Standard Output Research Articles

A Computer Vision System for an Automated Scoring of a Hand-drawn Geometric Figure

Visual Reproduction is a condition to measure Visual Spatial Memory as one of the cognitive domains commonly used to measure visuo-spatial memory. Geometric figures serve as stimulus material, and probands have to reproduce the figures from memory through a hand drawing. The scoring of the drawing has subjective elements. This study aims to evaluate the scoring criteria for the Figural Reproduction Test (FRT), part of the Indonesian Neuropsychological Test Battery, and to develop and evaluate an automated scoring system based on computer vision technology (FRT-CVAS). Scoring evaluation conducted by Cohen Kappa analysis, accuracy, sensitivity, and specificity. The analyzes of the three criteria of the manual confirmed a subjective element in the scoring of the shape of triangles by a moderate (0.74) inter-rater agreement; this agreement could be improved to 0.84 by a slight modification of its criteria. FRT-CVAS, based on computer vision’s identification of the different elements of the hand drawing, was developed and trained using 290 drawings. The system was additionally tested by comparing its scoring with the scoring of two independent raters on 120 drawings from a second data set. FRT-CVAS recognized all elements, and its comparison between human raters showed a high accuracy and sensitivity (minimally 0.91), while the specificity was 0.80 for one of the three criteria. FRT-CVAS offers a highly standardized, consistent, precise, and objective output from the first card in the FRT. This approach is advantageous to data-hungry alternatives such as deep learning when applied to the automated scoring of hand drawings with relatively little data available for training.

Reducing tree volume overestimation in quantitative structure models using modeled branch topology and direct twig measurements

Abstract Quantitative Structure Models (QSMs) are fit to tree point clouds to represent the topology of trees as a network of cylinders. QSMs allow for the calculation of metrics difficult to measure without destructive sampling, including total tree volume. Current limitations in terrestrial laser scanning technology make small branches difficult to accurately resolve, causing overestimation of small branch volume in QSMs, which can translate into overestimating tree biomass. We present a new method called Real Twig to correct overestimated small branch and twig cylinders in QSMs. Real Twig differs from current methods by using twig diameters measured directly from corresponding tree species to model a unique taper for every path in the QSM, using the QSM’s inherent branching topology, but without relying on predefined mathematical or allometric relationships. To test Real Twig, we generated QSMs for different sets of trees that had detailed dry mass and density measurements obtained via felling after scanning. QSM-based biomass estimates were obtained by multiplying the tree’s QSM-based volume estimate by the tree’s specific basic density value. We trained our method with high-quality data consisting of five northern red oak (Quercus rubra L.) and five red maple (Acer rubrum L.) trees, using two different versions of TreeQSM, a widely used algorithm for generating QSMs. We further tested our method on three publicly available datasets, including managed forests and large tropical trees, collected with both phase-shift or time-of-flight sensors. QSMs corrected with our Real Twig method showed a very large improvement in tree biomass estimation, with a relative mean error of −1.2%, a relative root mean square error of 10.5%, and a concordance correlation coefficient of 0.999, compared to a relative mean error 76.8%, a relative root mean square error of 48.7%, and a concordance correlation coefficient of 0.982, when using the standard outputs of TreeQSM.

Using health literacy principles to improve understanding of evolving evidence in health emergencies: Optimisation and evaluation of a COVID-19 vaccination risk-benefit calculator

