Surgical resection of bone is largely due to trauma and tumors, resulting in a critical size defect (CSD). Treatment of CSDs involves the use of synthetic or biological grafts to foster tissue growth and bone regeneration. Modern advancements in tissue engineering allow for the rapid creation of such biocompatible scaffolds to assist in bone regeneration. Gelatin methacryloyl (after dialysis) is biocompatible and fosters cell proliferation but it is printable using a 3D printer due to excessive water content. Here, we investigated a novel composite biomaterial comprising of Gelatin Methacryloyl (GelMA), Laponite (LP), and tween 80 for its 3D printing capability with cells and biomechanical characteristics. These cell-laden composite scaffolds demonstrated the ability to degrade slowly across 14 days and were able to retain large amounts of fluid within 24 hours. Cell proliferation was significantly improved 14 days with the presence LP in the GelMA scaffolds.
The present work proposes a vehicle monitoring system’s hardware and software design. The proposed electronic device transmits the data provided by the vehicle’s onboard computer through the OBDII port, using the CAN-BUS interface, respectively the location and the speed of the vehicle, to a web API which has the role of receiving, process, and storing the information in a database. Later, the user can view the given data on a webpage that exposes the information on a map. The device respects all the regulation laws in the transportation field and uses the latest technology, which permits easy upgrades in the future. This open-source implementation permits the end users to customize the visualization of the parameters of interest. Developing and implementing this device is to be used in fleet tracking.
Acquiring platform (Merchant Bank) today faces greater Merchant Fraud cases (catering to small medium business) post pandemic as new methods of payments are evolving. In many cases, the fraud merchants, present the legitimate Know Your Customer (KYC) documents and complete onboarding like a legitimate merchant profile. Acquirers need to mitigate the merchant risk in Payments (multiple method of payments – card Present, card Not Present, contactless Payments (Tap to Pay, Apple Pay, Samsung Pay, using Near-Field Communication (NFC) and biometric based payments), in Point of Sale (PoS) and in large ecommerce portals). To address this, acquiring platform needs to build a comprehensive risk management platform with large set of data from multiple sources (device activity, merchant activity over time, analyze merchant portfolio for a specific geography and finally line of business). Acquirer needs to also analyze KYC data as well as payment data (over period of time). While doing so, risk platform needs to address both account fraud and payments fraud. It also needs to balance accuracy in fraud detection and without compromising high authorization rates. As new payment methods emerge, it is critical that risk management platform gets data from various sources in addition to just KYC and payment data. These are device /peripherals – Internet of Things (IoT) data, leveraging estate management, tracking geo location, analyzing transaction pattern (over period of twenty hours), profiling merchants and merchant segmentation. Risk management platform needs to address PoS 3 needs specifically.
The object recognition in thermal infrared spectrum can possibly be enhanced by capturing radiation signals in narrower subbands of this spectrum and performing recognition in color or multiple channel thermal infrared images. In this work, we investigate possible benefits of 2-channel thermal infrared images captured by commercial cameras. We performed experiments on our collected images containing persons and cars. Fusion of object recognition results obtained in different channels separately, gives some improvement over the use of a recognizer with single channel full spectrum images. We also present a proof-of-concept design of adaptable thermal imager based on asymmetrically-doped double quantum well arrays, which can efficiently capture multiband images in the future.
The cryptographic security protocols are designed to ensure that the information transmission that is taking place between two entities is secured and is not impacted while interaction occurs among them. Security protocols are a set of rules that are formulated to ensure that the intruder or third party is not able to access the information that is transmitted. In this context, one of the algorithms that could be used is the RC4 algorithm. There are many discussions about the RC4 algorithm, but as of now is only theoretically dominant. It is a fact that the theory alone cannot complete the RCA algorithm process in a detailed manner. The applications reveal the input and do not produce any output of the processes involved. This research paper will illustrate the vulnerability and possible solutions to avoid the RC4 Algorithm attack.
The main goal of this research is to identify, demonstrate, and provide a secure approach to homeowners in creating fire-safety checklists and pre-incident plans of their homes, and uploading them through a secure cloud service to the centralized fire safety cloud-based system, ALIKE, which in turn provides updated and scalable views of the home interiors for their local fire departments. ALIKE is a cloud-based system that integrates 3D scans, AI, image recognition, and augmented reality to help homeowners improve the safety of their homes and conduct pre-incident planning with the local fire department. The ALIKE system provides for two modal approaches: the homeowner system approach and the firefighter system approach. Watson AI functionality within ALIKE system can identify potential fire safety issue in the home, examine the issue and recommend fire safe products to mitigate the potential hazard. Thus, the cause of residential fires would be reduced to a great extent, providing safer homes for residents.
