Search

In the medical diagnostics domain, pathology and histology are pivotal for the precise identification of diseases. Digital histopathology, enhanced by automation, facilitates the efficient analysis of massive amount of biopsy images produced on a daily basis, streamlining the evaluation process. This study focuses in Stain Color Normalization (SCN) within a Whole-Slide Image (WSI) cohort, aiming to reduce batch biases. Building on published graphical method, this research demonstrates a mathematical population or data-driven method that optimizes the dependency on the number of reference WSIs and corresponding aggregate sums, thereby increasing SCN process efficiency. This method expedites the analysis of color convergence 50-fold by using stain vector Euclidean distance analysis, slashing the requirement for reference WSIs by more than half. The approach is validated through a tripartite methodology: 1) Stain vector euclidean distances analysis, 2) Distance computation timing, and 3) Qualitative and quantitative assessments of SCN across cancer tumors regions of interest. The results validate the performance of data-driven SCN method, thus potential to enhance the precision and reliability of computational pathology analyses. This advancement is poised to enhance diagnostic processes, therapeutic strategies, and patient prognosis.

This study developed an automated matchmaking system using deep neural networks to enhance the efficiency of pairing researcher biosketches with funders’ requests for proposals (RFPs). In thus U.S., with over 900 federal grant programs and 86,000+ foundations, researchers often spend up to 200 hours on each application due to low success rates, forcing them to apply multiple times a year. Our approach improves on existing systems by fixing issues like unreliable keyword searches, and one-size-fits-all recommendations. We analyzed 12,991 biosketches from a research institution and 2,234 RFPs from the National Institutes of Health, spanning 2014 to 2019. Employing four advanced deep-learning models, utilizing cross and Siamese encoding strategies, we benchmarked their performance against conventional predictive models such as logistic regression and support vector machines. The most effective model integrated BERT with cross-encoding, a post-BERT BiLSTM layer, and back translation (BC2BT), achieving an F1-score of 71.15%. These results demonstrate the potential of sophisticated natural language processing techniques to automate complex matchmaking tasks in the research funding sector. This approach not only improves the precision of matching researchers to suitable funding opportunities but also sets a promising foundation for future advancements in automated funding mechanisms.

Modern navigation solutions rely on a combination of inertial measurement units (IMUs) and a global navigation satellite system (GNSS) receiver to estimate a navigating body's position. In order to produce a high-fidelity solution, the current approach is to utilize a single ultra-low bias navigational grade IMU in the system. However, these high-quality IMUs are expensive, bulky, heavy, and require significant power consumption. This article proposes the fusion of multiple lower-quality IMUs to achieve near-identical or better positional accuracy as a single high-quality sensor to minimize cost, size, weight, and power (C-SWaP) without sacrificing the positional estimation accuracy. The primary focus is on a generalized method to fuse multiple position estimations from multiple co-located IMUs for a single navigating body. The proposed fusion algorithm is applied to simulated data produced by three precision micro-electromechanical systems (MEMS) grade IMU modules (Analog Devices ADIS16465) from two different simulated flight paths. The absolute error is calculated between the position estimation generated by the proposed algorithm and the “truth” position provided by the simulation to determine the accuracy of the final result. The error of the proposed algorithm using the Analog Devices modules is then compared to the error between a single navigational grade IMU (NovAtel IMU-ISA-100 C) and the simulated “truth” position. The results show that using only three precision MEMS grade IMUs; the proposed method can produce identically accurate position estimations as a navigational grade IMU while drastically reducing C-SWaP. This result is further validated in measured data from an instrumented test setup using the above mentioned IMU configurations fused with a standard GPS and compared to a real-time kinematic GNSS setup used as a the third-party ground truth. In addition, the proposed method is further validated by integrating the two navigation systems with a Ku-band radar to produce synthetic aperture radar images. The image produced using the multi-IMU configuration as opposed to the navigation-grade IMU is more focused. Given these results, the proposed method has been validated in theory, simulation, and measurement, resulting in an order of magnitude reduction in cost, size, and power consumption, as well as a three-and-a-half times weight reduction of the overall IMU solution.

