articles+ search results

штучний інтелект, fpga, ші як сервіс, гетерогенні проєкти ші систем, апаратні прискорювачі ші, dpu, інструментальні засоби розробки ші, xrt, Motor vehicles. Aeronautics. Astronautics, and TL1-4050

The subject of study in this article is modern technologies, tools and methods of building AI systems as a service using FPGA as a platform. The goal is to analyze modern technologies and tools used to develop FPGA-based projects for systems that implement artificial intelligence as a service and to prepare a practical AI service prototype. Task: to analyze the evolution of changes in the products of leading manufacturers of programmable logic devices and experimental and practical examples of the implementation of the paradigm of continuous reprogramming of programmable logic; analyze the dynamics of changes in the development environment of programmable logic systems for AI; analyze the essential elements of building projects for AI systems using programmable logic. According to the tasks, the following results were obtained. The area of application of hardware implementation of artificial intelligence for on-board and embedded systems including airspace industry, smart cars and medical systems is analyzed. The process of programming FPGA accelerators for AI projects is analyzed. The analysis of the capabilities of FPGA with HBM for building projects that require enough of high speed memory is performed. Description languages, frameworks, the hierarchy of tools for building of hardware accelerators for AI projects are analyzed in detail. The stages of prototyping of AI projects using new FPGA development tools and basic DPU blocks are analyzed. The parameters of the DPU blocks were analyzed. Practical steps for building such systems are offered. The practical recommendations for optimizing the neural network for FPGA implementation are given. The stages of neural network optimization are provided. The proposed steps include pruning of branches with low priority and the use of fixed point computations with custom range based on the requirements of an exact neural network. Based on these solutions, a practical case of AI service was prepared, trained and tested. Conclusions. The main contribution of this study is that, based on the proposed ideas and solutions, the next steps to create heterogeneous systems based on the combination of three elements are clear: AI as a service, FPGA accelerators as a technology for improving performance, reliability and security, and cloud or Edge resources to create FPGA infrastructure and AI as service. The development of this methodological and technological basis is the direction of further R&D.

3D printing, Zinc submicron particles, Osteoinductivity, Anti-inflammatory, Bone defect repair, Materials of engineering and construction. Mechanics of materials, TA401-492, Biology (General), and QH301-705.5

Long-term nonunion of bone defects has always been a major problem in orthopedic treatment. Artificial bone graft materials such as Poly (lactic-co-glycolic acid)/β-tricalcium phosphate (PLGA/β-TCP) scaffolds are expected to solve this problem due to their suitable degradation rate and good osteoconductivity. However, insufficient mechanical properties, lack of osteoinductivity and infections after implanted limit its large-scale clinical application. Hence, we proposed a novel bone repair bioscaffold by adding zinc submicron particles to PLGA/β-TCP using low temperature rapid prototyping 3D printing technology. We first screened the scaffolds with 1 wt% Zn that had good biocompatibility and could stably release a safe dose of zinc ions within 16 weeks to ensure long-term non-toxicity. As designed, the scaffold had a multi-level porous structure of biomimetic cancellous bone, and the Young's modulus (63.41 ± 1.89 MPa) and compressive strength (2.887 ± 0.025 MPa) of the scaffold were close to those of cancellous bone. In addition, after a series of in vitro and in vivo experiments, the scaffolds proved to have no adverse effects on the viability of BMSCs and promoted their adhesion and osteogenic differentiation, as well as exhibiting higher osteogenic and anti-inflammatory properties than PLGA/β-TCP scaffold without zinc particles. We also found that this osteogenic and anti-inflammatory effect might be related to Wnt/β-catenin, P38 MAPK and NFkB pathways. This study lay a foundation for the follow-up study of bone regeneration mechanism of Zn-containing biomaterials. We envision that this scaffold may become a new strategy for clinical treatment of bone defects.