While the use of Quantum Key Distribution is very secure, some faults can occur, specifically, that can create a manin-the-middle attack. This paper examines how the man-in-the-middle attack can occur and why Quantum Key Distribution is still the superior encryption choice. Applying Post-Quantum Cryptography (FALCON Algorithm) to the Quantum Key Distribution should create more secure encryption to mitigate a man-in-the-middle attack, thusly protecting the communication from the eavesdropper.
Fault occurrence, location, and system performance information, based on the analysis of traveling waves is gaining importance as new levels of high frequency measurement equipment are being introduced to the power grid. In this work Continuous Wavelet Transformations with the complex Morlet wavelet analyzing functions was used to design a signal processing scheme to extract time and frequency characteristics from measured and recorded transient events that launched traveling waves on a 115 kV transmission system. The primary motivation was to extract baseline system performance information employing the traveling waves launched as a result of line re-energizations. This work is a proof of concept aimed extracting and baselining a system’s line specific characteristics or ‘fingerprints’ and determining if a useful level of fingerprint consistency exists for typical line energizations. Implementation of the proposed scheme on these specific signals, measured at 1.5 MHz sampling frequency, successfully extracted consistent fingerprints of the system’s time-frequency behavior. A database of these fingerprints could be used to determine a dynamic high-frequency model for the system as well as track changes in the physical system as the response changes in different source and system operational scenarios.
We developed a subspace classifier for measurement classification to provide an alternative to current deep learning approaches. Many modern neural networks cannot provide an understandable explanation for their classification. The subspace classifier provides a decomposition of the classification problem making it computationally simpler. We first use a Bayesian method in which all the class conditional probabilities for the entire measurement space can be stored in memory.Then we made experiments with simulated class conditional distributions and defined a subspace classifier that only stores the class conditional probabilities for the subspaces. This can use much larger distributions than the previous model as it uses much less memory so we expanded to cases where the measurement space is generated sequentially and everything does not have to be in the memory at the same time.For cases with distributions that fit in the memory we also compared a Bayesian approach with the subspace approach. The Bayesian subspace classifiers consistently outperformed the subspace classifiers without Bayes rule by a large margin. We also compare the subspace classifier with 3 Python Machine Learning Models, namely a Ridge Classifier, a Multi-Layer Perceptron (MLP) classifier (neural network), and a Support Vector Machine (SVM) on a set of tuples and class conditional probability distributions with 4 classes. The subspace classifier had an average probability of correct identification of 0.25172, the SVM model had an average accuracy of 0.20987, the neural network MLP classifier had an average accuracy of 0.2140 and the Ridge Classifier had an average accuracy of 0.2798 over 10,000 trials.
Across all scientific fields, the amount of data generated continues to grow exponentially, which creates both opportunities and challenges. Increased availability allows for new insights into phenomena and stimulates discovery. While traditional ways of interacting with data have focused mostly on flat displays, Extended Reality (XR) technologies have the potential to transform the way both research and education are done. They provide an immersive and interactive experience that allows participants to visualize and interact with data while also working collaboratively. In this paper, the research and development of immersive educational 3D visualization tools aligned with several science drivers including environmental science and sustainability and life sciences is described. Specifically, the development of two XR applications is discussed with emphasis on the innovative mechanisms introduced using head-mounted display technologies. Preliminary results from the prototype applications indicate that the use of the XR technologies significantly expands the visualization ability by integrating additional data sources and improving the user’s experience.
In this paper, we present an overview of quantum neural network (QNN) and quantum computing benefits in the Artificial Intelligence field with methods used to enhance the application of machine learning. The approach uses quantum computers to enhance Convolutional Neural Network (CNN) with Hybrid QNNs using IBM Qiskit called the Convolutional Neural Network Hybrid Qiskit (CNNHQ). This paper will compare and focus on using the quantum concepts of computation to develop and optimize the network process entirely depending on Activation functions used in the neural network approach.
