Craniomaxillofacial (CMF) fractures, often resulting from traffic accidents, falls, and head traumas, necessitate prompt diagnosis and analysis with CT images. Our study leverages a segmentation model named 3D Swin UNETR to develop an automated detection system for these fractures. The key finding of this study is the significant improvement in the quality of CMF fracture detection achieved by incorporating an additional input channel containing labels of skull regions, using an additional loss function named Proximity loss, and performing an ensemble inference approach using different models trained by different settings. Clinical evaluations were manually performed by experts where the best-performing model achieved the positive predictive value (PPV) of 82.49%, true positive rate (TPR) of 96.03%, false detection rate (FDR) of 17.51%, false negative rate (FNR) of 3.97%, and F1-score (F1) of 88.23%.

This paper presents design techniques for an energyefficient multi-lane receiver (RX) with baud-rate clock and data recovery (CDR), which is essential for high-throughput lowlatency communication in high-performance computing systems. The proposed low-power global clock distribution not only significantly reduces power consumption across multi-lane RXs but is capable of compensating for the frequency offset without any phase interpolators. To this end, a fractional divider controlled by CDR is placed close to the global phase locked loop. Moreover, in order to address the sub-optimal lock point of conventional baudrate phase detectors, the proposed CDR employs a background eye-climbing algorithm, which optimizes the sampling phase and maximizes the vertical eye margin (VEM). Fabricated in a 28 nm CMOS process, the proposed 4×32 Gb/s RX shows a low integrated fractional spur of-40.4 dBc at a 2500 ppm frequency offset. Furthermore, it improves bit-error-rate performance by increasing the VEM by 17 %. The entire RX achieves the energy efficiency of 1.8 pJ/bit with the aggregate data rate of 128 Gb/s.

With the Internet of Things (IoT) generating vast amounts of data, privacy breaches have become increasingly prevalent, exposing individuals to serious risks such as identity theft and life-threatening situations. This research addresses the challenge of identifying cybersecurity threats and vulnerabilities leading to privacy breaches, as evidenced by recent cyber-attacks on Australian Medibank, Optus, and hospital networks. We propose a machine learning (ML)-based approach to distinguish between legitimate and rogue privacy policies, defining fundamental concepts of privacy, security, and access control in the context of personal, confidential, and sensitive information breaches. Our methodology introduces zero-privacy (ZP) and binary question-answer (QA) models to discern legitimate versus illegitimate actions or interests within privacy policies. Our experiments utilise natural language processing (NLP)-based ML models to analyse the linguistics of privacy policies. In experiments conducted on a dataset from the top 100 Forbes-listed companies, including 67 rogue policies, our privacy classification approach demonstrates reliability, accurately distinguishing between legitimate and rogue policies. With a dataset split of 90% for training and 10% for testing, our model achieves accuracy and precision exceeding 94% and 91%, respectively. Additionally, we evaluate the probability of ZP occurrences in organisations’ privacy and service-level agreements, revealing significant privacy breach risks. Through case studies utilising our proposed binary QA model, we underscore the urgent need for enhanced privacy measures across various organisations’ policies. Introducing a novel approach to access control, we specify permissions under conditions of legitimate and rogue privacy policies, exemplifying the applicability of our proposed access control mechanism through security policy modelling.

Privacy breaches have become increasingly prevalent, exposing individuals to significant risks. These breaches can have far-reaching consequences, including identity theft and life-threatening situations. Several studies have analysed data and privacy breaches and presented detection or prevention techniques to combat these breaches. However, because the number and type of breaches have significantly increased, these studies have become less relevant or outdated. Previous research on data/privacy breaches compared the techniques and results of various studies, but none attempted to comprehensively analyse the type of information and the type and level of compromise that occurred after such breaches. In this survey, we examine the fundamental concepts of privacy and security and define the security incidents and data/privacy breaches. We propose a set of criteria to evaluate the published studies on privacy breaches. We thoroughly investigate the problem domains and security-related concerns considering six recent breach cases in Australia, elucidating the critical challenges and issues associated with privacy breaches. We comprehensively review and outline the trends of security incidents and data/privacy breaches from 2020 to 2023. Additionally, we review the current state-of-the-art countermeasures to safeguard against these breaches. Finally, we identify an open research direction to develop an artificial intelligence (AI)-powered security framework that can help analyse cyber threats, characterise attackers' behaviours, distinguish between legitimate and illegitimate privacy policies, and restrict access to individuals' information. Overall, this survey will help organisations to reassess and update their security and privacy measures.

