Advances In Hardware Technology Research Articles

Intelligent drone-assisted robust lightweight multi-factor authentication for military zone surveillance in the 6G era

In the diverse range of surveillance applications, large-scale deployment of next-generation communication technologies and the fast-growing development of unmanned aerial vehicles (UAVs) are envisioned as key innovations in the adoption of beyond-fifth generation (B5G) and 6G communication. Due to its self-reliance and versatility, a complex communication network can be formulated strategically to improve the application features of drone technology, including search-and-rescue, mission-critical services, and military surveillance. In recent times, technological advancements in hardware and software infrastructure have gained momentum toward seamless information interaction in aerial communication. Unfortunately, the recurrent process of user authentication causes severe communication instability in an unmanned aerial ad hoc network (UAANET) leading to some serious cyber threats, such as buffer overflow, denial of service, and spoofing. Therefore, building secure and reliable authentication is inevitable in order to protect drone-aided healthcare service environments. To protect aerial zones and improve security efficiency, this paper designs robust lightweight secure multi-factor authentication (RL-SMFA). The proposed RL-SMFA utilizes an AI-enabled, secure analytics phase to verify the genuineness of drone swarms for the ground control station. While protecting communication with drone vehicles, we also observe that power consumption by drones is reduced to a large extent. Using formal verification under a random oracle model, we show that the proposed RL-SMFA can functionally resist system vulnerabilities and constructively decrease the computation and communication costs of the UAANET. Lastly, the simulation study using ns3 shows that the proposed RL-SMFA achieves better performance efficiencies in terms of throughput rate, packet delivery ratio, and end-to-end delay than other state-of-the-art approaches to discovering a proper link establishment.

Deep Residual Separable Convolutional Neural Network for lung tumor segmentation

Lung cancer is one of the deadliest types of cancers. Computed Tomography (CT) is a widely used technique to detect tumors present inside the lungs. Delineation of such tumors is particularly essential for analysis and treatment purposes. With the advancement in hardware technologies, Machine Learning and Deep Learning methods are outperforming the traditional methods in the field of medical imaging. In order to delineate lung cancer tumors, we have proposed a deep learning-based methodology which includes a maximum intensity projection based pre-processing method, two novel deep learning networks and an ensemble strategy. The two proposed networks named Deep Residual Separable Convolutional Neural Network 1 and 2 (DRS-CNN1 and DRS-CNN2) achieved better performance over the state-of-the-art U-net network and other segmentation networks. For fair comparison, we have evaluated the performances of all networks on Medical Segmentation Decathlon (MSD) and StructSeg 2019 datasets. The DRS-CNN2 achieved a mean Dice Similarity Coefficient (DSC) of 0.649, mean 95 Hausdorff Distance (HD95) of 18.26, mean Sensitivity 0.737 and a mean Precision of 0.765 on independent test sets.

Deep Integration of Physical Health Education Based on Intelligent Communication Technology.

In recent years, intelligent medical communication technology has developed rapidly, and the advancement of hardware technology has accelerated the development of software technology. Physical exercise breaks the limits of all aspects of the human body and requires scientific and reliable information. Industrial cameras realize collection and high-resolution image processing functions and use artificial intelligence to create conditions for collecting scientific data. With the powerful driving force of artificial intelligence, it is believed that physical education will develop to a new height. The intellectualization of physical education makes people expect that it will also bring some inevitable problems, but as long as we continue to overcome and improve, we believe that artificial intelligence will better serve physical education and benefit mankind. Artificial intelligence is the current trend of social development. This paper takes physical education as a breakthrough point to discuss the application of artificial intelligence in physical education. This paper uses the grey GM prediction model and questionnaire survey method to build an online education platform and develop an effective education model, which can integrate the country's best educational resources and reduce time and space constraints. Providing a relaxing learning environment for education, one can learn anytime, anywhere. In a typical physical education system, the average self-efficacy score is converted from 20.48 points to 68.37 points, and the average self-score of physical education in the integrated artificial intelligence environment is converted from 30.25 points to 75.54 points. The data show that the quality of education has greatly improved.

Pediatric Cardiothoracic CT Guideline Provided by the Asian Society of Cardiovascular Imaging Congenital Heart Disease Study Group: Part 2. Contemporary Clinical Applications.