In this paper, we implement and experiment with a portion of our overall hybrid deep learning framework, An Affective hYbrid SPIking Neural Network [ARSPI-NET]. In order to motivate and show the usage of liquid state machines as an efficient, feature extraction framework we perform experiments on connection architecture as well as neuron model and their effect on overall classification performance. We perform our initial experimentation on the MNIST dataset and achieve a 87% classification using a liquid state machine and logistic regression classifier. Our results suggest that our framework can compare with current models in terms of accuracy, however, we outperform traditional deep learning methods in terms of energy consumption and the potential to move to energy-efficient neuromorphic platforms. In addition, our framework has the advantage of being more interpretable, as it allows us to model the spatiotemporal dynamics of signals through the usage of a liquid reservoir and an interpretable readout vector. This paper sets precedence for future experimentation and development of ARSPI-Net as a hybrid deep learning framework.
Bone tissues in critical size bone fracture cases do not re-generate naturally and require special treatment procedures, such as cast alignments, support plates and bone grafting. These procedures are risky and often have high rejection rates. Modern clinical procedures use scaffold implants that can facilitate the process of healing with a lesser risk, provide mechanical support and a porous, nutritious medium for bone tissue regeneration and recovery. They require special training, tools, and significant time to be manufactured, and are generally made at a dedicated laboratory. The whole process takes around a week to be manufactured and implanted in the fracture site. In this work a novel technique for automatic segmentation of bone fractures from CT scan images to facilitate the process of manufacturing the patient-specific scaffolds in a significantly shorter time, without the need of skilled personnel has been presented. To achieve this, the procedure of generating 3D printable models was automated using image processing and machine learning algorithms. For this, 3D CT (Computed Tomography) images were used as input (as a series of 2D slices), acquired using a micro-CT scanner for the approach. After pre-processing the acquired images (filtering), thresholding segmentation was applied to extract the bone from the scan. This step is followed by orientation optimization of the segmentation result, by taking advantage of a global optimization technique, namely Simulated Annealing, to ensure maximized visibility of the fracture in a projected view of the volume by projecting the volume on a 2D surface. Binary hole-filling techniques and bone thickness estimation is then used to create a 3D template (model) to be sent for scaffold printing to a compatible 3D printer. Experiments with both synthetic and real datasets show that the proposed method is an effective approach for creating rapid, precise, and patient-specific 3D scaffolds to treat critical-size bone fractures.
Historically, nuclear science and radiation detection fields of research used Pulse Shape Discrimination (PSD) to label gamma-ray and neutron interactions. However, PSD’s effectiveness relies greatly on the existence of distinguishable differences in an interaction’s measured pulse shape. In the fields of machine learning and data analytics, clustering algorithms provide ways to group samples with similar features without the need for labels. Clustering gamma-ray and neutron interactions may mitigate PSD’s pitfalls, since clustering methods view the total waveform rather than just the area under the tail and the total area under the pulse. However, traditional clustering methods, such as the k-means clustering algorithm, suffer from poor performance on high dimensional data. This study explores unsupervised machine learning methods using Deep Neural Networks (DNN) to cluster gamma-ray and neutron interaction measurements collected with an organic scintillation detector, in order to perform binary labeling of gamma-rays and neutrons. Using various network architectures, this research demonstrates the effectiveness of using autoencoder-based neural networks to cluster gamma-ray and neutron interactions when compared to shallow clustering algorithms. The results reveal the effectiveness of autoencoders on high energy gamma-ray and neutron pulses with an energy deposit greater than 0.80 MeVee whilst greatly outperforming k-means comparatively in all cases.
A working prototype of an environment-friendly multipurpose solar charging station is presented here. It is meant for use in places that have limited access to electricity but has abundance of solar irradiance. While solar charging stations have been available and in use, the charger presented here will be capable of tracking the path of sun to achieve greater efficiency through better harvesting of solar energy. It uses pulse width modulation technique to provide the desired output voltage for charging. We have developed an initial prototype that delivers up to 75% efficiency and can follow the path the sun during the day. Our goal is to improve on what we already achieved and further develop the product for commercial production.
The diversity of modern biomedical, industrial, and military requirements is increasing the level of innovation and workability required to deploy modern solutions. A key aspect of this technological innovation has been required in the photonics industry to explore potential applications and workability avenues. Across leading industrial areas, lensing technology is emerging as the primary method to bridge existing gaps and paving forward solutions that provide viable solutions to the existing gaps in technology and application.The aspects of the innovative lensed fibers are emerging as the cornerstone to fulfill existing gaps and pave forward new applications across security, biomedical and industrial performance utilizing the core aspects of laser technology through lensed fibers. The technological application using particularly perpendicular lensed fibers is a key aspect of the innovation explored through directed research and coordinated solution management in biomedical and defense areas.
