The latest pandemic situation (COVID-19) tremendously affected societies, having also a significant impact on higher education. The disruptions caused by the pandemic, including widespread school closures and limitations in physical classroom interactions, further emphasized the obstructions faced in the continuation of education due to the COVID-19 pandemic. The situation clearly indicated the vulnerability of education continuation and the problems of remote education. As a positive side effect, the COVID-19 situation accelerated the pace of developing digital-technology based innovative approaches to teaching and learning.
In this context, the research project aims to investigate the efficiency of utilizing semi-autonomous robotic assistants, which are advanced technological devices designed to assist in various domains, including healthcare, elderly care, telemedicine, and facility inspection. This study focuses on exploring the applicability of these robotic assistants for teaching technical courses in higher education, bridging the gap between innovative teaching approaches and the potential of robotic-assisted instruction.
By examining the role and effectiveness of these robotic assistants, the research aims to study how these devices can be employed for the purposes of enhancing the learning experience for students and improving overall educational outcomes. Through careful analysis and evaluation, the project will explore how robotic assistants can contribute to increased student engagement, personalized instruction, and effectively transfer technical knowledge.
The findings of this research will provide valuable insights and recommendations for integrating robotic assistants into teaching practices, ultimately innovating and optimizing the educational environment in higher education settings, e.g., technology universities. In addition, the study will explore potential challenges and barriers.
==> Abstract (PDF)Cash movements happen daily between banks, retailers, and the Federal Reserve Bank. These transactions necessitate the use of armored couriers to safely transport large amounts of currency and other valuables. This industry, however, functions without the aid of current technology, relying on paper receipts and faxed manifests.
This research proposes the adoption of RFID into cash handling to provide real-time tracking and visibility while decreasing costs and increasing system capacity.
For many years, academic institutions have offered so-called summer schools as a way to provide additional or standard teaching content to selected groups of students outside the regular curricular activities. The reasons are quite diverse.
Summer schools may allow students
In this presentation we will introduce and discuss a new international summer school, jointly organised by Kiel University of Applied Sciences and China Jiliang University (CJLU) in Hangzhou, China, and actively supported by several international partner institutions.
The general focus will be on Smart City (SC) Innovation Technologies, including: IT for Smart Cities, Ubiquitous and Pervasive Computing, Internet of Things (IoT, Building IoT), Data Science, Big Data, Artificial Intelligence (AI), Cyber Security, Creative Design, Innovation Management and Entrepreneurship.
The location of this summer school, Hangzhou, is the capital of the Zhejiang Province in the Eastern part of the People's Republic of China (PRC), in close vicinity to Shanghai and other important industry hubs. Zhejiang is considered one of the wealthiest provinces in China with Hangzhou being the innovation center of China.
Although the city’s wealth was originally based on its silk and textile industry, and the tea plantations, it is now the Chinese center for research, development, and production in the Information Technology area in general and particularly in the field of Artificial Intelligence (AI).
The e-commerce giant Alibaba Group has its headquarters in Hangzhou, maintaining several R&D centers as well as innovation and start-up hubs.
Hangzhou hosted the G20 summit 2016. Its most popular touristic attraction is the West Lake, a UNESCO World Heritage Site.
Observation:
This summer school was first proposed for 2020. Due to the Corona pandemic and the current world situation, it had to be postponed several times. It has not yet been decided whether the event can be offered in 2024 as currently planned.
The method calculates the food intake of adults and children under dietetic/medical guidance/monitoring. Dieticians apply this method, to keep in contact and guide patients/clients; still, can be applied in community-nutrition services, to support public health. For community reasons a platform may be developed for participants to receive nutritional recommendations; achieve weight loss and health promotion goals. Images of the actual food plates before and after the meal are recorded and transmitted using mainly cell phones or tablets, or computer applications; to be further analyzed and compared to standard pre-calculated meals.
Applications are developed, to calculate energy and nutrient intake, including macronutrients, as protein, lipids. Data may be transferred and analyzed retrospectively, in near real-time, or even in real time. The method may be incorporating computer automation to improve accuracy. However, reliability and validity of food intake measurement must be examined, in each case of integrated applications, since the technical aspects of the recording, as well as the transfer and analysis software may affect the overall results. The ability to compare a participants’ dietary intake, first with the reference food (i.e. the accuracy of these food composition); the comparisons within a group or between groups etc., represent major challenges. Furthermore, the benefit as well as the limits of the involvement of experts that cooperate with integrated intelligent systems, on the basis of not only the nutritional analysis but the nutritional guidance and even clinical nutrition support also, remains an area of applied dietetics to be further studied.