Human gait parameters reveal a lot about physical and psychological well-being. In addition, gait impairments significantly affect daily life activities and hamper the locomotive freedom of people with neurological or musculoskeletal disorders. However, there is still a need for a portable, user-friendly, costeffective gait characterization device. Therefore, in this study, a feature engineering-based portable gait characterization module is proposed, and a shank-mounted inertial measurement unit (IMU) is utilized for gait phases and event detection. The efficiency of the developed module is estimated on ten healthy subjects for plain terrain walking. A force sensing resistor (FSR) sensorized instrumented insole has been utilised as a reference system to validate the results estimated using the developed module. The performance is estimated with three different classifiers, support vector machine (SVM), K-nearest neighbor (KNN), and linear discriminant analysis (LDA). For gait event identifications, the average classification accuracies depicted by SVM, LDA, and KNN classifiers are 95.69±5.23%, 96.64±5.02%, and 93.63±4.84% (p − value < 0.05), respectively. Furthermore, the confusion matrix demonstrated the insight illustration of predicted and misclassified events for individual classifiers. In summary, the gait events and gait temporal parameters are reliably estimated using a single IMU with SVM or LDA classifier (p − value > 0.05). Additionally, the efficacy of the proposed model for sensor location and subject variability has been evaluated. The performance of LDA and SVM classifier for gait phase prediction has been found invariant (p − value > 0.05) towards sensor location and subject variability.

Voice-based apps can be effective among rural farmers, if it is in their own spoken language/dialect. Many voice-based apps were developed in the agricultural sector, and in each case, farmers had to either type in the queries or they had to communicate with the device which had the standard speech to deliver the solution which added to the challenge to comprehend the information. This paper presents the research work in developing the wake word detection system for major dialects based on 5 different regions in Karnataka, namely-Dharwad, Dogganal, Tulu, Kodagu and Urban Kannada.The customized wake word system is designed using 1D CNN model with 98% accuracy which showed better results over ANNs with 14.1% and RNNs with 48.1% accuracies. The diversity in regional dialects has been well identified using Conv1D model and with comparative analysis with RNNs to validate on the sequential data the predicted labels were compared and the performance of Conv1d reconciles well for the Dialect Dataset.

In an era characterized by the digital revolution and the widespread adoption of social media platforms, analysing user sentiments online has become essential for grasping public opinion, consumer preferences, and emerging trends. This project delves into sentiment analysis within social media, employing web scraping techniques to gather and evaluate data from diverse online platforms. The central aim of this project is to uncover valuable insights and trends by systematically examining user sentiments. With social media's pervasive influence, individuals and organizations now communicate, share information, and express opinions in unprecedented ways. This abundance of user-generated content presents a unique opportunity to access a vast repository of data that can offer meaningful insights. Sentiment analysis, also referred to as opinion mining, provides a means of understanding and categorizing the emotions, attitudes, and opinions expressed in online content. This project showcases the potential of data-driven approaches in making informed decisions and gaining a competitive advantage in today's digital landscape.

There is an unprecedented need to expand the toolbox of solutions to boost the scalability of clean power and energy systems. Amidst this challenge, nuclear energy is increasingly recognized as an important player in the path toward deep decarbonization of the global energy mix. This paper presents a technology review of small modular reactor (SMR) concepts currently under development and deployment. Both conventional and next-generation reactor technologies are evaluated with a focus on potential power system integration benefits, added system values, and provision of system-bearing services. Nevertheless, there are currently some uncertainties in the techno-economic competitiveness of SMRs, whether they can leverage economics of mass production over their inherent lack of economics of scale. To address this challenge, our paper provides some basic cost analysis of SMRs, taking into account their expected learning curves to evaluate the cost of future deployments and give insights into their economic competitiveness and potential role as a disruptive solution in the energy transition.

"Facial Expression Detection" is one of the most challenging tasks in the field of object detection. Object detection is a field of computer vision that deals with detection and recognition of semantic objects (such as cars, human face, buildings and etc) in images and videos. Nowadays, with advancements in hardware technologies and increase in image acquisition. Facial expression in human images can convey important information. Facial expression can be used for deciding marketing and teaching strategies. Facial expressions along with hand gestures can be used in VR applications. It can useful in providing inclusive healthcare services. In this paper, we propose a novel approach to extract facial features and recognize facial expression from an image. We display our hybrid approach to detect and measure facial features such as eyes, nose, mouth and eyebrows. We also describe various feature points and their associated distance which can be used for expression recognition by matching techniques using Supervised Machine Learning. We have used the JAFFE (Japanese Female Facial Expression) database for this project. We classify each image into seven categories of facial expression viz. 1) Angry, 2) Disgust, 3) Fear, 4) Happy, 5) Neutral, 6) Sad and 7) Surprise.