BackgroundRisk communication tools based on epidemiological models can help inform decision-making, but must be responsive to health literacy needs to be effective. To facilitate informed choice about risks and benefits of COVID-19 vaccination, an epidemiological model called the COVID-19 Risk Calculator (CoRiCal) tool was developed by a multi-disciplinary team. AimThis paper demonstrates how to use health literacy principles to improve consumer understanding of COVID-19 and vaccine effects, using a range of methods that could be applied to any health emergency. MethodsStage 1: Health literacy optimisation and user testing to reduce improve understandability (n = 19). Stage 2: Experiments to explore the effect of risk communication formats on perceived understanding including probability, graphs, evaluative labels and comparison risks (n = 207). Stage 3: Randomised controlled trial (n = 2005) with 4 arms: 1) standard government information; 2) standard CoRiCal output based on bar graphs; 3) animation explaining bar graphs in “x per million” format; 4) animation explaining bar graphs in “1 in x chance” format. The primary outcome was knowledge about COVID-19 risk. ResultsStage 1 reduced the complexity of the text and graphs. Stage 2 showed that different risk communication formats change perceived understanding, with a preference for evaluative labels across 2 experiments and some indication people with lower health literacy had a greater preference for bar graphs. Stage 3 showed both animations increased knowledge compared to standard government information. There was no difference between the probability formats, or by health literacy level. DiscussionThe results showed that simple explanations of complex epidemiological models improve knowledge about COVID-19 and vaccination. This demonstrates how health literacy design principles and short animations can be used to support informed decision making about health emergencies.

Slot-die coating method in organic solar panel and evaluation in polluted air

Renewable energy sources and pollution have a direct effect on each other. Solar panels development could enhance the adoption of green energy sources and contribute to a cleaner environment. Today's third generation organic photovoltaic panels (OPVs) offer good efficiency, low fabrication costs, and can be produces rapidly using slot-die coating methods. These advantages make OPVs increasingly attractive for broadening solar energy consumption. Therefore, related studies are being conducted to achieve this goal. This study aimed to evaluate real-world performance of OPVs in polluted air by testing a fully slot-die-coating OPV module in Tehran’s polluted environment, specifically at four air quality control stations. The results showed a power conversion efficiency (PCE) of 6.3% under standard test condition (STC) and maximum power output of 0.189 W in 5×6 cm module, which include ITO/PET, PEI, P3HT-PCBM, PEDOT-PSS, and AGNW layers. Testing in polluted environment revealed a significant dicrease in current and maximum power (Pmax) due to reduced sunlight reaching the module’s surface. Pmax dropped to 0.0229 W at least polluted site in Masodieh Street and to 0.0173 W at the most polluted site in Shahr-e Rey Street.

Capacitor voltage balancing, capacitance monitoring, and fast fault detection in a nested neutral point clamped (NNPC) converter with the reduced number of sensors

In multilevel converters (MCs), system components such as capacitors and semiconductor switches are very susceptible to damage. Furthermore, utilizing voltage sensors to balance capacitor voltages increases the system's cost and complexity while decreasing reliability. This paper presents methods for voltage balancing of capacitors, capacitance monitoring and open-circuit fault detection in nested neutral point-clamped (NNPC) converter with a reduced number of voltage and current sensors. In the proposed method, converter capacitors' voltage and current sensors are eliminated, and only the output sensors are employed to control the converter. In order to balance the voltage of capacitors, their voltages are estimated in three cases without using individual sensors. Comparing the capacitors' measured and estimated voltage change ratios is used to assess their health in the monitoring method. Thus, the capacitors' status can be checked by considering the system's health indicator during each period. Furthermore, the open-circuit fault detection method can identify defective switches within one processing cycle by comparing the converter's measured and standard output voltages in all switching states. The proposed methods have been evaluated in an experimental study of the system, and the results confirmed the effectiveness of the methods based on the new system configuration.© 2017 Elsevier Inc. All rights reserved.

A dendrite is a dendrite is a dendrite? Dendritic signal integration beyond the "antenna" model.