This paper presents a twin dual-axis robotic platform system which is designed for the characterization of postural balance under various environmental conditions and quantification of bilateral ankle mechanics in 2 degrees-of-freedom (DOF) during standing and walking. Methods: Validation experiments were conducted to evaluate performance of the system: 1) to apply accurate position perturbations under different loading conditions; 2) to simulate a range of stiffness-defined mechanical environments; and 3) to reliably quantify the joint impedance of mechanical systems. In addition, several human experiments were performed to demonstrate the system’s applicability for various lower limb biomechanics studies. The first two experiments quantified postural balance on a compliance-controlled surface (passive perturbations) and under oscillatory perturbations with various frequencies and amplitudes (active perturbations). The second two experiments quantified bilateral ankle mechanics, specifically, ankle impedance in 2-DOF during standing and walking. The validation experiments showed high accuracy of the platform system to apply position perturbations, simulate a range of mechanical environments, and quantify the joint impedance. Results of the human experiments further demonstrated that the platform system is sensitive enough to detect differences in postural balance control under challenging environmental conditions as well as bilateral differences in 2-DOF ankle mechanics. This robotic platform system will allow us to better understand lower limb biomechanics during functional tasks, while also providing invaluable knowledge for the design and control of many robotic systems including robotic exoskeletons, prostheses and robot-assisted balance training programs. Clinical and Translational Impact Statement— Our robotic platform system serves as a tool to better understand the biomechanics of both healthy and neurologically impaired individuals and to develop assistive robotics and rehabilitation training programs using this information.

Histopathologic evaluation of Hematoxylin & Eosin (H&E) stained slides is essential for disease diagnosis, revealing tissue morphology, structure, and cellular composition. Variations in staining protocols and equipment result in images with color nonconformity. Although pathologists compensate for color variations, these disparities introduce inaccuracies in computational whole slide image (WSI) analysis, accentuating data domain shift and degrading generalization. Current state-of-the-art normalization methods employ a single WSI as reference, but selecting a single WSI representative of a complete WSI-cohort is infeasible, inadvertently introducing normalization bias. We seek the optimal number of slides to construct a more representative reference based on composite/aggregate of multiple H&E density histograms and stain-vectors, obtained from a randomly selected WSI population (WSI-Cohort-Subset). We utilized 1,864 IvyGAP WSIs as a WSI-cohort, and built 200 WSI-Cohort-Subsets varying in size (from 1 to 200 WSI-pairs) using randomly selected WSIs. The WSI-pairs' mean Wasserstein Distances and WSI-Cohort-Subsets' standard deviations were calculated. The Pareto Principle defined the optimal WSI-Cohort-Subset size. The WSI-cohort underwent structure-preserving color normalization using the optimal WSI-Cohort-Subset histogram and stain-vector aggregates. Numerous normalization permutations support WSI-Cohort-Subset aggregates as representative of a WSI-cohort through WSI-cohort CIELAB color space swift convergence, as a result of the law of large numbers and shown as a power law distribution. We show normalization at the optimal (Pareto Principle) WSI-Cohort-Subset size and corresponding CIELAB convergence: a) Quantitatively, using 500 WSI-cohorts; b) Quantitatively, using 8,100 WSI-regions; c) Qualitatively, using 30 cellular tumor normalization permutations. Aggregate-based stain normalization may contribute in increasing computational pathology robustness, reproducibility, and integrity.

The illegal exploitation of protected marine environments has consistently threatened the biodiversity and economic development of coastal regions. Extensive monitoring in these– often remote– areas is challenging. Machine learning methods are useful in object detection and classification tasks and have the potential to underpin techniques for the development of robust monitoring systems to overcome this problem. However, development is hindered due to the limited number of publicly available labelled and curated datasets. Furthermore, there are relatively few open-source state-of-the-art methods to be used for evaluation. This paper presents an investigation of automated classification methods using underwater acoustic signals to infer the presence and type of vessels navigating in coastal regions. Various combinations of deep convolutional neural network architectures, and preprocessing filter layers, were evaluated using a new dataset based on a subset of the extensive open-source Ocean Networks Canada hydrophone data. Tests were conducted in which VGGNet and ResNet networks were applied to classify the input data. The data was preprocessed using either Constant Q Transform (CQT), Gammatone, Mel spectrogram, or a combination of these filters. With over 97% accuracy, using all three preprocessing representations simultaneously yielded the most reliable result. However, high accuracies of 94.95% were achieved using CQT as the preprocessing filter for a ResNet-based convolutional neural network, providing a trade-off between model complexity and accuracy; a result that is more than 10% higher than previously reported approaches. This more accurate classifier for underwater acoustics could be used as a reliable autonomous monitoring system in maritime environments.