Cognitive radio (CR) is an important wireless technology for contemporary spectrum-demanding applications, such as the fifth-generation and internet-of-things communication systems. The CR technology can dynamically alleviate the common problems of spectrum underutilization and scarcity and thus can optimize spectral efficiency and throughput. Unlicensed users are equipped with CR transceivers to monitor the real-time usage of the licensed channels so that they can “rendezvous” whenever there is any unoccupied channel. This paper reports the implementation of a CR experimental testbed using off-the-shelf modern software-defined radios (SDRs). To support the ad hoc nature of CR wireless communications, a family of shift-invariant (SI), asynchronous channel-hopping (CH) sequences that have the desirable properties of short maximum-time-to-rendezvous and full degree-of-rendezvous are employed in this first-of-its-kind testbed. The hardware set-up and procedures of two SDR experiments are detailed. The throughput performance of these CH sequences in the testbed is modeled, analyzed, and validated by the experimental and simulation results.
Over the past five years, almost every application of engineering has tried to implement an artificial neural network to improve performance in one way or another. The study of real neurons in the human brain on the other hand, which paved the way for such applications, has been ongoing for well over five decades. While significant progress has been made, artificial neurons fall well short of the capabilities of real neurons in some key areas, such as the ability to perform lifelong learning. This limitation can be traced back to fundamental differences in network architecture and overall implementation. To remedy this, attention must first be directed back to the foundational science behind how the human brain acquires, stores, uses, and selectively removes specific memory or knowledge. Next, novel architectural concepts and implementation strategies to address the aforementioned limitations must be developed. Here, this will be done through careful considerations of all aspects within the repetitive regime. The human brain, by design, is very reliant on continual lifelong learning to solve problems. Control strategies used today, artificial neural networks, and even combinations of both are unable to optimally engage in this fundamental process. This paper attempts to bridge the gap and push toward enhanced lifelong learning control strategies for future use.
The MADE (Molecular Augmented reality for Design and Engineering) application is an augmented reality (AR) tool that affords users the ability to visualize complex macromolecules in an interactive and exploratory environment. The process to develop and design MADE required a thorough analysis of user segments, which informed the commercial prospects of MADE’s smartphone and HoloLens application. To determine the user’s needs and corresponding value propositions, a set of customer discovery interviews were conducted with individuals from MADE’s potential user segments. Through the customer discovery process, active user engagement revealed unique value propositions and helped identify user segments of higher education students and professors and life science professionals. For the aforementioned user segments, the design of the MADE application serves as an educational tool for enhanced learning experience, as well as a collaboration tool for exploratory drug development and discovery. The discovery and design process for MADE revealed viability of the application for commercialization in the higher education and life science profession markets. This paper discusses the processes of research and design of the MADE application.
Nuclear counter proliferation and nuclear material stewardship are critical international processes to maintaining safety and international balance. Monitoring, locating, and investigating special nuclear material is integral to both of these aims. Advanced radiation detection systems contribute to nuclear stockpile safekeeping and prevent unstable military and geopolitical environments. A Time Projection Chamber (TPC) is a common piece of equipment in the world of high energy physics but never used in the realm of nuclear security. A TPC is capable of providing a direction to a source of fast neutrons, a signature unique to spontaneous fission sources and Weapons Grade Plutonium specifically. The ability to collect direction and energy information from a single incident fast neutron is unique to the time projection chamber design. A time projection chamber’s unique data outputs can be used to create a fast neutron detector capable of robust data collection and special nuclear material localization. Previous time projection chamber experimentation and design proves the concept’s effectiveness in lab environments. The Soldier Mounted Advanced Radiation Tracking Detector design uses printed circuit board technology for significant improvements in the field uniformity, allowing for an unprecedented active volume to total volume ratio. The structural improvements will provide the end user with a portable fast neutron detector. Previous work has demonstrated through simulations that a portable TPC can provide a source location accurate to within 60 cm after a single pass with a closest approach of two meters. The improvements discussed in this work result in a single user being capable carrying a TPC while simultaneously collecting data to locate and characterize a 1kg source of WGPu on a single pass from two meters.
