Standard Test Set Research Articles

A joint FrameNet and element focusing Sentence-BERT method of sentence similarity computation

As one of the fundamental research areas of natural language processing, sentence similarity computation attracts researchers’ attention. Considering two single independent sentences, it is difficult to measure the similarity between them without sufficient context information. To solve this issue, we propose a joint FrameNet and element focusing Sentence-BERT method of sentence similarity computation (FEFS3C). Considering the actual meaning of sentences, we adopt the frame semantics theory and adapt FrameNet in FEFS3C. Moreover, focusing on critical information conveyed in sentences, FEFS3C takes the superiority of deep learning technologies and proposes a new sentence representation model element focusing Sentence-BERT (EF-SBERT) which improves traditional sentence representations. Two primary considerations of sentences in FEFS3C “sentence meaning” and “critical sentence information” aim to better utilize the influence of sentences context. To evaluate the performance of FEFS3C, we carried out experiments on the standard test set “STS-B”. Results show that FEFS3C has obtained better Spearman correlation compared with traditional methods.

Building Standard Levels to Assess the Level of Some Elements of Fitness Related to Health among Junior Footballers

This study aimed to build standard levels to assess the level of some elements of physical fitness related to health among the juniors of Al-Faisali Football Club Academy, as well as to identify the differences in the tests due to the variable of height, weight and age. The researcher used a set of standardized tests that have a high degree of validity and reliability to measure the physical fitness elements of the subject of the study. The sample included (123) young players registered in the academy. The study reached to build standard levels for the juniors of the Al-Faisaly Football Club Academy, as well as the results of the study showed that there are differences between the averages of tests of explosive power, agility, transitional speed and grip strength of the right and left hand due to the variables of age, height and weight, while there were no differences in the 1000m running tests and the 30-second stomach repeat test due to the study variables. The study recommends building standard levels to measure the elements of fitness for other games, specifically for juniors.
 
 Received: 7 December 2021 / Accepted: 26 January 2022 / Published: 5 March 2022

OC09: Patient reported experience with multimodal peripheral electrophysiological studies in children and adults with Spinal Muscular Atrophy (SMA)

Purpose: Several electrophysiological tests have been proposed to aid in characterization of motor unit integrity and function in patients with Spinal Muscular Atrophy (SMA), but systematic studies on tolerability and test burden are lacking. Therefore, we systematically assessed patient reported experience with a standardized set of electrophysiological tests in a prospective SMA cohort. Methods: We recruited consecutive patients seen at the SMA Expertise Centre in Utrecht. SMA types 1c-4, age ≥12 years were eligible for inclusion. All patients underwent a standardized set of electrophysiological tests on the median nerve. We determined tolerability and test burden using visual analogue scale (VAS) scores (range 0 – 10), obtained at the end of each test-set. The protocol was defined as feasible when all examinations could be ended with satisfactory results for both patient and investigator. Results: We included 43 patients with SMA (21 males), mean age 39 years (range 13-67 years). 10/43 patients were ambulant at time of inclusion. 14/43 had (severe) contractures. Median VAS score was 2 with 70% (n=30) scoring ≤3 points. Only two participants (adults) scored 5 and 6. The full electrophysiological test-set was feasible in all SMA patients, who were also willing to undergo the same protocol at follow-up. Conclusion: Our study shows that combined electrophysiological studies on motor unit physiology are very well tolerated in children aged ≥12 years and adults with SMA, and even feasible with severe contractures.

Investigating the relation between gravitational wave tests of general relativity

Gravitational wave observations of compact binary coalescences provide precision probes of strong-field gravity. There is thus now a standard set of null tests of general relativity (GR) applied to LIGO-Virgo detections and many more such tests proposed. However, the relation between all these tests is not yet well understood. We start to investigate this by applying a set of standard tests to simulated observations of binary black holes in GR and with phenomenological deviations from GR. We consider four types of tests: residuals, inspiral-merger-ringdown consistency, parameterized phasing (two varieties), and modified dispersion relation. We also check the consistency of the unmodeled reconstruction of the waveforms with the waveform recovered using GR templates. These tests are applied to simulated observations similar to GW150914 with both large and small deviations from GR and similar to GW170608 just with small deviations from GR. We find that while very large deviations from GR are picked up with high significance by almost all tests, more moderate deviations are picked up by only a few tests, and some deviations are not recognized as GR violations by any test at the moderate signal-to-noise ratios we consider. Moreover, the tests that identify various deviations with high significance are not necessarily the expected ones. In particular, the parameterized tests recover PN test parameters much closer to zero than their true values in some cases. Additionally, we find that of the GR deviations we consider, the residuals test is only able to detect extreme deviations from GR that no longer look like binary black hole coalescences in GR. The reconstruction comparison shows more promise for detecting relatively small GR deviations in an unmodeled framework, at least for high-mass systems.