The use of pediatric cardiothoracic CT for congenital heart disease (CHD) was traditionally limited to the morphologic evaluation of the extracardiac thoracic vessels, lungs, and airways. Currently, the applications of CT have increased, owing to technological advancements in hardware and software as well as several dose-reduction measures. In the previously published part 1 of the guideline by the Asian Society of Cardiovascular Imaging Congenital Heart Disease Study Group, we reviewed the prerequisite technical knowledge for clinical applications in a user-friendly and vendor-specific manner. Herein, we present the second part of our guideline on contemporary clinical applications of pediatric cardiothoracic CT for CHD based on the consensus of experts from the Asian Society of Cardiovascular Imaging CHD Study Group. This guideline describes up-to-date clinical applications effectively in a systematic fashion.

Network Intrusion Detection using a Deep Learning Approach

At present situation network communication is at high risk for external and internal attacks due to large number of applications in various fields. The network traffic can be monitored to determine abnormality for software or hardware security mechanism in the network using Intrusion Detection System (IDS). As attackers always change their techniques of attack and find alternative attack methods, IDS must also evolve in response by adopting more sophisticated methods of detection .The huge growth in the data and the significant advances in computer hardware technologies resulted in the new studies existence in the deep learning field, including ID. Deep Learning (DL) is a subgroup of Machine Learning (ML) which is hinged on data description. The new model based on deep learning is presented in this research work to activate operation of IDS from modern networks. Model depicts combination of deep learning and machine learning, having capacity of wide range accurate analysis of traffic network. The new approach proposes non-symmetric deep auto encoder (NDAE) for learning the features in unsupervised manner. Furthermore, classification model is constructed using stacked NDAEs for classification. The performance is evaluated using a network intrusion detection analysis dataset, particularly the WSN Trace dataset. The contribution work is to implement advanced deep learning algorithm consists IDS use, which are efficient in taking instant measures in order to stop or minimize the malicious actions.

Artificial intelligence driven next-generation renal histomorphometry.

Successful integration of artificial intelligence into extant clinical workflows is contingent upon a number of factors including clinician comprehension and interpretation of computer vision. This article discusses how image analysis and machine learning have enabled comprehensive characterization of kidney morphology for development of automated diagnostic and prognostic renal pathology applications. The primordial digital pathology informatics work employed classical image analysis and machine learning to prognosticate renal disease. Although this classical approach demonstrated tremendous potential, subsequent advancements in hardware technology rendered artificial neural networks '(ANNs) the method of choice for machine vision in computational pathology'. Offering rapid and reproducible detection, characterization and classification of kidney morphology, ANNs have facilitated the development of diagnostic and prognostic applications. In addition, modern machine learning with ANNs has revealed novel biomarkers in kidney disease, demonstrating the potential for machine vision to elucidate novel pathologic mechanisms beyond extant clinical knowledge. Despite the revolutionary developments potentiated by modern machine learning, several challenges remain, including data quality control and curation, image annotation and ontology, integration of multimodal data and interpretation of machine vision or 'opening the black box'. Resolution of these challenges will not only revolutionize diagnostic pathology but also pave the way for precision medicine and integration of artificial intelligence in the process of care.

Single-particle cryo-electron microscopy: Mathematical theory, computational challenges, and opportunities.

In recent years, an abundance of new molecular structures have been elucidated using cryo-electron microscopy (cryo-EM), largely due to advances in hardware technology and data processing techniques. Owing to these new exciting developments, cryo-EM was selected by Nature Methods as Method of the Year 2015, and the Nobel Prize in Chemistry 2017 was awarded to three pioneers in the field. The main goal of this article is to introduce the challenging and exciting computational tasks involved in reconstructing 3-D molecular structures by cryo-EM. Determining molecular structures requires a wide range of computational tools in a variety of fields, including signal processing, estimation and detection theory, high-dimensional statistics, convex and non-convex optimization, spectral algorithms, dimensionality reduction, and machine learning. The tools from these fields must be adapted to work under exceptionally challenging conditions, including extreme noise levels, the presence of missing data, and massively large datasets as large as several Terabytes. In addition, we present two statistical models: multi-reference alignment and multi-target detection, that abstract away much of the intricacies of cryo-EM, while retaining some of its essential features. Based on these abstractions, we discuss some recent intriguing results in the mathematical theory of cryo-EM, and delineate relations with group theory, invariant theory, and information theory.

Intermediate Deep Feature Compression: Toward Intelligent Sensing.