Vision-based haptic sensors have become an increasingly common tool for providing both coarse and delicate tactile feedback in robotic fingers. We present a soft, affordable, vision-based tactile sensor named Subblescope, which provides the magnitude and direction of the applied force. The sensor consists of a single layer of Ecoflex incorporating an air bubble inside it, which is injected into the sample while curating to stabilize the bubble. An LED array lights the elastomer, while a camera records the shape change of the bubble during normal and shear force interaction. The sensor is calibrated using a micro-tensile UTM. In the case of normal loading, the sensor has a sensitivity of 248.6 µN over a range of 0.02 N to 2.9 N, with a repeatability of 1.09 % at 0.6 N, and hysteresis of 26.23 µJ between 0.1 N to 0.6 N. In the case of shear loading, the range is 0.02 N to 0.15 N with a sensitivity of 31.31 µN, better than the existing vision-based haptic sensors. The multidirectional loading onto the sensor provides 3D shapes of the bubble simulated in COMSOL. The present concept and design are more compliant for sensing bi-directional force interactions while maneuvering interactive surfaces. Multiple bubbles can be populated in the elastomer for further study towards spatio-temporal sensing.

We introduce an innovative deep learning-based method that uses a denoising diffusion-based model to translate low-resolution images to high-resolution ones from different optical sensors while preserving the contents and avoiding undesired artifacts. The proposed method is trained and tested on a large and diverse data set of paired Sentinel-II and Planet Dove images. We show that it can solve serious image generation issues observed when the popular classifier-free guided Denoising Diffusion Implicit Model (DDIM) framework is used in the task of Image-to-Image Translation of multi-sensor optical remote sensing images and that it can generate large images with highly consistent patches, both in colors and in features. Moreover, we demonstrate how our method improves heterogeneous change detection results in two urban areas: Beirut, Lebanon, and Austin, USA. Our contributions are: i) a new training and testing algorithm based on denoising diffusion models for optical image translation; ii) a comprehensive image quality evaluation and ablation study; iii) a comparison with the classifier-free guided DDIM framework; and iv) change detection experiments on heterogeneous data.

This paper underscores the significance of safety and reliability in the realm of Unmanned Aerial Vehicle (UAV) technologies, and how regulations play a pivotal role in ensuring their responsible use. We have analyzed safety incidents and trends both in Canada and globally, noting a decline in incidents attributed to enhanced regulations. Our comparative analysis of different UAV technologies identified batteries as the most reliable power supply, Global Navigation Satellite System (GNSS) as the most effective navigation system, and Light Detection and Ranging (LiDAR) as the optimal optical sensor due to regulatory compliance and system redundancies. We also examined the regulatory framework in Canada, comparing it with the riskbased approach of the European Union Aviation Safety Agency (EASA) and the efforts of Joint Authorities for Rule-making on Unmanned Systems (JARUS) towards global harmonization. Furthermore, we highlighted emerging trends in automation and flight control technologies, with a focus on European regulations shaping UAV automation trends. In conclusion, by adhering to best practices from other regulatory bodies, embracing emerging trends, and adopting a risk-based approach, Canada can promote the growth of the UAV industry while ensuring safety and reliability in UAV technologies.

The paper presents an analysis of the heat transfer process on heating surfaces made with different materials in a fire tube water boiler supplied by natural gas. The analysis was based on analytical calculations and numerical simulations using ANSYS Fluent software. The analytical calculations were performed to investigate the flue gas composition as a product of the natural gas combustion process and to apply these results in numerical modeling of heat transfer phenomena between hot exhaust gas and water in the boiler's heat exchanger. The obtained results show that optimization of heat exchanger design is crucial for efficient heat transfer process in every boiler, which has important implications for the energy efficiency of those systems and environmental protection due to fuel consumption reduction. This research provides valuable insights into the importance of heat exchanger design and its impact on the efficiency of water boilers.

This work presents the utilization of square ring slots as resonators on active and passive metasurfaces operating within the K a-band. Thereby, fully functional prototypes of reflectors and lenses based on a two-layer printed circuit board were fabricated and verified. The analytical model of the square ring slot structure is depicted extensively, which allows finding simulation optimized designs that can be applied to the passive prototypes. Additionally, the gained information on self-resonances is used for combinations with layers, activated through lumped varactor diodes. In the end, the reflection and transmission parameters of single active and passive designs as well as beam patterns of the lenses are measured and interpreted, resulting in a switching dynamic range of 15 dB at a bandwidth of 1 GHz for active reflectors and a maximum gain of 22 dBi at 26 GHz for the passive lens.

