Background and aims: Chest X-rays are widely used, non-invasive, cost effective imaging tests. However, the complexity of interpretation and global shortage of radiologists have led to reporting backlogs, delayed diagnosis and a compromised quality of care. A fully automated, reliable artificial intelligence system that can quickly triage abnormal images for urgent radiologist review would be invaluable in the clinical setting. The aim was to develop and validate a deep learning Convoluted Neural Network algorithm to automate the detection of 13 common abnormalities found on Chest X-rays. Method: In this retrospective study, a VGG 16 deep learning model was trained on images from the Chest-ray 14, a large publicly available Chest X-ray dataset, containing over 112,120 images with annotations. Images were split into training, validation and testing sets and trained to identify 13 specific abnormalities. The primary performance measures were accuracy and precision. Results: The model demonstrated an overall accuracy of 88% in the identification of abnormal X-rays and 87% in the detection of 13 common chest conditions with no model bias. Conclusion: This study demonstrates that a well-trained deep learning algorithm can accurately identify multiple abnormalities on X-ray images. As such models get further refined, they can be used to ease radiology workflow bottlenecks and improve reporting efficiency. Napier Healthcare’s team that developed this model consists of medical IT professionals who specialize in AI and its practical application in acute & long-term care settings. This is currently being piloted in a few hospitals and diagnostic labs on a commercial basis.

Neuromuscular disorders in Cerebral Palsy (CP) patients lead to foot deformities and affect foot biomechanics leading to compromised gait. Thus, measurement of the foot kinematic measurement is of particular interest to understand and characterize the walking pattern among CP patients. The objective of the present work is to develop a wearable instrument to measure foot kinematics such as foot-to-ground angle in three-dimensional planes and to measure the foot clearance i.e., toe and heel clearances. A template-based outsole was developed that incorporated an optical distance sensor located anatomically on the outsole and the magnetometer to measure the foot kinematics. The developed system was validated against the reference marker-based motion capture system (from Noraxon). The data from eight able-bodied participants were acquired simultaneously from both the systems (developed and the reference system) at three different walking speeds. A CoP based feedback was presented to the participants to shift the sagittal CoP anteriorly, posteriorly and normal to simulate the walking pattern of CP patients with three different foot landing strategies. Pearson's correlation coefficient of more than or equal to 0.62, root mean square error of less than or equal to 7.81 degrees and limit of agreement of more than or equal to 95% is found. Furthermore, a wireless wristband is developed and validated for real-time vibrotactile feedback. The measurement accuracy reported with outsole while participants simulated CP gait shows the potential of present work in real-time foot kinematics detection in CP patients. The instrumentation is wearable, low-cost, easy to use and implement.

In the process of exploring the essence of light, Newton initially agreed with the particle interpretation of light while Huygens argued for the wave theory. Hence, these two theories had been disputed in Newton’s time. In the beginning people accepted the particle theory, but after Thomas Young’s experiment and Augustin Jean Fresnel’s experiment, people began to accept the wave theory. Until Einstein proposed the quanta concept, which was later called photon, and, even later, De Broglie proposed the wave nature of matter, subsequently, people began using particle-wave duality to explicate all phenomena in micro world. Thus here appears a paradox: how can one particle exist in two forms? To solve this enigma, I have done some experiments; discover that moving photons create force; this effect reveal the phenomenon of light wave property - the inference fringes is caused by force which moving photons produced. The essence of light is particle but not particle-wave duality.

We present an application of fully anisotropic hp-adaptivity over quadrilateral meshes for H(curl)-conforming discretizations in Computational Electromagnetics (CEM). Traditionally, anisotropic h-adaptivity has been difficult to implement under the constraints of the Continuous Galerkin Formulation; however, Refinement-by-Superposition (RBS) facilitates anisotropic mesh adaptivity with great ease. We present a general discussion of the theoretical considerations involved with implementing fully anisotropic hp-refinement, as well as an in-depth discussion of the practical considerations for 2-D FEM. Moreover, to demonstrate the benefits of both anisotropic h- and p-refinement, we study the 2-D Maxwell eigenvalue problem as a test case. The numerical results indicate that fully anisotropic refinement can provide significant gains in efficiency, even in the presence of singular behavior, substantially reducing the number of degrees of freedom required for the same accuracy with isotropic hp-refinement. This serves to bolster the relevance of RBS and full hp-adaptivity to a wide array of academic and industrial applications in CEM

