. Conclusions and Relevance Fluorescein angiography does not image the radial peripapillary or the deep capillary networks well. However, OCT angiography can image all layers of the retinal vasculature without dye injection. Therefore, OCT angiography, and the findings generated, have the potential to affect clinical evaluation of the retina in healthy patients and patients with disease.

Abstract: Semi-supervised learning lately has shown much promise in improving deep learning models when labeled data is scarce. Common among recent approaches is the use of consistency training on a large amount of unlabeled data to constrain model predictions to be invariant to input noise. In this work, we present a new perspective on how to effectively noise unlabeled examples and argue that the quality of noising, specifically those produced by advanced data augmentation methods, plays a crucial role in semi-supervised learning. By substituting simple noising operations with advanced data augmentation methods such as RandAugment and back-translation, our method brings substantial improvements across six language and three vision tasks under the same consistency training framework. On the IMDb text classification dataset, with only 20 labeled examples, our method achieves an error rate of 4.20, outperforming the state-of-the-art model trained on 25,000 labeled examples. On a standard semi-supervised learning benchmark, CIFAR-10, our method outperforms all previous approaches and achieves an error rate of 5.43 with only 250 examples. Our method also combines well with transfer learning, e.g., when finetuning from BERT, and yields improvements in high-data regime, such as ImageNet, whether when there is only 10% labeled data or when a full labeled set with 1.3M extra unlabeled examples is used. Code is available at this https URL.

Abstract: We introduce Group equivariant Convolutional Neural Networks (G-CNNs), a natural generalization of convolutional neural networks that reduces sample complexity by exploiting symmetries. G-CNNs use G-convolutions, a new type of layer that enjoys a substantially higher degree of weight sharing than regular convolution layers. G-convolutions increase the expressive capacity of the network without increasing the number of parameters. Group convolution layers are easy to use and can be implemented with negligible computational overhead for discrete groups generated by translations, reflections and rotations. G-CNNs achieve state of the art results on CI- FAR10 and rotated MNIST.

Abstract: Significance: Chronic wounds include, but are not limited, to diabetic foot ulcers, venous leg ulcers, and pressure ulcers. They are a challenge to wound care professionals and consume a great deal of healthcare resources around the globe. This review discusses the pathophysiology of complex chronic wounds and the means and modalities currently available to achieve healing in such patients. Recent Advances: Although often difficult to treat, an understanding of the underlying pathophysiology and specific attention toward managing these perturbations can often lead to successful healing. Critical Issues: Overcoming the factors that contribute to delayed healing are key components of a comprehensive approach to wound care and present the primary challenges to the treatment of chronic wounds. When wounds fail to achieve sufficient healing after 4 weeks of standard care, reassessment of underlying pathology and consideration of the need for advanced therapeutic agents should be undertaken. However, selection ...

Abstract: Systematic reviews of patient-reported outcome measures (PROMs) differ from reviews of interventions and diagnostic test accuracy studies and are complex. In fact, conducting a review of one or more PROMs comprises of multiple reviews (i.e., one review for each measurement property of each PROM). In the absence of guidance specifically designed for reviews on measurement properties, our aim was to develop a guideline for conducting systematic reviews of PROMs. Based on literature reviews and expert opinions, and in concordance with existing guidelines, the COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN) steering committee developed a guideline for systematic reviews of PROMs. A consecutive ten-step procedure for conducting a systematic review of PROMs is proposed. Steps 1–4 concern preparing and performing the literature search, and selecting relevant studies. Steps 5–8 concern the evaluation of the quality of the eligible studies, the measurement properties, and the interpretability and feasibility aspects. Steps 9 and 10 concern formulating recommendations and reporting the systematic review. The COSMIN guideline for systematic reviews of PROMs includes methodology to combine the methodological quality of studies on measurement properties with the quality of the PROM itself (i.e., its measurement properties). This enables reviewers to draw transparent conclusions and making evidence-based recommendations on the quality of PROMs, and supports the evidence-based selection of PROMs for use in research and in clinical practice.

Abstract: Unlike images which are represented in regular dense grids, 3D point clouds are irregular and unordered, hence applying convolution on them can be difficult. In this paper, we extend the dynamic filter to a new convolution operation, named PointConv. PointConv can be applied on point clouds to build deep convolutional networks. We treat convolution kernels as nonlinear functions of the local coordinates of 3D points comprised of weight and density functions. With respect to a given point, the weight functions are learned with multi-layer perceptron networks and the density functions through kernel density estimation. A novel reformulation is proposed for efficiently computing the weight functions, which allowed us to dramatically scale up the network and significantly improve its performance. The learned convolution kernel can be used to compute translation-invariant and permutation-invariant convolution on any point set in the 3D space. Besides, PointConv can also be used as deconvolution operators to propagate features from a subsampled point cloud back to its original resolution. Experiments on ModelNet40, ShapeNet, and ScanNet show that deep convolutional neural networks built on PointConv are able to achieve state-of-the-art on challenging semantic segmentation benchmarks on 3D point clouds. Besides, our experiments converting CIFAR-10 into a point cloud showed that networks built on PointConv can match the performance of convolutional networks in 2D images of a similar structure.