Artificial intelligence, Machine learning, AutoML, Business analytics, Data-driven decision making, Digital transformation, Cybernetics, Q300-390, Electronic computers. Computer science, and QA75.5-76.95

The realization that AI-driven decision-making is indispensable in today's fast-paced and ultra-competitive marketplace has raised interest in industrial machine learning (ML) applications significantly. The current demand for analytics experts vastly exceeds the supply. One solution to this problem is to increase the user-friendliness of ML frameworks to make them more accessible for the non-expert. Automated machine learning (AutoML) is an attempt to solve the problem of expertise by providing fully automated off-the-shelf solutions for model choice and hyperparameter tuning. This paper analyzed the potential of AutoML for applications within business analytics, which could help to increase the adoption rate of ML across all industries. The H2O AutoML framework was benchmarked against a manually tuned stacked ML model on three real-world datasets. The manually tuned ML model could reach a performance advantage in all three case studies used in the experiment. Nevertheless, the H2O AutoML package proved to be quite potent. It is fast, easy to use, and delivers reliable results, which come close to a professionally tuned ML model. The H2O AutoML framework in its current capacity is a valuable tool to support fast prototyping with the potential to shorten development and deployment cycles. It can also bridge the existing gap between supply and demand for ML experts and is a big step towards automated decisions in business analytics. Finally, AutoML has the potential to foster human empowerment in a world that is rapidly becoming more automated and digital.

Printing, Three-Dimensional, Hydrogen Peroxide, and Luminescence

Low force stereolithography is exploited for the first time for one-step facile fabrication of chemiluminescence (CL) flow-through cells that bear unrivalled features as compared to those available through milling or blowing procedures or alternative 3D printing technologies. A variety of bespoke cross-section geometries with polyhedral features (namely, triangular, square, and five-side polygon) as well as semicircular cross-section are herein critically evaluated in terms of analytical performance against the standardcircular cross-section in a flat spirally-shape format. The idea behind is to maximize capture of elicited light by the new designs while leveraging 3D printing further for fabrication of (i) customized gaskets that enable reliable attaching of the active mixing zone of the CL cell to the detection window, (ii) in-line 3D-printed serpentine reactors, and (iii) flow confluences with tailorable shapes for enhancing mixing of samples with CL reagents. Up to twenty transparent functional cells were simultaneously fabricated without inner supports following post-curing and surface treatment protocols lasting less than 5 h. In fact, previous attempts to print spirally-shaped cells in one-step by resorting to less cost effective photopolymer inkjet printing technologies were unsuccessful because of the requirement of lengthy procedures (>15 days) for quantitative removal of the support material. By exploiting the phthalazinedione-hydrogen peroxide chemistry as a model reaction, the five-side irregular pentagon cell exhibited superior analytical figures of merit in terms of LOD, dynamic range and intermediate precision as compared to alternative designs. Computational fluid dynamic simulations for mapping velocities at the entry region of the spiral cell corroborated the fact that the 5-side polygon cross-section flow-cell with Y-type confluence permitted the most efficient mixing of reagents and sample while enabling larger flow velocities near the inlet that contribute to a more efficient capture of the photons from the flash-type reaction. The applicability of the 3D-printed 5-side polygon CL cell for automatic determination of hydrogen peroxide using a computerized hybrid flow system was demonstrated for the analysis of high matrix samples, viz., seawater and saliva, with relative recoveries ranging from 83 to 103%.
Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
(Copyright © 2023 Elsevier B.V. All rights reserved.)

Brain diagnostic imaging, Algorithms, Image Processing, Computer-Assisted methods, Magnetic Resonance Imaging methods, and Software

Purpose: Introduce Shimming Toolbox ( https://shimming-toolbox.org), an open-source software package for prototyping new methods and performing static, dynamic, and real-time B0 shimming as well as B1 shimming experiments.
Methods: Shimming Toolbox features various field mapping techniques, manual and automatic masking for the brain and spinal cord, B0 and B1 shimming capabilities accessible through a user-friendly graphical user interface. Validation of Shimming Toolbox was demonstrated in three scenarios: (i) B0 dynamic shimming in the brain at 7T using custom AC/DC coils, (ii) B0 real-time shimming in the spinal cord at 3T, and (iii) B1 static shimming in the spinal cord at 7T.
Results: The B0 dynamic shimming of the brain at 7T took about 10 min to perform. It showed a 47% reduction in the standard deviation of the B0 field, associated with noticeable improvements in geometric distortions in EPI images. Real-time dynamic xyz-shimming in the spinal cord took about 5 min and showed a 30% reduction in the standard deviation of the signal distribution. B1 static shimming experiments in the spinal cord took about 10 min to perform and showed a 40% reduction in the coefficient of variation of the B1 field.
Conclusion: Shimming Toolbox provides an open-source platform where researchers can collaborate, prototype and conveniently test B0 and B1 shimming experiments. Future versions will include additional field map preprocessing techniques, optimization algorithms, and compatibility across multiple MRI manufacturers.
(© 2022 International Society for Magnetic Resonance in Medicine.)