This paper proposes a new estimation and compensation approach to mitigate several linear and widely linear effects in coherent optical systems using digital signal processing (DSP) algorithms. Compared to most of the available strategies that employ local estimation and/or compensation algorithms, this approach performs a global impairments estimation and compensation based on Nonlinear Least Squares. The proposed method estimates and compensates for the chromatic dispersion (CD), carrier frequency offset (CFO), in-phase/quadrature (IQ) imbalance, and laser phase noise (PN) in two steps. Firstly, it estimates the quasi-static parameters related to the CD, CFO, and both transmitter and receiver IQ imbalance. Secondly, it estimates both transmitter and receiver lasers’ phases and compensates for all the imperfections by using a Zero-Forcing (ZF) equalizer. Simulations show the effectiveness of the approach in terms of statistical performance and computational time. The estimation performance is assessed by computing the Cramér Rao Lower Bound (CRLB), while the detection performance is compared to a modified Clairvoyant equalizer.

Artificial intelligence has proven a matching technology to support multiple business system applications. This technology has been in use to reshape current business models and to provide innovative management strategies. Growing business competition among major players is also strengthening its applications. The business information system with use of artificial intelligence raises the competitiveness of enterprises in the global market. The fast pace of merging artificial intelligence and automation are propelling strategists to revise business models and to explore new possibilities to meet the customer expectations. This paper focuses on the impact of artificial intelligence on business systems. The article also presents an overview of influential academic achievements and innovations in the field of artificial intelligence and their solutions for entrepreneurial activities. In the end, we intend to discuss important points spurred around us in today's scenario about the challenges and future of artificial intelligence.

Gait disorders in children with cerebral palsy (CP) affect their mental, physical, economic, and social lives. Gait assessment is one of the essential steps of gait management. It has been widely used for clinical decision making and evaluation of different treatment outcomes. However, most of the present methods of gait assessment are subjective, less sensitive to small pathological changes, time-taking and need a great effort of an expert. This work proposes an automated, comprehensive gait assessment score (A-GAS) for gait disorders in CP. Kinematic data of 356 CP and 41 typically developing subjects is used to validate the performance of A-GAS. For the computation of A-GAS, instance abnormality index (AII) and abnormality index (AI) are calculated. AII quantifies gait abnormality of a gait cycle instance, while AI quantifies gait abnormality of a joint angle profile during walking. AII is calculated for all gait cycle instances by performing probabilistic and statistical analyses. Abnormality index (AI) is a weighted sum of AII, computed for each joint angle profile. A-GAS is a weighted sum of AI, calculated for a lower limb. Moreover, a graphical representation of the gait assessment report, including AII, AI, and A-GAS is generated for providing a better depiction of the assessment score. Furthermore, the work compares A-GAS with a present rating-based gait assessment scores to understand fundamental differences. Finally, A-GAS's performance is verified for a high-cost multi-camera set-up using nine joint angle profiles and a low-cost single camera set-up using three joint angle profiles. Results show no significant differences in performance of A-GAS for both the set-ups. Therefore, A-GAS for both the set-ups can be used interchangeably.

A static metasurface reflector based on a novel coupled resonator configuration is proposed to independently control the reflection phase and magnitude of linearly polarized incident fields, and is demonstrated experimentally in the millimeter-wave Ka-band around 30 GHz. The proposed concept is illustrated using a unit cell design consisting of a rectangular ring coupled with a rectangular slot resonator backed by a grounded dielectric slab. By geometrically tuning various dimensions of the two resonators, a near-perfect amplitude-phase coverage is achieved at a fixed design frequency of 30 GHz. To demonstrate the flexible beam-forming capability of the proposed metasurface reflectors, illustrative examples of fixed beam steering with varying reflection magnitudes, and asymmetric dual-beam patterns with specified reflection magnitude, reflection angles and beam-widths, are successfully shown. Compared to the standard method based on polarization rotation and resistive loadings with discrete values, the proposed technique does not generate undesired cross-polarization field reflection, and provides a continuous magnitude tuning including full absorption, along with wide phase coverage.

This paper demonstrates the design and implementation of an inductorless analog band-pass filter (BPF). Band-pass filters are widely used in communication systems, wireless transceivers and audio systems, they only pass signals within a desired frequency range. The principles mentioned in this article can be generalized to design any analog filter regardless of its order, approximation and prototype. The design procedure can be broken down into three main parts, first of all, a passive low-pass filter (LPF) is implemented, then the passive LPF is converted into a passive BPF. Finally, the passive BPF is transformed into an active BPF by adding operational amplifiers. The active BPF is then modified into two different topologies, the first in which the inductors are replaced with simulated- inductors (gyrators), while in the second topology, less operational amplifiers are used.