Towards improving the robustness of sequential labeling models against typographical adversarial examples using triplet loss

AbstractMany fundamental tasks in natural language processing (NLP) such as part-of-speech tagging, text chunking, and named-entity recognition can be formulated as sequence labeling problems. Although neural sequence labeling models have shown excellent results on standard test sets, they are very brittle when presented with misspelled texts. In this paper, we introduce an adversarial training framework that enhances the robustness against typographical adversarial examples. We evaluate the robustness of sequence labeling models with an adversarial evaluation scheme that includes typographical adversarial examples. We generate two types of adversarial examples without access (black-box) or with full access (white-box) to the target model’s parameters. We conducted a series of extensive experiments on three languages (English, Thai, and German) across three sequence labeling tasks. Experiments show that the proposed adversarial training framework provides better resistance against adversarial examples on all tasks. We found that we can further improve the model’s robustness on the chunking task by including a triplet loss constraint.

Secure machine learning against adversarial samples at test time

Deep neural networks (DNNs) are widely used to handle many difficult tasks, such as image classification and malware detection, and achieve outstanding performance. However, recent studies on adversarial examples, which have maliciously undetectable perturbations added to their original samples that are indistinguishable by human eyes but mislead the machine learning approaches, show that machine learning models are vulnerable to security attacks. Though various adversarial retraining techniques have been developed in the past few years, none of them is scalable. In this paper, we propose a new iterative adversarial retraining approach to robustify the model and to reduce the effectiveness of adversarial inputs on DNN models. The proposed method retrains the model with both Gaussian noise augmentation and adversarial generation techniques for better generalization. Furthermore, the ensemble model is utilized during the testing phase in order to increase the robust test accuracy. The results from our extensive experiments demonstrate that the proposed approach increases the robustness of the DNN model against various adversarial attacks, specifically, fast gradient sign attack, Carlini and Wagner (C&W) attack, Projected Gradient Descent (PGD) attack, and DeepFool attack. To be precise, the robust classifier obtained by our proposed approach can maintain a performance accuracy of 99% on average on the standard test set. Moreover, we empirically evaluate the runtime of two of the most effective adversarial attacks, i.e., C&W attack and BIM attack, to find that the C&W attack can utilize GPU for faster adversarial example generation than the BIM attack can. For this reason, we further develop a parallel implementation of the proposed approach. This parallel implementation makes the proposed approach scalable for large datasets and complex models.

Blood pressure variability and neuroplasticity in patients with type 2 diabetes mellitus

Analysis of the role of blood pressure (BP) variability in the formation of neuroplasticity in patients with type 2 diabetes mellitus (DM). 100 patients with type 2 DM were examined, which were divided into groups depending on the presence of cognitive impairment (CI), the control group consisted of 25 people. All examined patients underwent a clinical examination, a standard set of biochemical blood tests, plasma osteopontin levels, 24-hour blood pressure monitoring (ABPM) for 24-26 h MRI of the brain (dynamic contrast and arterial spin marks, proton spectroscopy, tractography). Patients with type 2 diabetes and CI had higher body mass index, blood levels of glycated hemoglobin, glucose, alanine aminotransferase, low-density lipoprotein, triglycerides, total cholesterol, osteopontin, and lower levels of high-density lipoprotein (p≤0.05). The level of osteopontin was higher in patients with overweight, hyperglycemia, dyslipidemia, and in patients with CI in patients with BP variability. When assessing 24-hour blood pressure monitoring (ABPM), a significant difference was found in all standard indicators, while patients with type 2 diabetes were referred to as «non-dipper», in the presence of CI they noted significantly higher values of the index of time and area of stay in the suprathreshold state. BP and variability in SBP and DBP at night, as well as the risk of occult hypertension. A decrease in cerebral blood flow was revealed according to the data of contrast and non-contrast assessment of perfusion in cortical (especially in the frontal lobe) and subcortical (mainly in the putamen) structures, associated with changes in ABPM parameters. Mean SBP and DBP day and night, as well as the index of BP variability, also affect the integrity of the corticospinal, uncinate, lower longitudinal tracts, and arcuate fasciculus. The same parameters change the metabolism of the hippocampus in terms of choline (Cho), creatine (Cr), creatine phosphate (Cr2), as well as the ratio of N-acetylaspartate (NAA)/Cho, NAA/Cr, Cho/Cr. In patients with type 2 diabetes, BP variability contributes to the formation of CI through a pro-inflammatory mechanism (osteopontin), leading to impaired brain vascularization in general, white matter structure, and hippocampal metabolism.