Humans, Creatinine chemistry, Limit of Detection, Smartphone, Electrochemical Techniques, Electrodes, Graphite chemistry, and Nanoparticles chemistry

3D printing technologies are an attractive for fabricating electrochemical sensors due to their ease of operation, freedom of design, fast prototyping, low waste, and low cost. We report the fabrication of a simple 3D-printed electrochemical sensing device for non-enzymatic detection of creatinine, an important indicator of renal function. To create the 3D-printed electrodes (3DE), carbon black/polylactic acid (CB/PLA) composite filament was used. The 3DE was activated using 0.5 M NaOH via amperometry prior to use to improve electrochemical performance. To give selectivity for creatinine, the activated 3DE was modified with a copper oxide nanoparticle-ionic liquid/reduced graphene oxide (CuO-IL/rGO) composite. The modified 3DE was characterized using microscopy and electrochemistry. Cyclic voltammetry and amperometry were used to evaluate sensor performance. The modified 3DE provided electrocatalytic activity towards creatinine without enzymes. Under optimal conditions, the modified 3DE directly coupled with a portable smartphone potentiostat exhibited the linear detection range of 0.5-35.0 mM, and the limit of detection was 37.3 μM, which is sufficient for detecting creatinine in human urine samples. Furthermore, the other physiological compounds present in human urine were not detected on the modified 3DE. Therefore, the modified 3DE could be a tool for effective creatinine screening in the urine.
Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
(Copyright © 2022 Elsevier B.V. All rights reserved.)

Purpose: to evaluate the efficacy of dual-purpose computer-generated splint in guiding the proximal and the distal segment in bilateral sagittal split osteotomy.
Patients and Method: It was a prospective case series study directed on 8 class III patients indicating the need of maxillary advancement and mandibular set back by bilateral sagittal split osteotomy. A CAD/CAM splint is generated to guide the distal segment to the stable maxilla and at the same time a grooved extension to engage the proximal segment ensuring the condyle in its planned position during fixation. The primary outcome was measured by calculating the difference between the pre- and post-operative condylar segment position.
Results: The present study included five female patient and three male patient with mean age of 28.4 ± 5.1 years. The accuracy of the splint in positioning the mandibular proximal segment showed promising results ranging from 2.59 to 0.49.
Conclusion: The dual-purpose splint introduced in this study showed satisfied results in maintaining the pre-operative condylar position while securing the distal segment in the desired plan.
Competing Interests: Conflict of interestThe authors declare that there is no conflict of interest (Mamdouh Ahmed declares that there is no conflict of interest, Sherif Ali declares that there is no conflict of interest and Sara Soliman declares that there is no conflict of interest).
(© This is a U.S. government work and not under copyright protection in the U.S.; foreign copyright protection may apply 2022.)

Metal additive manufacturing (MAM), Laser powder bed fusion (L-PBF), Machine learning (ML), Process parameter optimization, Anomaly detection, and Technology

Metal Additive Manufacturing (MAM) applications are growing rapidly in high-tech industries such as biomedical and aerospace, and in many other industries including tooling, casting, automotive, oil and gas for production and prototyping. The onset of Laser Powder Bed Fusion (L-PBF) technology proved to be an efficient technique that can convert metal additive manufacturing into a reformed process if anomalies occurred during this process are eliminated. Industrial applications demand high accuracy and risk-free products whereas prototyping using MAM demand lower process and product development time. In order to address these challenges, Machine Learning (ML) experts and researchers are trying to adopt an efficient method for anomaly detection in L-PBF so that the MAM process can be optimized and desired final part properties can be achieved. This review provides an overview of L-PBF and outlines the ML methods used for anomaly detection in L-PBF. The paper also explains how ML methods are being used as a step forward toward enabling the real-time process control of MAM and the process can be optimized for higher accuracy, lower production time, and less material waste. Authors have a strong believe that ML techniques can reform MAM process, whereas research concerned to the anomaly detection using ML techniques is limited and needs attention.This review has been done with a hope that ML experts can easily find a direction and contribute in this field.