In this letter, we study the temporal and spatial variability of ionospheric irregularities by generating high-resolution maps of the observed amplitude scintillation index (S4) using data from a multi-constellation and multi-frequency GNSS receiver. The study is located in the Arabian Peninsula region, which falls under the northern crest of the equatorial ionization anomaly (EIA). Even though the study was conducted during a solar minimum period, considerable pre-sunset scintillation occurrences have been observed between 15-17 local time, particularly during the winter solstices. While most scintillation occurrences have been observed at low elevation (20 to 25 degrees), a considerable number of scintillation causing ionospheric irregularities have been observed towards the north, east, and southeast of the receiver location, for elevation ranging from 40 to 60 degrees. Out of all the GNSS constellations with medium-earth-orbit (MEO) satellites, GPS was the most impacted by amplitude scintillation, while BeiDou and Galileo satellites were the least affected. It is anticipated that the patches of ionospheric irregularities reported in this work will be further enhanced as solar activity increases in the coming years. Therefore, this work can serve as a reference for future studies during periods of increased geomagnetic activity.

This paper presents the RADNN algorithm. The RADNN is a robust to imperceptible adversarial attack algorithm that uses the concept of data density and similarities to detect attacks on real-time. Differently from traditional deep learnings that need be trained on the attacks to be able to detect, RADNN has a mechanism that detects data patterns changes. In order to evaluate the proposed method, we considered the PerC attacks and a 1000 images from the Imagenet dataset. The RADNN could correctly identify 97.2% of the attacks.

The advantage of computational resources in edge computing near the data source has kindled growing interest in delay-sensitive Internet of Things (IoT) applications. However, the benefit of the edge server is limited by the uploading and downloading links between end-users and edge servers when these end-users seek computational resources from edge servers. The scenario becomes more severe when the user-end’s devices are in the shaded region resulting in low uplink/downlink quality. In this paper, we consider a reconfigurable intelligent surface (RIS)-assisted edge computing system, where the benefits of RIS are exploited to improve the uploading transmission rate. We further aim to minimize the delay of worst-case in the network when the end-users either compute task data in their local CPU or offload task data to the edge server. Next, we optimize the uploading bandwidth allocation for every end-user’s task data to minimize the maximum delay in the network. The above optimization problem is formulated as quadratically constrained quadratic programming. Afterward, we solve this problem by semidefinite relaxation. Finally, the simulation results demonstrate that the proposed strategy is scalable under various network settings.

Synchronization with other machines is one of those tasks that an intelligent machine should be able to perform. To do so, a general method of synchronization must be defined and that is the ambition of this article. For this purpose, we recall what the main concepts of systemic modelling consist of. Then we define what a Synchronization Problem is and distinguish three types of synchronization problems : predetermined (PSP), stochastic (SSP) and asymptotic (ASP). So, to find the best solution, we define three optimized synchronization problems : Optimized Predetermined Synchronization Problems , Optimized Stochastic Synchronization Problems (OSSP) and Optimized Asymptotic Synchronization Problems (OASP). A general method for solving an optimized synchronization problem and thus synchronize an out of sync system, is then given; it includes four steps : step 1, identify the synchronization functions; step 2, identify the synchronization case : predetermined, stochastic or asymptotic; step 3, select an optimization criterion and solve the optimized synchronization problem corresponding to the chosen synchronization case; step 4, find a control which permits to track the system trajectory, optimal solution of the selected optimized synchronization problem. In the next section, we present a heuristic algorithm that checks the constraints of an ASP for any initial state of the system to be synchronized. The principle of this algorithm is to pursue or track a perfectly synchronized solution. Finally, we present and solve two synchronization problems in the field of Mobile Robot Systems: a synchronized satelliteization problem and the Horse Carousel Problem. We apply to these two problems the general method given above by choosing the asymptotic synchronization case and our tracking algorithm. Simulation results show the efficiency of the method and the relevance of using a tracking algorithm.

This paper presents simple techniques to significantly reduced energy consumption of DNNs: Operating at reduced voltages offers substantial energy efficiency improvement but at the expense of increasing the probability of computational errors due to hardware faults. In this context, we targeted Deep Neural Networks (DNN) as emerging energy hungry building blocks in embedded applications. Without an error feedback mechanism, blind voltage down-scaling will result in degraded accuracy or total system failure. To enable safe voltage down-scaling, in this paper two solutions based on Self-Supervised Learning (SSL) and Algorithm Based Fault Tolerance (ABFT) were developed. A DNN model trained on MNIST data-set was deployed on a Field Programmable Gate Array (FPGA) that operated at reduced voltages and employed the proposed schemes. The SSL approach provides extremely low-overhead (≈0.2%) fault detection at the cost of lower error coverage and extra training, while ABFT incurs less than 8%overheads at run-time with close to 100% error detection rate. By using the solutions, substantial energy savings, i.e., up to 40.3%,without compromising the accuracy of the model was achieved