Construction of the 3D Reconstruction System of Building Construction Scene Based on Deep Learning

The increasing complexity and enormity of construction projects, as well as the fact that the actual operation of construction schedule management still mainly relies on traditional manual management methods, have led to low efficiency of construction schedule management and caused many construction projects to have cost overruns and legal disputes due to schedule delays. Existing 3D reconstruction algorithms often lead to significant voids, distortions, or blurred parts in the reconstructed 3D models, while the machine learning-based 3D reconstruction algorithms are often only to reconstruct simple separated objects and represent them as 3D boxes. A novel architecture of semisupervised 3D reconstruction algorithm is proposed. The algorithm iteratively improves the quality of the original 3D reconstruction model by training a generative adversarial network model to a converged state. Only the prior observed 2D images are required as weakly supervised samples, without any dependence on prior knowledge of the 3D structure shape or reference observations. Experimental results show that this algorithmic framework has significant advantages over the current state-of-the-art 3D reconstruction methods on the standard 3D reconstruction test set.

Treatment of clay soil with paper ash

Abstract The mineralogy of fine soils such as clays has always posed problems and remains an uncontrollable phenomenon in the presence of water and causes destructible damage throughout the world. In order to minimize the cost of implementation, it is necessary to find practical and less expensive solutions to ensure the stabilization of these soils by the valorisation of local waste available in nature. This article concerns an experimental study on the treatment of reconstituted soil by the addition of paper ash of different proportions on a set of standardized tests, the preliminary results show that the paper ash to the clay soil improves its swelling potential, its plasticity, its compaction characteristics, and its shear strength parameters.

High-order numerical solutions to the shallow-water equations on the rotated cubed-sphere grid

A novel numerical approach to solving the shallow-water equations on the sphere using high-order numerical discretizations in both space and time is proposed. A space-time tensor formalism is used to express the equations of motion covariantly and to describe the geometry of the rotated cubed-sphere grid. The spatial discretization is done with the direct flux reconstruction method, which is an alternative formulation to the discontinuous Galerkin approach. The equations of motion are solved in differential form and the resulting discretization is free from quadrature rules. It is well known that the time step of traditional explicit methods is limited by the phase velocity of the fastest waves. Exponential integration is employed to enable integrations with significantly larger time step sizes and improve the efficiency of the overall time integration. New multistep-type exponential propagation iterative methods of orders 4, 5 and 6 are constructed and applied to integrate the shallow-water equations in time. These new schemes enable time integration with high-order accuracy but without significant increases in computational time compared to low-order methods. The exponential matrix functions-vector products used in the exponential schemes are approximated using the complex-step approximation of the Jacobian in the Krylov-based KIOPS (Krylov with incomplete orthogonalization procedure solver) algorithm. Performance of the new numerical methods is evaluated using a set of standard benchmark tests.

A Competitive Nature-Inspired Cuckoo Search Algorithm for Solving the Resource Constrained Project Scheduling Problem

Since its advent, the resource constrained project scheduling problem has been a favorite NP-Hard problem studied by many researchers. In this paper, we examine this problem with a focus on the total project time minimization, usually known as the makespan minimization. To this end, a nature-inspired Cuckoo Search Algorithm (CSA) has been introduced that works by simulating the brooding parasitism of some cuckoo birds in nature. In a nutshell, the proposed cuckoo search algorithm has two main capabilities including: a) a global random search using Levy flights, and b) a local random refinement strategy. In addition, a double justification method has been employed to improve the generated solutions. Simultaneous use of these features can maximize the efficiency of resource searches in uncertain environments. Experimental results on the standard J30, J60, and J120 test sets from the PSPLIB are also presented, showing that the CSA has satisfying results compared with the other well-reported algorithms.

