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.

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.

cardiopulmonary resuscitation, Mechanical ventilation, Mechanical chest-compression, Respiratory monitoring, Thiel embalmed cadaver, Biomechanics, Computer applications to medicine. Medical informatics, R858-859.7, Science (General), and Q1-390

The data presented in this article relate to the research article, “Reliability of mechanical ventilation during continuous chest compressions: a crossover study of transport ventilators in a human cadaver model of CPR” [1].This article contains raw data of continuous recordings of airflow, airway and esophageal pressure during the whole experiment. Data of mechanical ventilation was obtained under ongoing chest compressions and from repetitive measurements of pressure-volume curves. All signals are presented as raw time series data with a sample rate of 200Hz for flow and 500 Hz for pressure. Additionally, we hereby publish extracted time series recordings of force and compression depth from the used automated chest compression device. Concomitantly, we report tables with time stamps from our laboratory book by which the data can be sequenced into different phases of the study protocol.We also present a dataset of derived volumes which was used for statistical analysis in our research article together with the used exclusion list.The reported dataset can help to understand mechanical properties of Thiel-embalmed cadavers better and compare different models of cardiopulmonary resuscitation (CPR). Future research may use this data to translate our findings from bench to bedside. Our recordings may become useful in developing respiratory monitors for CPR, especially in prototyping and testing algorithms of such devices.

aged, deprescribing, frailty, polypharmacy, primary health care, referral and consultation, Medicine (General), R5-920, Public aspects of medicine, and RA1-1270

Abstract Background In older people living with frailty, polypharmacy can lead to preventable harm like adverse drug reactions and hospitalization. Deprescribing is a strategy to reduce problematic polypharmacy. All stakeholders should be actively involved in developing a person‐centred deprescribing process that involves shared decision‐making. Objective To co‐design an intervention, supported by a logic model, to increase the engagement of older people living with frailty in the process of deprescribing. Design Experience‐based co‐design is an approach to service improvement, which uses service users and providers to identify problems and design solutions. This was used to create a person‐centred intervention with the potential to improve the quality and outcomes of the deprescribing process. A ‘trigger film’ showing older people talking about their healthcare experiences was created and facilitated discussions about current problems in the deprescribing process. Problems were then prioritized and appropriate solutions were developed. The review located the solutions in the context of current processes and procedures. An ideal care pathway and a complex intervention to deliver better care were developed. Setting and Participants Older people living with frailty, their informal carers and professionals living and/or working in West Yorkshire, England, UK. Deprescribing was considered in the context of primary care. Results The current deprescribing process differed from an ideal pathway. A complex intervention containing seven elements was required to move towards the ideal pathway. Three of these elements were prototyped and four still need development. The complex intervention responded to priorities about (a) clarity for older people about what was happening at all stages in the deprescribing process and (b) the quality of one‐to‐one consultations. Conclusions Priorities for improving the current deprescribing process were successfully identified. Solutions were developed and structured as a complex intervention. Further work is underway to (a) complete the prototyping of the intervention and (b) conduct feasibility testing. Patient or Public Contribution Older people living with frailty (and their informal carers) have made a central contribution, as collaborators, to ensure that a complex intervention has the greatest possible potential to enhance the experience of deprescribing medicines.

Perforations, Subtractive manufacturing, Negative Poisson’s ratio, 3D auxetic structures, Materials of engineering and construction. Mechanics of materials, and TA401-492

Prototyping of three-dimensional mechanical metamaterials that exhibit negative Poisson’s ratio is usually performed through additive manufacturing. Although this technique has a huge potential, its use to engineer mechanical metamaterials for consumer products is still challenging. In this work, a novel design method is being proposed where 3D auxetic metamaterials can be produced by introducing continuous voids of constant cross-sectional area. Such voids would be inserted at strategic positions in different perpendicular planes of a solid block to obtain a continuous three-dimensional mechanical metamaterial that can exhibit the desired mechanical characteristics. The use of continuous voids to design the 3D meatamaterial makes it possible to use additive manufacturing, subtractive manufacturing as well as casting to produce these systems. The proposed design method is explained by using continuous voids having a diamond shaped cross-sectional area. The resulting group of structures can be described as connected polygons and were found to exhibit a negative or zero Poisson’s ratio. The analysed systems were also found to have a strain independent Poisson’s ratio up to at least 7% strain. The proposed design method can thus facilitate the availability of three dimensional auxetic metamaterials in the consumer market which to date is conspicuous by their absence.

Bioprinting, Engineered living materials, Biosensors, Synthetic biology, Engineered bacteria, Medicine (General), R5-920, Biology (General), and QH301-705.5

The intertwined adoption of synthetic biology and 3D bioprinting has the potential to improve different application fields by fabricating engineered living materials (ELMs) with unnatural genetically-encoded sense & response capabilities. However, efforts are still needed to streamline the fabrication of sensing ELMs compatible with field use and improving their functional complexity. To investigate these two unmet needs, we adopted a workflow to reproducibly construct bacterial ELMs with synthetic biosensing circuits that provide red pigmentation as visible readout in response to different proof-of-concept chemical inducers. We first fabricated single-input/single-output ELMs and we demonstrated their robust performance in terms of longevity (cell viability and evolutionary stability >15 days, and long-term storage >1 month), sensing in harsh, non-sterile or nutrient-free conditions compatible with field use (soil, water, and clinical samples, including real samples from Pseudomonas aeruginosa infected patients). Then, we fabricated ELMs including multiple spatially-separated biosensor strains to engineer: level-bar materials detecting molecule concentration ranges, multi-input/multi-output devices with multiplexed sensing and information processing capabilities, and materials with cell-cell communication enabling on-demand pattern formation. Overall, we showed successful field use and multiplexed functioning of reproducibly fabricated ELMs, paving the way to a future automation of the prototyping process and boosting applications of such devices as in-situ monitoring tools or easy-to-use sensing kits.

Electrical engineering. Electronics. Nuclear engineering, TK1-9971, Computer software, and QA76.75-76.765

Establishing the indoor and outdoor humidity values in a greenhouse allows us to describe the crop yield during its entire developmental cycle. This study seeks to develop a predictive model of indoor relative humidity values in a greenhouse with high accuracy and interpretability through the use of optimized fuzzy inference systems, in order to offer greenhouse users a clear and simple description of their behaviour. The three-phase methodology applied made use of descriptive statistics techniques, correlation analysis, and prototyping paradigm for the iterative and incremental development of the predictive model, validated through error measurement. The research resulted in six models which define the behaviour of humidity as a result of temperature, CO2, and soil moisture, with percentages of effectiveness above 90%. The implementation of a Mamdani-type fuzzy inference system, optimized by a hybrid method combining genetic and interior point algorithms, allowed to predict the relative humidity in greenhouses with high interpretability and precision, with an effectiveness percentage of 90.97% and MSE (mean square error) of 8.2e − 3.