The recent advances of hardware technology have made the intelligent analysis equipped at the front-end with deep learning more prevailing and practical. To better enable the intelligent sensing at the front-end, instead of compressing and transmitting visual signals or the ultimately utilized top-layer deep learning features, we propose to compactly represent and convey the intermediate-layer deep learning features with high generalization capability, to facilitate the collaborating approach between front and cloud ends. This strategy enables a good balance among the computational load, transmission load and the generalization ability for cloud servers when deploying the deep neural networks for large scale cloud based visual analysis. Moreover, the presented strategy also makes the standardization of deep feature coding more feasible and promising, as a series of tasks can simultaneously benefit from the transmitted intermediate layer features. We also present the results for evaluations of both lossless and lossy deep feature compression, which provide meaningful investigations and baselines for future research and standardization activities.

Survey on intrusion detection systems based on deep learning

Intrusion Detection Systems (IDSs) have a significant role in all networks and information systems in the world to earn the required security guarantee. IDS is one of the solutions used to reduce malicious attacks. As attackers always changing their techniques of attack and find alternative attack methods, IDS must also evolve in response by adopting more sophisticated methods of detection. The huge growth in the data and the significant advances in computer hardware technologies resulted in the new studies existence in the deep learning field, including intrusion detection. Deep learning is sub-field of Machine Learning (ML) methods that are based on learning data representations. In this paper, a detailed survey of various deep learning methods applied in IDSs is given first. Then, a deep learning classification scheme is presented and the main works that have been reported in the deep learning works is summarized. Utilizing this approach, we have provided a taxonomy survey on the available deep architectures and algorithms in these works and classify those algorithms to three classes, which are: discriminative, hybrid and generative. After that, chosen deep learning applications are reviewed in a wide range of fields of intrusion detection. Finally, popular types of datasets and frameworks are discussed.

Can AI Help Make Screening Mammography "Lean"?

Can AI Help Make Screening Mammography "Lean"?

Experimental evaluation of reconstruction algorithms for scintillation crystal array based on charge projection readout

Abstract Although the technological advances in hardware of gamma camera lead to an obvious increase of detector performance, the final reconstructed image is still strongly restricted by reconstruction algorithms. In this work, an experiment evaluation of four popular algorithms, i.e., truncated Center-of-Gravity (TCOG), Raise-to-Power (RTP), Least Squares Estimator based on Particle Swarm Optimization (PSO-LSE) and Gaussian fitting (GF), for estimating the position of scintillation events detected by scintillation crystal array based on charge projection readout is performed. The detector of the gamma camera consists of a 22 × 22 array of CsI(Tl) scintillation crystals and an 8 × 8 matrix of SiPM pixels. The readout circuit, known as symmetric charge division (SCD) circuit, reduces an 8 × 8 array of signal channels into 8+8 signal channels coming from rows and columns projections. The imaging performances, in terms of flood image, average peak-to-valley ratio of line profile, useful field-of-view (UFOV) and position linearity response are measured. The result indicates that the four presented algorithms show good applicability for charge projection readout. RTP and PSO-LSE algorithms perform better in pixel identification of flood image and peak-to-valley of line profile, and PSO-LSE and GF methods are helpful to increase the UFOV. Among the four implemented reconstruction algorithms, PSO-LSE is a potential and promising algorithm that allows achieving excellent pixel identification of flood image, peak-to-valley ratio of line profile, spatial resolution, UFOV and position linearity response at the same time. The application of the PSO-LSE algorithm may offer the possibility to obtain high-quality reconstructed image in scintillation crystal array based on charge projection readout.

Magnetic Resonance Spectroscopy in ALS.

Proton magnetic resonance spectroscopy (MRS) provides a means of measuring cerebral metabolites relevant to neurodegeneration in vivo. In amyotrophic lateral sclerosis (ALS), neurochemical changes reflecting neuronal loss or dysfunction (decreased N-actylaspartate [NAA]) is most significant in the motor cortex and corticospinal tracts. Other neurochemical changes observed include increased myo-inositol (mIns), a putative marker of gliosis. MRS confirmation of involvement of non-motor regions such as the frontal lobes, thalamus, basal ganglia, and cingulum are consistent with the multi-system facet of motor neuron disease with ALS being part of a MND-FTD spectrum. MRS-derived markers exhibit an encouraging discriminatory ability to identify patients from healthy controls, however more data is needed to determine its ability to assist with the diagnosis in early stages when upper motor neuron signs are limited, and in distinguishing from disease mimics. Longitudinal change of NAA and mIns do not appear to be reliable in monitoring disease progression. Technological advances in hardware and high field scanning are increasing the number of accessible metabolites available for interrogation.