Large Scale Implementation of a Core Set of Outcome Measures is Possible

<h3>Research Objectives</h3> To implement a core set of standardized tests to: 1) measure individual patient outcomes, 2) aggregate data for analysis, and 3) build the foundation for practice-based evidence study. <h3>Design</h3> Stakeholders from physical (PT) and occupational (OT) therapy determined important constructs to measure and identified tests through evidence review. A set of tests was piloted in 6 facilities and modified secondary to feasibility constraints. The final set was: Short Physical Performance Battery (SPPB), 6-minute Walk Test (6MWT), grip strength (Grip), and the Continuity Assessment Record Evaluation (CARE) mobility and self-care item set. The St. Louis Mental Status Examination (SLUMS) was implemented to stratify patients. A data pipeline was established through electronic health record (EHR) customization, extraction and transformation rules, data warehouse creation, and dashboard visualization using PowerBI (TM) software. <h3>Setting</h3> Infinity Rehab is a post-acute therapy provider. This project was initially implemented in 180 skilled nursing facilities (SNF) treating approximately 15000 patients/year. <h3>Participants</h3> 480 PT and 279 OT clinicians completed the initial web-based training over 4-months. A new-hire process was developed to ensure ongoing training. <h3>Interventions</h3> Clinical knowledge tools were developed: guides for set-up, best practices, interpretive comments, data sheets, support for goal writing, fact sheets, frequently asked questions, chart audits, and adherence reporting. Operational and clinical incentives were aligned, and company leaders engaged in an enthusiasm campaign. <h3>Main Outcome Measures</h3> Clinician grouped adherence to each required outcome measure. <h3>Results</h3> Initial adherence was 38.3%, 28.4%, and 51.7% for SPPB, 6MWT, and Grip respectively. (Reporting for SLUMS and CARE was not available). Adherence improved over the following 11 months and stabilized at 70% (SPPB), 60% (6MWT), and 80% (Grip). Adherence targets were established at 70% for SPPB and 6MWT and 80% for Grip. These targets are now a routine operational metric. <h3>Conclusions</h3> It is possible to implement and sustain collection of a core set of standardized tests on a large scale. Currently, data are used for the first 2 stated objectives with plans for the 3rd. <h3>Author(s) Disclosures</h3> The authors have nothing to disclose.

Deep learning based HEVC in-loop filter and noise reduction

The lossy compression techniques at low bit rate often create ringing and contouring effects on the output images and introduce various blurring and distortion at block bounders. To overcome those compression artifacts different neural network based post-processing techniques have been experimented with over the last few years. The traditional loop-filter methods in the HEVC frame-work support two post-processing operations namely a de-blocking filter followed by a sample adaptive offset (SAO) filter. These operations usually introduce extra signaling bits and become overhead to the network with high-resolution video processing. In this study, we came up with a new deep learning-based algorithm for SAO filtering operations and substantiated the merits of the proposed method. We introduced a variable filter size sub-layered dense CNN (SDCNN) to improve the denoising operation and incorporated large stride deconvolution layers for further computation improvement. We demonstrate that our deconvolution model can effectively be trained by leveraging the high-frequency edge features learned in a shallow network using residual learning and data augmentation techniques. Extensive experiments show that our approach outperformed other state-of-the-art approaches in terms of SSIM, Bjøntegaard delta bit-rate (BD-BR), BD-PSNR measurements on the standard video test set and achieves an average of 8.73 % bit rate saving compared to HEVC baseline.