It is vital for firms to measure, and, report on, the impact of eco-innovation towards their environmental performance. Additive Manufacturing is an eco-innovation that contributes towards reducing the environmental impact of manufacturing. However, there is a lack of practical tools for management to monitor and report on firms’ eco-innovation implementation and performance. This paper presents a prototype Eco-innovation Management Information System (EiMIS), developed to enable a firm’s management to systematically manage and report on eco-innovation performance. Using a top-down approach, the EiMIS is developed through utilizing the C# and SQL programming languages for the Graphical User Interface (GUI), Database Management System (DBMS) and other technologies to collect, store (into a relational database) and manage eco-innovation performance data. A 4-tier architecture enables the EiMIS to: store, view and retrieve or generate eco-innovation performance reports; and generally, manage firms’ eco-innovation performance data. This developed prototype could, help introduce information system platforms that: assist firms to collect data that can model their additive manufacturing; ease firms’ eco-innovation performance reporting; improve environmental data management capabilities; and complement firms’ digital transformation strategy.

Global positioning system (GPS) and inertial measurement units (IMUs) are often combined to produce navigation systems for airborne imaging platforms. The current state-of-the-art radar technology allows for radars to pulse at very high rates. GPS and IMU update rates are not fast enough to accurately report the platform position for each radar pulse. Independent GPS and IMUs cannot provide positional accuracy for long term stability. Traditional techniques, such as the Kalman and particle filter, are used to fuse GPS and IMU measurements. The Kalman filter excels for linear and Gaussian systems whereas the particle filter excels at non-linear and non-Gaussian systems. Sensor fusion techniques are used to help correct for IMU errors and provide the positional accuracy required for synthetic aperture radar (SAR) imaging applications. However, SAR requires the fusion algorithms to provide faster update rates. This paper explores the use of an up-sampled particle filter (UPF) for SAR to provide highly accurate position estimates at sampling frequencies comparable to radar pulse rates and overcome the limitations of standard interpolation techniques. This up-sampled particle filter is proven through simulations and instrumentation with a NovAtel GPS and IMU. The UPF technique allows for the GPS/IMU sampling rate to be different from the radar pulse repetition frequency (PRF) while providing accurate position solutions for each radar pulse capable of compensating for the phase history required for focusing a SAR image. The algorithms are instrumented in a SAR system and the position estimates are further validated and demonstrated through captured SAR images.

Background: Lyme disease (caused by Borrelia burgdorferi) is an infectious disease transmitted to humans by a bite from infected blacklegged ticks (Ixodes scapularis) in eastern North America. Lyme disease can be prevented if antibiotic prophylaxis is given to a patient within 72 hours of a blacklegged tick bite. Therefore, recognizing a blacklegged tick could facilitate the management of Lyme disease. Methods: In this work, we build an automated detection tool that can differentiate blacklegged ticks from other tick species using advanced computer vision approaches in real-time. Specially, we use convolution neural network models, trained end-to-end, to classify tick species. Also, advanced knowledge transfer techniques are adopted to improve the performance of convolution neural network models. Results: Our best convolution neural network model achieves 92% accuracy on unseen tick species. Conclusion: Our proposed vision-based approach simplifies tick identification and contributes to the emerging work on public health surveillance of ticks and tick-borne diseases. In addition, it can be integrated with the geography of exposure and potentially be leveraged to inform the risk of Lyme disease infection. This is the first report of using deep learning technologies to classify ticks, providing the basis for automation of tick surveillance, and advancing tick-borne disease ecology and risk management.

In this paper, an optimization procedure for synthesizing shaped beams with an arbitrary geometry phased array system with real-time capability is presented. The algorithm involves a combination of an arbitrary extension of the Woodward-Lawson synthesis procedure with the magnitude least-squares optimization method. This simple combination significantly reduces the optimization time, enabling real-time beam shaping in response to changing requirements on beam shape or evolving geometry in distributed arrays. The technique places no restrictions on element positioning in all three spatial dimensions and is demonstrated to accurately reproduce the desired beam shape in both angular dimensions. The algorithm is demonstrated for multiple beam cases with a large, randomly generated array in computational simulation. The applicability of the algorithm to practical phased array hardware is also demonstrated using full-wave electromagnetic simulations and measurements of a realized arbitrary phased array system using a near-field scanner in an anechoic chamber.