Neurons in central nervous systems receive multiple synaptic inputs and transform them into a largely standardized output to their target cells-the action potential. A simplified model posits that synaptic signals are integrated by linear summation and passive propagation towards the axon initial segment, where the threshold for spike generation is either crossed or not. However, multiple lines of research during past decades have shown that signal integration in individual neurons is much more complex, with important functional consequences at the cellular, network, and behavioral-cognitive level. The interplay between concomitant excitatory and inhibitory postsynaptic potentials depends strongly on the relative timing and localization of the respective synapses. In addition, dendrites contain multiple voltage-dependent conductances, which allow scaling of postsynaptic potentials, non-linear input processing, and compartmentalization of signals. Together, these features enable a rich variety of single-neuron computations, including non-linear operations and synaptic plasticity. Hence, we have to revise over-simplified messages from textbooks and use simplified computational models like integrate-and-fire neurons with some caution. This concept article summarizes the most important mechanisms of dendritic integration and highlights some recent developments in the field.

Creating an automated tool for a consistent and repeatable evaluation of disability progression in clinical studies for multiple sclerosis

Background: The lack of standardized disability progression evaluation in multiple sclerosis (MS) hinders reproducibility of clinical study results, due to heterogeneous and poorly reported criteria. Objective: To demonstrate the impact of using different parameters when evaluating MS progression, and to introduce an automated tool for reproducible outcome computation. Methods: Re-analyzing BRAVO clinical trial data (NCT00605215), we examined the fluctuations in computed treatment effect on confirmed disability progression (CDP) and progression independent of relapse activity (PIRA) when varying different parameters. These analyses were conducted using the msprog package for R, which we developed as a tool for CDP assessment from longitudinal data, given a set of criteria that can be specified by the user. Results: The BRAVO study reported a hazard ratio (HR) of 0.69 (95% confidence interval (CI): 0.46–1.02) for CDP. Using the different parameter configurations, the resulting treatment effect on CDP varied considerably, with HRs ranging from 0.59 (95% CI: 0.41–0.86) to 0.72 (95% CI: 0.48–1.07). The treatment effect on PIRA varied from an HR = 0.62 (95% CI: 0.41–0.93) to an HR = 0.65 (95% CI: 0.40–1.04). Conclusions: The adoption of an open-access tool validated by the research community, with clear parameter specification and standardized output, could greatly reduce heterogeneity in CDP estimation and promote repeatability of study results.

On more insightful dimensionless numbers for computational viscoelastic rheology

Abrupt contraction flows involving viscoelastic fluids represent a longstanding computational challenge within the field of non-Newtonian fluid mechanics. Despite the apparent simplicity of the geometry, these flows have given rise to intricate discussions in the study of viscoelastic phenomena. This study aims to re-examine the numerical solutions for flows through abrupt contractions, offering a fresh interpretation through the lens of reformulated dimensionless numbers. These numbers are designed to consider the characteristic shear rate of the problem, providing a more comprehensive understanding of the underlying dynamics.When investigating models with intermediate levels of complexity, such as the Giesekus and Phan-Thien-Tanner constitutive equations, the usual comparison with the corresponding Oldroyd-B model becomes inadequate because it tends to rely on the nominal relaxation time (λ) and the nominal total viscosity (η) instead of their effective counterparts when defining the Reynolds number (Re), the Weissenberg number (Wi) and the ratio of solvent to total viscosities (β) (β plays a role only in rheological models involving a solvent contribution). If these dimensionless numbers are tailored to account for the characteristic shear rate specific to the problem under investigation, the choice of the corresponding Oldroyd-B flow, at the adequate values of Re, Wi, and β allows for significantly better quantification of the correct effects of nonlinear viscoelasticity of the original model.We show the conventional approach tends to overemphasize the role of the nonlinear parameter in nonlinear constitutive equations, like the Giesekus and PTT models, when examining standard abrupt contraction flow outputs such as the Couette correction and vortex size. This overestimation occurs because the conventional method does not allow the Reynolds and Weissenberg numbers (and possibly β) to carry the portion of the nonlinear effect that can potentially be captured by the linear Oldroyd-B model through the use of characteristic shear rate-based values. We believe the present approach provides a better perspective of the role played by the nonlinear parameter and its extension to more general flows is also discussed.