Measuring molecular interactions in solution using multi-wavelength analytical ultracentrifugation: combining spectral analysis with hydrodynamics.

In 1926, the Swedish scientist Theodor Svedberg was awarded the Nobel Prize in Chemistry for his work on a disperse system, and for studying the colloidal properties of proteins. This work was, to a large extent, made possible by his invention of a revolutionary tool, the analytical ultracentrifuge. These days, technological advances in hardware and computing have transformed the field of analytical ultracentrifugation (AUC) by enabling entirely new classes of experiments and modes of measurement unimaginable by Svedberg, making AUC once again an indispensable tool for modern biomedical research. In this article these advances and their impact on studies of interacting molecules will be discussed, with particular emphasis on a new method termed multi-wavelength analytical ultracentrifugation (MWL-AUC). Novel detectors allow us to add a second dimension to the separation of disperse and heterogeneous systems: in addition to the traditional hydrodynamic separation of colloidal mixtures, it is now possible to identify the sedimenting molecules by their spectral absorbance properties. The potential for this advance is significant for the study of a large range of systems. A further advance has occurred in data management and computational capabilities, opening doors to improved analysis methods, as well as direct networking with the instrument, facilitating data acquisition and data handling, and significant increases in data density from faster detectors with higher resolution capability.

LocalRec 2018 workshop report the second ACM SIGSPATIAL workshop on recommendations for location-based services and social networks * Seattle, Washington, USA - November 6, 2018

Driven by technological advances in hardware (positioning systems, environmental sensors), software (standards, tools, network services), and aided by various open movements (open, linked, government data) and the ever-growing mentality of sharing for the greater good (crowdsourcing, crowdfunding, collaborative and volunteered geographic information), the amount of available geo-referenced data has seen dramatic explosion over the past few years. Human activities generate data and traces that are now often transparently annotated with location and contextual information. At the same time, it has become easier than ever to collect and combine rich and diverse information about locations. Exploiting this torrent of geo-referenced data provides a tremendous potential to materially improve existing and offer novel types of recommendation services, with clear benefits in many domains, including social networks, marketing, and tourism.

A Secured and Efficient Communication Scheme for Decentralized Cognitive Radio-Based Internet of Vehicles

The advancements in hardware technologies have driven the evolution of vehicular ad hoc networks into the Internet of Vehicles (IoV). The IoV is a decentralized network of IoT-enabled vehicles capable of smooth traffic flow to perform fleet management and accident avoidance. The IoV has many commercial applications due to improved security and safety on the roads. However, the rapidly increasing number of wireless applications have challenged the existing spectrum bands allocated to IoV. The IoV has only six communication channels that are congested during the peak hours. The limited number of channels and the presence of congestion on these channels are the challenging issues that affect the safety of vehicles on the road. To mitigate the congestion, Cognitive Radio (CR) can be an optimal solution for the existing IoV Paradigm. In this paper, we propose a secured and efficient communication scheme for a decentralized CR-based IoV (CIoV) network. In this scheme, the Roadside Unit (RSU) senses the spectrum using an energy detection method. Each vehicle independently predicts the Primary User (PU) activity pattern using a hidden Markov model (HMM). Once a vehicle detects a licensed channel free from the PUs, it informs the RSU to store the channel in a database alongside the dedicated direct short-range communication (DSRC) channels for data transmission. The RSU and vehicles are registered with a trusted authority and they mutually authenticate each other. Upon mutual authentication, the RSU assigns communication channels to the vehicles on the road, based on their density. When the density of the vehicles is high, the detected licensed channels are used, otherwise, the DSRC channels are used. We evaluate the performance of CIoV in terms of packet delivery and packet loss ratio, end-to-end delay, and throughput, using NS-2. The simulation results show that the CR-based approach of CIoV outperforms the existing schemes and significantly enhances the performance of the underlying network.

Nuclear magnetic resonance structure-based drug design.

Nuclear magnetic resonance structure-based drug design.

Validity of Consumer Activity Wristbands and Wearable EEG for Measuring Overall Sleep Parameters and Sleep Structure in Free-Living Conditions.