The landing point of a saccade defines the new fixation region, the new region of interest. We asked whether it was possible to predict the saccade landing point early in this very fast eye movement. This work proposes a new algorithm based on LSTM networks and a fine-grained loss function for saccade landing point prediction in real-world scenarios. Predicting the landing point is a critical milestone toward reducing the problems caused by display-update latency in gaze-contingent systems that make real-time changes in the display based on eye tracking. Saccadic eye movements are some of the fastest human neuro-motor activities with angular velocities of up to 1,000°/s. We present a comprehensive analysis of the performance of our method using a database with almost 220,000 saccades from 75 participants captured during natural viewing of videos. We include a comparison with state-of-the-art saccade landing point prediction algorithms. The results obtained using our proposed method outperformed existing approaches with improvements of up to 50% error reduction. Finally, we analyzed some factors that affected prediction errors including duration, length, age, and user intrinsic characteristics.

We develop an optimization approach for the planning problem of configuring the positions of sensors within a sensor network for minimization of the travel length required to service the sensor locations. This is in contrast to existing approaches whereby the coverage of the sensor network is an objective of the optimization; in this new optimization approach the level of coverage is treated as a constraint. By beginning with an over-populated sensor network and then alternating between sensor repositioning and sensor removal, we create an optimization procedure that solves this difficult practical planning problem. A formal rule for switching between the repositioning and removal components of the optimization strategy is developed. Numerical example problems are presented to illustrate the method and show its performance against a heuristic optimization approach.

Demand response for electrical power networks is a mature field that has yielded efficiency and resilience benefits like managing the peaks and valleys of electricity usage and reduction in peak electricity usag. However, only recently has the study of the counterpart to demand response for electric power in natural gas started to receive similar levels of attention. Natural gas systems are increasingly operating at or near capacity, which challenges these systems to meet all the needs for gas, especially during severe winter weather. However, unlike demand response programs in electrical networks, demand response in gas networks cannot shift peak usage. Here, we develop analogues to demand response that can help improve the resilience of natural gas systems by reducing peak consumption and thereby limiting potential disruptions such events can cause. This paper develops a mathematical formulation to support a residential-level demand response analogue for natural gas based on current and anticipated smart thermostat technologies. The mathematical formulation takes the form of an optimal control problem (OCP) that leverages physics constraints to model temperature changes, balance equitable service, and optimize the gas consumption for collections of houses. On test problems, the formulation demonstrates significant benefits, including the ability to cut peak demand by 15% while still ensuring equitable service to customers.

Faced with the COVID-19 pandemic, the US system for developing and testing technologies was challenged in unparalleled ways. This article describes the multi-institutional, transdisciplinary team of the “RADxSM Tech Test Verification Core” and its role in expediting evaluations of COVID-19 testing devices. Expertise related to aspects of diagnostic testing was coordinated to evaluate testing devices with the goal of significantly expanding the ability to mass screen Americans to preserve lives and facilitate the safe return to work and school. Focal points included: laboratory and clinical device evaluation of the limit of viral detection, sensitivity, and specificity of devices in controlled and community settings; regulatory expertise to provide focused attention to barriers to device approval and distribution; usability testing from the perspective of patients and those using the tests to identify and overcome device limitations, and engineering assessment to evaluate robustness of design including human factors, manufacturability, and scalability.

We evaluated a 16-channel loop + dipole (LD) transceiver antenna array with improved specific absorption rate (SAR) efficiency for 10.5 Tesla (T) human head imaging applications. Three different array designs with equal inner dimensions were considered: an 8-channel dipole antenna, an 8-channel loop, and a 16-channel LD antenna arrays. Signal-to-noise ratio (SNR) and B1+ efficiency (in units of μT per √W) were simulated and measured in 10.5 T magnetic resonance imaging (MRI) experiments. For the safety validation, 10 g SAR and SAR efficiency (defined as the B1+ over √ (peak 10 g SAR)) were calculated through simulation. Finally, high resolution porcine brain images were acquired with the 16-channel LD antenna array, including a fast turbo-spin echo (TSE) sequence incorporating B1 shimming techniques. Both the simulation and experiments demonstrated that the combined 16-channel LD antenna array showed similar B1+ efficiency compared to the 8-channel dipole antenna and the 8-channel loop arrays in a circular polarized (CP) mode. In a central 2 mm × 2 mm region of the phantom, however, the 16-channel LD antenna array showed an improvement in peak 10 g SAR of 27.5 % and 32.5 % over the 8-channel dipole antenna and the 8-channel loop arrays, respectively. We conclude that the proposed 16-channel head LD antenna array design is capable of achieving ~7% higher SAR efficiency at 10.5 T compared to either the 8-channel loop-only or the 8-channel dipole-only antenna arrays of the same dimensions.