Optimization of fast output from SiPM array for pulse shape discrimination techniques at high counting rates

Plastic scintillators with pulse shape discrimination (PSD) capability, when combined with silicon photo-multipliers (SiPMs), enable the development of compact neutron detectors with lower power consumption and insensitivity to magnetic fields. However, the SiPM array has a significantly large pulse width of the output signal, which leads to discrimination difficulties at high counting rates. There are two outputs of a SiPM array: a standard output for energy measurement and a fast output for timing measurement. The pulse width of the standard output signal is in the range of hundreds nanoseconds, while the pulse width of the other one is in the range of tens nanoseconds. Fast output signals have traditionally been considered suitably for fast timing, however utilizing them for PSD has been little explored. In our previous work, we successfully implemented PSD by using fast output signals from a SiPM. Despite the obvious negative overshoot in fast signals, we achieved PSD of fast signals by using longer sampling times. However, this approach was still limited at high counting rates. This study proposes a compensation network for fast output signals, which is composed of a compensation unit consisting of a resistor and a capacitor. This design effectively compensates the overshoot of the fast output and shortens the pulse width from 60 ns to 44 ns. Additionally, it highlights the frequency amplitude differences between the neutron and gamma signals from fast output in the high-frequency component, thus improving the PSD capability of the fast output.

Simulated TEM imaging of a heavily irradiated metal

We recast the Howie–Whelan equations for generating simulated transmission electron microscope (TEM) images, replacing the dependence on local atomic displacements with atomic positions only. This allows rapid computation of simulated TEM images for arbitrarily complex atomistic configurations of lattice defects and dislocations in the dynamical two beam approximation directly from standard atomistic simulation output files. Large-scale massively-overlapping cascade simulations, performed with molecular dynamics, are used to generate representative high-dose nanoscale defect and dislocation microstructures in tungsten at room temperature. We then compare the simulated TEM images to experimental TEM images with similar irradiation and imaging conditions. The simulated TEM shows ‘white-dot’ damage in weak-beam dark-field imaging conditions, in agreement with our experimental observations and as expected from previous studies, and in bright-field conditions a dislocation network is observed. In this work we also compare the simulated images to the nanoscale lattice defects in the original atomic structures, and find that at high dose the white spots are not only created by small dislocation loops, but rather arise from nanoscale fluctuations in strains around curved sections of dislocation lines.

Probabilistic Neural Networks for Ensemble Postprocessing

Abstract Accurate temperature forecasts are critical for various industries and sectors. We propose a probabilistic neural network (PNN), an extension of the distributional regression network (DRN), for 2-m temperature forecasts, consisting of three variants with different inputs and target variables. The first variant, standardized anomaly probabilistic neural network (SAPNN), employs a two-step approach involving standardized anomalies and global PNN modeling to effectively capture underlying features and anomalies. The second variant, PNN with geographical predictors (PNNGE), incorporates raw and static geographical predictors to enhance predictive performance. The third variant, PNN with station one-hot encoding (PNNEN), utilizes raw with station one-hot encoding predictors to represent geographical information effectively. We compare three PNN variants with two benchmarks: 1) standardized anomaly model output statistics (SAMOS) and 2) three DRN variants identical to those applied to PNN. These evaluations utilize ECMWF data from 2019 to 2020 at 6-h intervals up to 72 h over Hebei, China. Results show that SAPNN and PNNGE are better than SAMOS, while PNNEN notably exhibits a significant 14% improvement in the continuous ranked probability skill score (CRPSS). Moreover, the PNN variants exhibit comparable or superior performance to DRN regarding forecast accuracy, CRPSS, and reliability, showcasing a better-calibrated spread–error relationship. This study highlights the value of the proposed PNN variants with a distribution output in capturing nonlinear relationships within different sources of predictors and improving temperature forecast skills. Significance Statement This study aims to improve the accuracy, skills, and reliability of 2-m temperature forecasts, which are crucial in agriculture and energy management. To achieve this, we extend a popular artificial intelligence framework and explore four data sources with two schemes to systematically compare the predictive performances in making temperature forecasts. The findings of this research are vital as they offer novel ways to improve forecast skills. Imagine having a weather app that is significantly more accurate, enabling you to plan your day better. This study is about discovering innovative approaches to enhance forecast skills and reliability, which could benefit various aspects of our daily lives. One of the new methods even exhibits a 14% improvement in forecast skills.