Android, Communication, Deaf, Disability, Dumb, Hardware, and Science

People living with disability constitute a significant percentage of the world population. For many people with disabilities, assistance and support are prerequisites for participating in societal activities. This research work developed a hardware prototype of a robotic hand forfor sign language communication with persons living with hard-of-hearing disabilities (deaf and/or dumb). The prototype has three basic modules: the input unit, the control unit, and the robotic hand. The input unit is designed as an Android-based mobile application with speech recognition capabilities while the control unit is ATMEGA 2560 microcontroller board. The robotic hand is constructed using locally available materials (bathroom Slippers, expandable rubber, straw pipe, and tiny rope) together with three servo motors and is designed to look and perform movements similar to a human hand. The prototype was evaluated quantitatively in terms of empirical accuracy and response time. It was also evaluated qualitatively by thirty-five (35) users which included fifteen (15) experience ASL users, eighteen (18) non-experience ASL users, and two (2) ASL experts, who completed questionnaires to rate the prototype on a 5-point Likert scale in terms of five parameters: functionality, reliability, ease of use, efficiency, and portability. An accuracy of 78.43% with an average response time of 2 s was obtained from empirical experiments. Statistical analysis of user responses showed that 97%, 68%, 77%, 80%, and 83% of users rated the system as above average for functionality, reliability, ease of use, efficiency, and portability, respectively. The robotic hand effectively communicates American Sign Language which includes English Alphabets, numbers (1–9), and some selected common words, which can be demonstrated with a single hand for hard of hearing persons. To the best of our knowledge, this work is the first ASL robotic hand that is based on locally sourced cost-effective materials, and we build on flaws from existing literature, most of which are either template-based, not real-time, or expensive. In terms of future work, the prototype can be improved by extending the single robotic hand to a fully robotic body with two hands.

Hydrodynamics, Kinetics, Ultraviolet Rays, Disinfection methods, and Water Purification methods

We developed and studied one of the first high-flow UV-LED water disinfection reactors applicable to point-of-entry (POE) water disinfection. A multiphysics computational model was created to predict the performance of UV reactor design concepts by modeling the synergic effect of radiation, hydrodynamics, and the inactivation kinetics of microorganisms. The geometrical optics that describe light propagation in terms of rays were employed to model the radiation profile of multiple UV-LEDs with optical components in complex reactor geometries, the first account of such an approach. The computational solution of the mass, momentum, and species equations was applied to model the hydrodynamics and kinetics. We designed a reactor through a detailed computational study of the optical and hydrodynamic performance of various design strategies. Highly efficient UV fluence distribution in the reactor was achieved by creating nearly collimated UV radiation beams across the reactor and managing the hydrodynamics using a flow distributor. We fabricated a prototype of the optimized reactor design for experimental studies. Biodosimetry tests were conducted for various flow rates and UV transmittances (UVTs), and the experimental results were compared with the model predictions. The design, which employed 14 UV-LEDs assembled over custom-made optical modules, resulted in a reduction equivalent dose (RED) of 65 mJ/cm 2 at a flow rate of 20 liters per minute (LPM) while consuming about 50 W energy. This reactor design required only 0.05 W radiant power per LPM flow rate to achieve an NSF Class A UV dose equivalent of 40 mJ/cm 2 . The findings of this study provide insights into UV-LED reactor development strategies as well as the creation and application of reactor virtual prototyping tools for designing and optimizing highly efficient UV-LED reactors.
Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
(Copyright © 2023. Published by Elsevier Ltd.)