The In-Memory Computing (IMC) architecture based on 6T, 8T, 10T SRAM fails under process-variation and suffers from compute-disturb, compute-failure, half-select issue, respectively, which affect the reliability of IMC operation. To overcome these problems, local bit-line sharing Dual-Port 8T (SDP8T) SRAM with Virtual VSS is proposed to improve the stability and energy efficiency of IMC architecture. The decouple read-write path with high-Vth transistor is used to improve the read-margin by 2.11× and reduce the read-energy by 36.35% as compared to Transpose-8T SRAM. The virtual VSS write assist is used in SDP8T SRAM to improve the write-margin by 26.49%, and lower the leakage power by 47.95% as compared to Transpose-8T SRAM. Furthermore, IMC architecture is proposed using SDP8T SRAM. In addition to the SRAM function, SDP8T-IMC architecture performs In-memory Boolean computation(IMBC) operations without compute-disturbance and compute-failure. The remarkable feature of SDP8T-IMC architecture is that it performs IMBC operation on four operands simultaneously using all four bit-line ports in a single cycle, thus doubling the throughput and obtain 11.04 fJ/bit average energy consumption at 1 V supply. The maximum operating frequency of the proposed IMC architecture is 1050 MHz at 1 V. Cumulatively, the proposed SDP8T-IMC architecture has 32.22%, 27.03%, 60.10%, 50.93%, 60.48%, 35.05%, and 65.28% reduction in energy consumption as compared to C6T, 6TCSRAM, 8+T, 8T, 10T, 12T, and 4+2T SRAM based IMC architectures, respectively. Moreover, the proposed IMC architecture is configured as Binary Content Addressable Memory (BCAM) for searching applications which achieves 0.60fJ energy consumption per search/bit at 1 V.

COVID-19 has caused hundreds of millions of infections and hundreds of deaths, and even though vaccinations are increasing, the mutation of the virus makes the pandemic even difficult to control. Existing manual, operator and Bluetooth-based technologies for epidemiological investigation and close contact tracing suffer from high cost, low accuracy, and difficulty in scaling up. Viruses such as Delta variants have a greater ability to survive and spread, making many of the existing human-human close contacts tracing less effective. Also, it is easy to overlook the fact that there is still a large segment of the world's population that does not have access to the Internet and is proficient in using smartphones, which makes the performance of smart device-based tracing much less effective. Inspired by Health Code and Tracetogether, which have been widely accepted in China and Singapore, we propose a LoRa and blockchain-based contact tracing method LBTrace, which is low-power, lightweight, and operation-free. The experimental results demonstrate the high stability and accuracy of our proposed method, which can be used as a complement to existing methods to help some governments effectively control COVID-19 and future outbreaks under certain emergency conditions.

The growing complexity of wireless networks has sparked an upsurge in the use of artificial intelligence (AI) within the telecommunication industry in recent years. In network slicing, a key component of 5G that enables network operators to lease their resources to third-party tenants, AI models may be employed in complex tasks, such as short-term resource reservation (STRR). When AI is used to make complex resource management decisions with financial and service quality implications, it is important that these decisions be understood by a human-in-the-loop. In this paper, we apply state-of-the art techniques from the field of Explainable AI (XAI) to the problem of STRR. Using real-world data to develop an AI model for STRR, we demonstrate how our XAI methodology can be used to explain the real-time decisions of the model, to reveal trends about the model’s general behaviour, as well as aid in the diagnosis of potential faults during the model’s development. In addition, we quantitatively validate the faithfulness of the explanations across an extensive range of XAI metrics to ensure they remain trustworthy and actionable.

Traditionally, speech emotion recognition (SER) research has relied on manually handcrafted acoustic features using feature engineering. However, the design of handcrafted features for complex SER tasks requires significant manual effort, which impedes generalisability and slows the pace of innovation. This has motivated the adoption of representation learning techniques that can automatically learn an intermediate representation of the input signal without any manual feature engineering. Representation learning has led to improved SER performance and enabled rapid innovation. Its effectiveness has further increased with advances in deep learning (DL), which has facilitated deep representation learning where hierarchical representations are automatically learned in a data-driven manner. This paper presents the first comprehensive survey on the important topic of deep representation learning for SER. We highlight various techniques, related challenges and identify important future areas of research. Our survey bridges the gap in the literature since existing surveys either focus on SER with hand-engineered features or representation learning in the general setting without focusing on SER.