Effect of a 3-second Off-label Exposure on the Depth of Cure of Eight Resin-based Composites.

This study evaluated the depth of cure (DoC) of eight resin-based composites (RBCs) photocured using one multipeak light-curing unit (LCU) on the standard output setting for the manufacturer's RBC recommended exposure time and at a higher irradiance for 3 seconds. Three conventional RBCs: Tetric EvoCeram (Evo), Tetric N-Ceram (Cer), Tetric Prime (Pri); and five bulk-fill: Tetric N-Ceram Bulk Fill (CerBF), Opus Bulk Fill APS (OpusBF), Opus Bulk Fill Flow APS (OpusF), Tetric PowerFill (PFill) and Tetric PowerFlow (PFlow) were examined. Only PFill and PFlow are formulated to be photocured in 3 seconds. The RBCs were packed into a metal mold and photocured using a Bluephase PowerCure LCU for the RBC manufacturer's recommended exposure time on the standard mode and using the 3-second high irradiance mode. After photocuring, the specimens were immersed in a solvent for 1 hour. The length of the remaining RBC was measured and divided by 2. Data were analyzed using two-way analysis of variance (ANOVA) followed by the Tukey post hoc multiple comparison test (α=0.05). There was no significant difference in the DoC values for PFill and PFlow when photocured using the 3-second high irradiance protocol compared to the lower irradiance standard mode protocol. All other RBCs had significantly lower DoC values (p<0.001) when photocured off-label using the 3-second high irradiance mode. Of the eight RBCs tested, only PFill and PFlow achieved the same DoC when the high irradiance 3-second curing method was used compared to when their longer lower irradiance protocol was used.

Analysis of Working Time Measurement and Incentives Work to Increase Shoe Upper Production at CV. Esa Kalen Jaya

As time goes by, competition in the industrial world is increasing, every company is required to be able to compete with other companies. CV.Esa Kalen Jaya is a company that produces semi-finished goods, namely shoe uppers. Where there are obstacles in the form of not being able to fulfill the production targets that have been determined by the company which results in the company's productivity not being optimal. This research was conducted to calculate standard production time through direct observation using a stopwatch time study as well as improving the regulation of the number of working hours and providing work incentives using the piecework plan method. as an effort to increase company productivity.. Results obtained. The standard output obtained is 4588.48 units/month. And after increasing the number of production working hours to 8 working hours, the results were 5,234 units/month, which means the production target has been achieved. And based on the calculation of incentives that the company will provide in standard hourly wages, namely IDR 7,051.41/hour. It can be concluded that by determining the standard production time and determining the number of working hours, the production results have achieved the set production targets.

Dual-stage color calibration of UAV imagery using multivariate regression and deep learning

Accurate color representation in UAV-based imagery is critical in high-throughput plant phenotyping and precision agriculture for monitoring crop health and identifying nutrient deficiencies or water stress. However, achieving color accuracy is challenging due to varying lighting conditions and flight altitudes in UAV operations. To overcome these challenges, we propose an automated dual-phase approach for consistent color calibration in UAV imagery, integrating traditional color calibration methods with cutting-edge deep learning algorithms. In the first phase, we derived a color calibration matrix (M) for four color boards placed within the imaging area. This involved mapping the observed color values in the UAV images to their true color values, as measured by a colorimeter, using multivariate linear regression (MLR). The individual matrices from each board were then merged into a single, weighted average matrix, which was then applied to the UAV images for color correction, resulting in standardized color outputs. In the second phase, we developed a deep learning model, trained on the MLR-corrected images as a ground truth. This model assigns varying weights for different pixels across the image scene, translating the image from a distorted to a color-corrected image using image-to-image regression without manual intervention. We validated this approach using 184 UAV images of various sizes, ranging from 3969 × 2589 to 15257 × 15098 pixels, captured under a wide array of lighting conditions and altitudes. The results show that our deep learning model significantly outperforms existing methods, with a marginal color error of 1.52 %. This two-phase color correction method greatly improves operational efficiency by removing the necessity for physical color boards in each image and allowing for automated processing after training.