Consumer sleep tracking technologies offer an unobtrusive and cost-efficient way to monitor sleep in free-living conditions. Technological advances in hardware and software have significantly improved the functionality of the new gadgets that recently appeared in the market. However, whether the latest gadgets can provide valid measurements on overall sleep parameters and sleep structure such as deep and REM sleep has not been examined. In this study, we aimed to investigate the validity of the latest consumer sleep tracking devices including an activity wristband Fitbit Charge 2 and a wearable EEG-based eye mask Neuroon in comparison to a medical sleep monitor. First, we confirmed that Fitbit Charge 2 can automatically detect the onset and offset of sleep with reasonable accuracy. Second, analysis found that both consumer devices produced comparable results in measuring total sleep duration and sleep efficiency compared to the medical device. In addition, Fitbit accurately measured the number of awakenings, while Neuroon with good signal quality had satisfactory performance on total awake time and sleep onset latency. However, measuring sleep structure including light, deep, and REM sleep remains to be challenging for both consumer devices. Third, greater discrepancies were observed between Neuroon and the medical device in nights with more disrupted sleep and when the signal quality was poor, but no trend was observed in Fitbit Charge 2. This study suggests that current consumer sleep tracking technologies may be immature for diagnosing sleep disorders, but they are reasonably satisfactory for general purpose and non-clinical use.

Localrec 2017 workshop report: the First ACM SIGSPATIAL Workshop on Recommendations for Location-Based Services and Social Networks

Driven by technological advances in hardware (positioning systems, environmental sensors), software (standards, tools, network services), and aided by various open movements (open, linked, government data) and the ever-growing mentality of sharing for the greater good (crowdsourcing, crowdfunding, collaborative and volunteered geographic information), the amount of available geo-referenced data has seen dramatic explosion over the past few years. Human activities generate data and traces that are now often transparently annotated with location and contextual information. At the same time, it has become easier than ever to collect and combine rich and diverse information about locations. Exploiting this torrent of geo-referenced data provides a tremendous potential to materially improve existing and offer novel types of recommendation services, with clear benefits in many domains, including social networks, marketing, and tourism.

Multimodal nanoparticle imaging agents: design and applications.

Molecular imaging, where the location of molecules or nanoscale constructs can be tracked in the body to report on disease or biochemical processes, is rapidly expanding to include combined modality or multimodal imaging. No single imaging technique can offer the optimum combination of properties (e.g. resolution, sensitivity, cost, availability). The rapid technological advances in hardware to scan patients, and software to process and fuse images, are pushing the boundaries of novel medical imaging approaches, and hand-in-hand with this is the requirement for advanced and specific multimodal imaging agents. These agents can be detected using a selection from radioisotope, magnetic resonance and optical imaging, among others. Nanoparticles offer great scope in this area as they lend themselves, via facile modification procedures, to act as multifunctional constructs. They have relevance as therapeutics and drug delivery agents that can be tracked by molecular imaging techniques with the particular development of applications in optically guided surgery and as radiosensitizers. There has been a huge amount of research work to produce nanoconstructs for imaging, and the parameters for successful clinical translation and validation of therapeutic applications are now becoming much better understood. It is an exciting time of progress for these agents as their potential is closer to being realized with translation into the clinic. The coming 5-10 years will be critical, as we will see if the predicted improvement in clinical outcomes becomes a reality. Some of the latest advances in combination modality agents are selected and the progression pathway to clinical trials analysed.This article is part of the themed issue 'Challenges for chemistry in molecular imaging'.

Trends in cochlear implant candidacy in children

Abstract In the last 30 years, the field of cochlear implantation has rapidly evolved. Technological advances in hardware and corresponding developments in surgical techniques, along with new sound processing strategies and innovative rehabilitation, have combined to maximise functional outcomes. Informed by the evidence base of such outcomes, the inclusion criteria for children who might benefit from a cochlear implant (CI) has been refined and expanded. CIs are now the standard of care for children with severe to profound hearing loss where the desired outcome is spoken language. There are several emerging trends within the field of paediatric CIs that have already translated into clinical practice in some countries, but have not yet been universally adopted. These include, but are not limited to, the expansion of audiometric CI candidacy criteria for cohorts with: (1) more residual hearing; (2) partial hearing which may benefit from electric-acoustic stimulation (EAS); and (3) asymmetric hearing levels. This review will describe the historical context and provide an overview of the candidacy trends as they relate to children.