Intelligent Recognition Method of Vehicle Path with Time Window Based on Genetic Algorithm

Based on particle filter and improved cuckoo genetic algorithm, an algorithm for intelligent vehicle path recognition with a time window is designed. Particle filter (PF) is an influential visual tracking tool; it relies on the Monte Carlo Chain framework and Bayesian probability, which are essential for intelligent monitoring systems. The algorithm first uses particle filters for visual tracking and then obtains the current operating environment of the vehicle, then performs cluster analysis on customer locations, and finally performs path recognition in each area. The algorithm not only introduces particle filters, which are advanced visual tracking, but also improves the cuckoo search algorithm; when the bird’s egg is found by the bird’s nest owner, it needs to randomly change the position of the entire bird’s nest, which speeds up the search speed of the optimal delivery route. Analyze and compare the hybrid intelligent algorithm and the cuckoo search algorithm. Finally, the international standard test set Benchmark Problems is used for testing. The experimental outcomes indicated that the new hybrid intelligent approach is an effective algorithm for handling vehicle routing tasks with time windows.

A soft neuroprosthetic hand providing simultaneous myoelectric control and tactile feedback.

Neuroprosthetic hands are typically heavy (over 400 g) and expensive (more than US$10,000), and lack the compliance and tactile feedback of human hands. Here, we report the design, fabrication and performance of a soft, low-cost and lightweight (292 g) neuroprosthetic hand that provides simultaneous myoelectric control and tactile feedback. The neuroprosthesis has six active degrees of freedom under pneumatic actuation, can be controlled through the input from four electromyography sensors that measure surface signals from residual forearm muscles, and integrates five elastomeric capacitive sensors on the fingertips to measure touch pressure so as to enable tactile feedback by eliciting electrical stimulation on the skin of the residual limb. In a set of standardized tests performed by two individuals with transradial amputations, we show that the soft neuroprosthetic hand outperforms a conventional rigid neuroprosthetic hand in speed and dexterity. We also show that one individual with a transradial amputation wearing the soft neuroprosthetic hand can regain primitive touch sensation and real-time closed-loop control.

Single Image Super-resolution Reconstruction Algorithm Based on Deep Learning

Recover the high-resolution image from a single low-resolution image is known as single image super-resolution reconstruction (SISR). SSIR is currently a research hotspot in computer vision. In this paper, we study the feature learning module of the magnification-arbitrary network for Super-Resolution(Meta-SR), and propose an improved method of image feature learning. Its idea is multi-scale residual dense block (MRDB). We integrate it into the existing structure of the Meta-SR network. As a result, the same network has increased the receptive field, and it can obtain more detailed features of the image. This structure can be extended to other existing SISR network models. In the published and standard test set Set14, compared with the interpolation method, the average PSNR value of the algorithm in this paper increases by 3.79dB, 3.01dB and 2.85dB at amplification factors 2, 3 and 4, respectively. Through the analysis of SISR results, it can be obtained that the improved method proposed in this paper has a significant improvement in all indicators, and can more fully extract the image detail features, thereby improving the quality of the reconstructed image.

An AI Approach for Identifying Patients With Cirrhosis.

The goal of this study was to evaluate an artificial intelligence approach, namely deep learning, on clinical text in electronic health records (EHRs) to identify patients with cirrhosis. Accurate identification of cirrhosis in EHR is important for epidemiological, health services, and outcomes research. Currently, such efforts depend on International Classification of Diseases (ICD) codes, with limited success. We trained several machine learning models using discharge summaries from patients with known cirrhosis from a patient registry and random controls without cirrhosis or its complications based on ICD codes. Models were validated on patients for whom discharge summaries were manually reviewed and used as the gold standard test set. We tested Naive Bayes and Random Forest as baseline models and a deep learning model using word embedding and a convolutional neural network (CNN). The training set included 446 cirrhosis patients and 689 controls, while the gold standard test set included 139 cirrhosis patients and 152 controls. Among the machine learning models, the CNN achieved the highest area under the receiver operating characteristic curve (0.993), with a precision of 0.965 and recall of 0.978, compared with 0.879 and 0.981 for the Naive Bayes and Random Forest, respectively (precision 0.787 and 0.958, and recalls 0.878 and 0.827). The precision by ICD codes for cirrhosis was 0.883 and recall was 0.978. A CNN model trained on discharge summaries identified cirrhosis patients with high precision and recall. This approach for phenotyping cirrhosis in the EHR may provide a more accurate assessment of disease burden in a variety of studies.