A lesion‐aware automated processing framework for clinical stroke magnetic resonance imaging

AbstractMagnetic resonance imaging (MRI) and MRI based computational modelling studies provide insights into severity and recovery of ischemic stroke patients. The presence of brain lesions, however, can heavily distort and impair state‐of‐the‐art processing frameworks due to abnormal intensity values and tissue distortions. In this study, we introduce and validate the novel “Lesion Aware automated Processing Pipeline (LeAPP).” LeAPP automatically processes clinical stroke MRI data while significantly reducing the impact of pathological artefacts on processing outputs such as structural and functional connectomes (SC, FC). This helps to advance the identification of biomarkers for recovery and mechanism‐based intervention planning using MRI as well as computational approaches. We extended existing frameworks, such as the Human Connectome Project (HCP) minimal processing pipeline, introducing correction steps, and implementing and modifying functional and diffusion processing to cope with MRI acquisition protocols more typical for the clinical context. A total of 51 participants (36 patients, 15 age‐matched controls) were processed across four time points for patients (3–5, 30–40, 85–95, 340–380 days after stroke onset) and one time point for controls. We validated performance using artificial lesioned brains (N = 81), derived from healthy brains and informed by real stroke lesions. The quality of reconstructing the ground truth was quantified on whole brain level and for lesion affected and unaffected regions‐of‐interest (ROIs) for brain parcellations and for SCs. Volume based agreement was evaluated using metrics such as dice coefficient, volume difference or ROI center‐of‐gravity distance while SC based agreement was defined as the difference in network metrics (e.g., node strength, clustering coefficient, or centrality). The observed deviations in reconstructed ground truth brain parcellations and structural connectomes from lesioned brains were significantly reduced for LeAPP, compared to the performance of existing pipelines such as HCP minimal processing pipeline. For instance, in the case of lesion affected ROIS we achieved a mean dice coefficient (where a value of one represents total agreement as defined by an exact overlap of both ROIs) of 0.81 with LeAPP, compared to 0.75 for HCP (p &lt; .0001*). Additionally, the average measured ROI distance (where a value of zero represents no difference) for lesion ROIs was 0.87 for LeAPP in contrast to 1.7 for HCP (p &lt; .0001*), indicating an overall superior performance of LeAPP. The pipeline generates standardized output files ready for brain network modelling for instance with The Virtual Brain software. This novel open‐source automated processing framework contributes to reproducible research and provides a robust framework for automated processing of clinical stroke MRI data, supporting the identification of brain network‐based biomarkers of stroke recovery.

Sitting Pressure Measurements in Wheelchair Users-Can the Effects of Daily Relief Activities Be Depicted?