A common challenge forward-looking utilities are facing when deploying advanced applications that facilitate voltage optimization and service restoration is to provide adequate sensor data for a Distribution System State Estimator (DSSE) so that it provides sufficiently accurate estimates of the system states to enable these applications in an operational environment. We developed a stochastic method that informs telemetry and operational forecasting requirements by quantifying the DSSE performance in supporting advanced applications. The performance metric used is the α risk, which is the likelihood of a DSSE giving a false positive when determining if voltage and loading constraints are met. We applied this method to six real-world industrial/commercial/residential distribution circuits and evaluated α risk improvements provided by circuit-level sensors and operational forecasting. The results show that a combination of sensor deployment schemes was needed to reduce the α risk for undervoltage to effectively zero. Also, sensors deployed at large loads significantly reduce c risks on industrial/commercial circuits while operational forecasting consistently reduces α risks on all circuits. The practical method does not require advanced mathematics and can be readily used by utilities to inform grid modernization investments in sensor technologies so that advanced applications can be executed optimally and violation-free.

The Compact Ocean Wind Vector Radiometer (COWVR) was developed at the Jet Propulsion Laboratory as a proof of concept technology demonstration mission for the Air Force. COWVR is a fully polarimetric imaging radiometer system operating at 18.7, 23.8, and 33.9 GHz . Its receiver subsystem is based on the Jason-3 advanced microwave radiometer, which was launched in early 2016 and is currently in operation. Its enabling components are presented in this article, including the phase-matched waveguide components, noise injection, receivers, and polarimetric backend units.

Objectives: Worldwide efforts to protect front line providers performing endotracheal intubation during the COVID-19 pandemic have led to innovative devices. Authors evaluated the aerosol containment effectiveness of a novel intubation aerosol containment system (IACS) compared with a recently promoted intubation box and no protective barrier. Methods: In a simulation center at the authors' university, the IACS was compared to no protective barrier and an intubation box. Aerosolization was simulated using a commercial fog machine and leakage of aerosolize mist was visually assessed. Results: The IACS appeared to contain the aerosolized mist, while the intubation box allowed for mist to contact the laryngoscopist and contaminate the clinical space through arm port holes and the open caudal end. Both devices protected the laryngoscopist better than no protective barrier. Discussion: The IACS with integrated sleeves and plastic drape appears to offer superior protection for the laryngoscopist and assistant providers from aerosolized particles.

This paper presents a selective metallization method for the newly developed 3D printable thermoplastic polyurethane elastomers (TPU): FilaFlex ®, SemiFlexTM, PolyFlexTM, and NinjaFlex®. Silver nanoparticles were fabricated in-situ by photo-reduction on the surface of TPUs and acted as catalysts for copper ion adsorption. This method demonstrates the successful fabrication of copper patterns on flexible TPU filaments. Furthermore, PolyflexTM and acrylonitrile butadiene styrene (ABS) filaments printed in the same part have been used to enable selective electroless plating directly on PolyflexTMmaterial. Electroless copper deposited onto PolyflexTMhas a sheet resistance of (139.4 ± 7.2) m Ω /□ and a copper conductivity of (1.1 ± 0.1) × 107 S/m that is comparable with bulk copper. Copper-plated PolyflexTMinterconnects were fabricated as a proof of concept demonstrators.

Current trends in spectrum regulation show that more and more unlicensed and shared spectrum bands are poised to be opened up for mobile communication. However, the question remains how to best utilize this spectrum and build efficient networks, and if the time has come for newer approaches to be considered for the next generation system. In this work, we propose a coordinated shared spectrum framework that can be considered for next generation cellular standardization. In designing the framework, we aim to improve upon the current unlicensed access schemes toward increasing spectral efficiency in highly-dense networks. To this end, we demonstrate that with the proposed framework both throughput and access delay can be significantly improved over the state-of-the-art LAA system. Also, by optimizing access delay and improving inter-operator resource fairness, the system is designed to be more amenable for operators to invest in deploying networks using shared spectrum. We further show that, by taking advantage of small timescale variations in traffic demand, large statistical multiplexing gains are possible through dynamic sharing instead of static, hard splitting of shared spectrum, as in the current CBRS system.