Non-fungible Tokens (NFTs) are emblematic of unique digital assets that offer a distinct proprietorship and value creation, whilst Self-sovereign Identity (SSI) imbues individuals with the capacity to govern their personal data and identifying information. The conceptualization of an NFT-based Self-sovereign Identity is still under contemplation within the scholarly community. Despite initial assertions by researchers concerning the potential advantages of integrating NFTs with Self-sovereign identity, the manifestation of systems based on this amalgamation has yet to materialize in pragmatic circumstances. This research paper explores the prospective benefits and impediments associated with the synthesis of Non-Fungible Tokens (NFTs) and Self-Sovereign Identity (SSI). It further unveils a generalized operation flow of an NFT-based SSI, the various iterations of NFTs that can be deployed in the realm of SSI, and an array of refined and expanded SSI principles specific to NFT-based SSI. The paper delves into the technical, legal, standardization, and governance challenges that might ensue from this amalgamation. This research paper accentuates the significance of tackling these obstacles to harness their potential for industry transformation and individual empowerment. Moreover, the paper examines prospective research and developmental trajectories, including technological progressions, potential reverberations on industries and societal structures, as well as the emergence of new business models and revenue streams. It also reiterates the importance of addressing these challenges to actualize the full potential of these emerging technologies. This research endeavor is a pioneering exploration in the realm of NFT-based Self-sovereign Identity.

Adaptation of morphology in response to varying environments is a crucial feature seen in biological organisms. While some robots emulate adaptability through the use of adaptive body parts, practical implementation of morphological transformations in robotics remains limited. This limitation arises due to the complexity of such transformations, demanding the fusion of advanced materials, control systems, and design approaches. In our paper, we introduce a bioinspired quadruped robot equipped with a laterally undulating spine, designed to adapt its posture specifically for navigating complex terradynamic environments. Leveraging a symmetrical parallelogram mechanism, this robot can alter both height and width, enabling traversal across varied surfaces, collision avoidance, passage through narrow channels, and obstacle negotiation. Additionally, our robot's innovative design strategically positions its center of gravity within its support triangle throughout the gait cycle using lateral undulation, eliminating the need for posture-stabilizing sensors or learning algorithms.

Recently, deformable convolutional neural network is commonly used in computer vision tasks, achieving remarkable results. The existing method DCNv3, focuses more on heavyweight models rather than lightweight ones. These heavyweight models are not suitable for small computing devices, which are limited by their hardware to deploy lightweight convolutional neural networks (CNNs). In this article, we focus on applying the DCNv3 operation to lightweight CNNs. To explore the performance of lightweight CNNs based on DCNv3, we conduct experiments and find that DCNv3 does not fully utilize its advantages with lightweight CNNs due to sparse sampling. Yet the traditional solution of increasing kernel size boosts computational load, making it unsuitable. Based on this situation, we solve this dilemma from two levels, the core operation and the visual feature extraction module. At the core operation level, we propose Deformable Strip Convolution (DSCN). As a simplified version of DCNv3 with large kernel, DSCN has only 63.2% computational load of the original with respect to the deformable sampling method. DSCN further avoids a quadratic increase in computational load with kernel size by limiting the deformation sampling kernels to single axis. At the visual feature extraction module level, we propose Deformable Spatial Attention (DSA) constructed from DSCN as a replacement for DCNv3. Specifically, we observe the similarity between the modulation mask branch in DCNv3 and spatial attention, and use spatial attention instead of modulation mask branch based on this similarity to reduce parameters and memory consumption. Finally, in order to verify the effectiveness of our improved design, we further propose a lightweight CNN backbone named DSAN. After conducting numerous extensive experiments, we find that DSA has an inference speed that is 2.1 times faster than that of DCNv3 with large kernel. In dense prediction tasks such as semantic segmentation, DSAN-S with a lightweight decoder achieves 48.8% mIoU on ADE20K, which is higher than the result of InternImage-T based on DCNv3 with a heavyweight decoder, while the number of parameters and computation is only 35.0% and 9.1% of its. Our code is available at https://github.com/MarcYugo/DSAN-Deformable-Spatial-Attention.

As the world rapidly evolves, the integration of cutting-edge technologies like Artificial Intelligence (AI) and Machine Learning (ML) has become increasingly crucial in shaping the future of higher education. This essay delves into the pressing need for Nigerian state universities to adopt these transformative tools within their undergraduate curriculum, empowering students to thrive in the dynamic job market and driving innovation across various sectors.