Seat pressure measurements in wheelchair users have been available for some time; however, repeated measurements from a commercially available pressure mat over 90 min did not differ in the pressure-loaded measurement area or the coordinates of the center of pressure, even in participants who were able to reposition themselves in the wheelchair. The question therefore arises: to what extent are there other parameters that reflect the activity of wheelchair users with the pressure mat? To investigate this, a commercial pressure mat (BodiTrak®) was used to perform the measurements of pressure of 33 adult wheelchair-dependent people with spinal cord injury after 30 and 90 min sitting on the cushion. In addition to the standard output of the pressure mat, graph-based surface analyses (calculation of the area of maximum pressure, calculation of the pressure-loaded measurement area, and pressure-area ratio) was performed retrospectively using Python 3.7. The analysis of the measurements after 30 and 90 min was performed by distinguishing the participants between those who could actively change their position (N = 24) and those who could not (N = 9). The parameters of the pressure mat and the graph-based analyses remained unchanged for active participants. In participants who were unable to actively change their position, the area of maximum pressure and the pressure-area ratio (ratio of maximum pressure area and total pressure-loaded area) increased. Significant differences between minutes 30 and 90 are only found for the pressure-area ratio. Thus, when measuring the seat pressure of wheelchair users, the pressure-area ratio should be taken into account as it reflects the daily relief activities of wheelchair users.

Wind-driven suspended triboelectric-electromagnetic hybrid generator with vibration elimination for environmental monitoring in the high-voltage power transmission line

The sustainable development of the high-voltage power transmission system puts forward more significant requirements IoT sensor network. Currently, a large number of the IoT sensors are installed on the high-voltage transmission lines, and environmental wind energy harvesting around the outdoor power transmission lines enables a promising and feasible solution to supply distributed sensors. However, the traditional floor-mounted wind energy nanogenerator with high requirements for placement and fixation is no longer applicable to the suspension scenario of the high-voltage power transmission lines. In this work, a wind-driven suspended triboelectric-electromagnetic hybrid generator (WS-TEHG) is designed as a suspended wind cup with a three-layered integrated structure, which can be integrated with the damping structure that hanged and installed on the power transmission line, demonstrating the simultaneously achieving high efficiency wind energy harvesting and the operation stability of the device. After the structural and material optimization, WS-TEHG can achieve efficient energy collection under wide-range of the wind speed from 2.6 to 17.3 m s−1, whose TENG and EMG produced instantaneous peak power from 0.2 to 7.9 mW and 0.6–85.1 mW, respectively. Moreover, a standardized power management based on LTC 3588-based double-channel circuit was established to co-manage the outputs of the two generating modules in the hybrid generator and enhance the continuous stability of the standardized voltage output, demonstrating a shorter undervoltage charge accumulation, a faster start-up process, a lower power consumption compared with the traditional power management. Finally, a self-powered environmental monitoring system by a WS-TEHG equipped with a standardized power management circuit was developed to provide the weather and climate information around the power transmission lines, thereby demonstrating a potential and feasible for large-scale self-powered application in the field of the high-voltage power transmission lines.

Accelerated gradient descent using improved Selective Backpropagation

An improved version of the Selective Backpropagation (SBP+) is described which significantly reduces the training time and enhances the generalization. The algorithm eliminates correctly predicted instances from the Backpropagation (BP) in an iteration. The minimum useful content level to define a correct prediction is determined using the hyperparameter c. A new performance measure dm is introduced, which denotes the average of the fraction of remaining instances in the BP during an epoch. With a proper selection of c, a model can be adjusted to learn dominant patterns (mostly found in instances located further away from a decision hyperplane) and skip minor details (the very details which are due to the random error and can be found in instances close to a decision hyperplane). Since the SBP+ assigns a lower correct softmax score to instances located closer to a decision hyperplane, it is shown that — under certain conditions — a Neural Networks (NN) standard crisp output can be converted to a fuzzy membership scores set. The regulating effect of the content level c is compared with other standard techniques and its advantages are outlined. Lastly, the statistical process leading to minimization of the useful total error is explicated and the SBP+ ability to pinpoint outliers is illustrated with an example.