Crosslinking of the Bioplast plastic material as a means of increasing its biomechanical stability and the effectiveness of keratoplasty for corneal ulcers

Purpose:to assess the effect of ultraviolet (UVA) crosslinking of Bioplast plastic material (silicon-dried human cornea) on its biomechanical stability and effectiveness for keratoplasty in corneal ulcers. Material and methods. To determine the maximum rehydration time, 12 Bioplast samples were incubated in physiological solution (NaCl 0.9 %) for 12 hours and monitored for thickness using optical coherence tomography (OCT Spectralis, Heidelberg Engineering). After the epithelium was scarified rehydrated corneas were treated withof 0.1% riboflavin solution for 30 minutes, and after it 8 rehydrated samples (experimental group) were treated with UVA for 30minutes (wavelength 370 nm, power 3 mW/cm2) using a UV-X device system (IROC AG, Switzerland). In the control group (4 rehydrated samples), no UVA crosslinking was performed. The elastic-strength indicators of the experimental and control group samples were assessed by applying uniaxial tension using a deformation machine (Autograph AGS-H, Japan). To examine the patients (presented in clinical examples)before and after keratoplasty with Bioplast subjected to UVA crosslinking, a set of standard ophthalmic tests was used. Results.Rehydration of Bioplast was the most intensive in the first three hours of incubation, during which time tissue thickness increased uniformly by 50–55 μm/hour to 700–710 μm, while further incubation showed no significant additional effect on the fluid content in the cornea. After UVA crosslinking, the strength indices and the elastic modulus of the rehydrated Bioplast samples significantly increased in comparison with the control group. Clinical examples demonstrate the effectiveness of modified Bioplast for organ-preserving penetrating keratoplasty in case of perforatedcorneal ulcer. Conclusion. UVA crosslinking of rehydrated Bioplast increases its biomechanical stability. The data of laboratory studies and the first clinical tests demonstrate that modified Bioplast may be an effective replacement of native donor cornea in urgent keratoplasty of perforated ulcers.

Microbiome Preprocessing Machine Learning Pipeline.

Background16S sequencing results are often used for Machine Learning (ML) tasks. 16S gene sequences are represented as feature counts, which are associated with taxonomic representation. Raw feature counts may not be the optimal representation for ML.MethodsWe checked multiple preprocessing steps and tested the optimal combination for 16S sequencing-based classification tasks. We computed the contribution of each step to the accuracy as measured by the Area Under Curve (AUC) of the classification.ResultsWe show that the log of the feature counts is much more informative than the relative counts. We further show that merging features associated with the same taxonomy at a given level, through a dimension reduction step for each group of bacteria improves the AUC. Finally, we show that z-scoring has a very limited effect on the results.ConclusionsThe prepossessing of microbiome 16S data is crucial for optimal microbiome based Machine Learning. These preprocessing steps are integrated into the MIPMLP - Microbiome Preprocessing Machine Learning Pipeline, which is available as a stand-alone version at: https://github.com/louzounlab/microbiome/tree/master/Preprocess or as a service at http://mip-mlp.math.biu.ac.il/Home Both contain the code, and standard test sets.

Smoothed particle radiation hydrodynamics: two-moment method with local Eddington tensor closure

ABSTRACTWe present a new smoothed particle hydrodynamics-radiative transfer method (sph-m1rt) that is coupled dynamically with sph. We implement it in the (task-based parallel) swift galaxy simulation code but it can be straightforwardly implemented in other sph codes. Our moment-based method simultaneously solves the radiation energy and flux equations in sph, making it adaptive in space and time. We modify the m1 closure relation to stabilize radiation fronts in the optically thin limit. We also introduce anisotropic artificial viscosity and high-order artificial diffusion schemes, which allow the code to handle radiation transport accurately in both the optically thin and optically thick regimes. Non-equilibrium thermochemistry is solved using a semi-implicit sub-cycling technique. The computational cost of our method is independent of the number of sources and can be lowered further by using the reduced speed-of-light approximation. We demonstrate the robustness of our method by applying it to a set of standard tests from the cosmological radiative transfer comparison project of Iliev et al. The sph-m1rt scheme is well-suited for modelling situations in which numerous sources emit ionizing radiation, such as cosmological simulations of galaxy formation or simulations of the interstellar medium.