Computer-based facial recognition as an assisting diagnostic tool to identify children with Noonan syndrome

BackgroundNoonan syndrome (NS) is a rare genetic disease, and patients who suffer from it exhibit a facial morphology that is characterized by a high forehead, hypertelorism, ptosis, inner epicanthal folds, down-slanting palpebral fissures, a highly arched palate, a round nasal tip, and posteriorly rotated ears. Facial analysis technology has recently been applied to identify many genetic syndromes (GSs). However, few studies have investigated the identification of NS based on the facial features of the subjects.ObjectivesThis study develops advanced models to enhance the accuracy of diagnosis of NS.MethodsA total of 1,892 people were enrolled in this study, including 233 patients with NS, 863 patients with other GSs, and 796 healthy children. We took one to 10 frontal photos of each subject to build a dataset, and then applied the multi-task convolutional neural network (MTCNN) for data pre-processing to generate standardized outputs with five crucial facial landmarks. The ImageNet dataset was used to pre-train the network so that it could capture generalizable features and minimize data wastage. We subsequently constructed seven models for facial identification based on the VGG16, VGG19, VGG16-BN, VGG19-BN, ResNet50, MobileNet-V2, and squeeze-and-excitation network (SENet) architectures. The identification performance of seven models was evaluated and compared with that of six physicians.ResultsAll models exhibited a high accuracy, precision, and specificity in recognizing NS patients. The VGG19-BN model delivered the best overall performance, with an accuracy of 93.76%, precision of 91.40%, specificity of 98.73%, and F1 score of 78.34%. The VGG16-BN model achieved the highest AUC value of 0.9787, while all models based on VGG architectures were superior to the others on the whole. The highest scores of six physicians in terms of accuracy, precision, specificity, and the F1 score were 74.00%, 75.00%, 88.33%, and 61.76%, respectively. The performance of each model of facial recognition was superior to that of the best physician on all metrics.ConclusionModels of computer-assisted facial recognition can improve the rate of diagnosis of NS. The models based on VGG19-BN and VGG16-BN can play an important role in diagnosing NS in clinical practice.

Task-based automatic keV selection: leveraging routine virtual monoenergetic imaging for dose reduction on clinical photon-counting detector CT * *Portions presented at RSNA 2021 and 2022.

Objective. Photon-counting detector (PCD) CT enables routine virtual-monoenergetic image (VMI) reconstruction. We evaluated the performance of an automatic VMI energy level (keV) selection tool on a clinical PCD-CT system in comparison to an automatic tube potential (kV) selection tool from an energy-integrating-detector (EID) CT system from the same manufacturer. Approach. Four torso-shaped phantoms (20–50 cm width) containing iodine (2, 5, and 10 mg cc−1) and calcium (100 mg cc−1) were scanned on PCD-CT and EID-CT. Dose optimization techniques, task-based VMI energy level and tube-potential selection on PCD-CT (CARE keV) and task-based tube potential selection on EID-CT (CARE kV), were enabled. CT numbers, image noise, and dose-normalized contrast-to-noise ratio (CNRd) were compared. Main results. PCD-CT produced task-specific VMIs at 70, 65, 60, and 55 keV for non-contrast, bone, soft tissue with contrast, and vascular settings, respectively. A 120 kV tube potential was automatically selected on PCD-CT for all scans. In comparison, EID-CT used x-ray tube potentials from 80 to 150 kV based on imaging task and phantom size. PCD-CT achieved consistent dose reduction at 9%, 21% and 39% for bone, soft tissue with contrast, and vascular tasks relative to the non-contrast task, independent of phantom size. On EID-CT, dose reduction factor for contrast tasks relative to the non-contrast task ranged from a 65% decrease (vascular task, 70 kV, 20 cm phantom) to a 21% increase (soft tissue with contrast task, 150 kV, 50 cm phantom) due to size-specific tube potential adaptation. PCD-CT CNRd was equivalent to or higher than those of EID-CT for all tasks and phantom sizes, except for the vascular task with 20 cm phantom, where 70 kV EID-CT CNRd outperformed 55 keV PCD-CT images. Significance. PCD-CT produced more consistent CT numbers compared to EID-CT due to standardized VMI output, which greatly benefits standardization efforts and facilitates radiation dose reduction.