==> Abstract (PDF)Purpose: The purpose of this paper is to assess if blockchain technology, which is hailed for its potential to improve supply chain transparency and efficiency, can deliver to this promise (Abreu et al., 2021; Mahyuni et al., 2020; Manzoor et al., 2022; Rennie, 2022; Yiannas, 2018). Supply chains are increasingly complex and full transparency is difficult to achieve and may sometimes be undesired, yet governments worldwide are demanding more transparency especially with respect to the social and environmental sustainability of supply chains (Francisco & Swanson, 2018; Franke, 2021; Najjar, 2021). If blockchain technology allows to close this gap, without compromising corporate confidentiality and efficiency requirements, it might change supply chain management drastically. Most other research in this field remains rather superficial with respect to the fundamentals of blockchain technology and supply chain management. Our study will add to the very limited theoretically founded research on blockchain potentials in supply chains.
Study Design: Our paper is conceptual. Based on the extant literature, we compare blockchain technology and its advantages to the characteristics of supply chains and challenges of supply chain management to identify a theoretical fit. We investigate if published case studies of blockchain technology in supply chain management support this. Finally, we devise a conceptual framework for the application of blockchain technology in supply chain management.
Findings: There is, currently, very limited use for blockchain technology in supply chain management. It does not provide an answer to several of the challenges that exist in supply chain management. Blockchain technology was primarily devised as encryption technology for virtual entities and transactions involving them and not for tangible objects. Published case studies describing the use of blockchain technology in supply chain management do not contradict these findings.
They rely on private blockchain solutions lacking the key benefits of blockchain technology. Our conceptual framework constitutes a reference for a future public blockchain solution for supply chains. It offers solutions for some of the challenges but also cannot solve some fundamental challenges, like power imbalances in supply chains.
Limitations: Blockchain technology undergoes rapid development. Future blockchain technology may be faster, less energy-intensive and more integrated with other technologies. This may be aided by developments in hardware, like quantum computing. The findings of this paper are limited to currently available technology. Implications: Companies and governments alike should be very cautious when considering investing in currently available blockchain solutions for supply chain management.
Last year, a marvelous technology was born and became world-famous within weeks. Its name - ChatGPT - an AI. While we had encountered other AI systems before, this one was distinctly different. It was not limited to one particular topic, and one could even engage in discussions in a human-like manner.
In this talk, I aim to highlight some fascinating discoveries, whether found elsewhere or uncovered through my own exploration, that have proven to be quite intriguing. Side note: This abstract has been cross-checked by ChatGPT.
Despite substantial advancement over the past ten years, learning opportunities are still unevenly distributed and many people are omitted of the educational system. Deaf people need special educators and sign language experts. However, there is a lack to well-trained educators for deaf people. Computer Assisted Teaching in Sign Language (CAT-SL) was a three-year EU funded project that fulfills this gap and develops an innovative and affordable system for interactive SL teaching for students in Special Education/Pedagogical departments and primary school education.
CAT-SL project developed a curricula and guides for teaching SL using the CAT-SL system for three multilingual courses, in partners countries - Greece, Cyprus, Portugal and Netherlands- to support the social inclusion of the deaf children. The CAT-SL system is based on computer-vision, machine-learning and linguistic technology developed by all the involved partners in several ongoing or recently completed national and EU-funded projects. The project can serve as a basis for future similar projects and can be easily extended to more international Sign Languages.
==> Abstract (PDF)With the prevalence of sophisticated cyberattacks, it is imperative to utilize advanced techniques to protect computer systems from potential threats. Intrusion detection tend to plays a crucial role in the security of computer systems by detecting potential attacks and enabling proactive responses to them. Various methods have been proposed to detect attacks, including signature-based, anomaly-based, and behavior-based methods. However, traditional intrusion detection methods based on signatures and predefined rules reach their limits when faced with complex and evolving attacks.
In this work we analyze, experiment and compare the effectiveness of different models e.g., Artificial Neural Networks, Support Vector Machines and Random forests for network intrusion detection. The obtained results from the conducted experiments on a benchmark dataset, revealed that although the Random Forest classifier outperformed the rest of the aforementioned methods, machine learning algorithms offer a promising approach to improve the accuracy and efficiency of intrusion detection systems.
==> Abstract (PDF)Current wireless sensing systems aim to accurately and efficiently detect, estimate and extract useful physical information/features of designated targets by exploiting radio wave transmission, reflection, diffraction and scattering. However, existing wireless sensing systems face critical challenges in practice (non-availability of line-of-sight link, limited sensing range due to high path loss, etc.). Furthermore, wireless sensing is expected to become a major service of the next generation (6G) wireless networks, in addition to wireless communication services. In such an Integrated Sensing and Communication (ISAC) framework, the emerging technology of Reconfigurable Intelligent Surfaces (RISs) has been recently proposed to reconfigure the radio propagation environment for not only boosting the wireless communication coverage and capacity, but also enhancing sensing in an ambient way.
This tutorial presentation will cover design issues of the aforementioned envisioned ambient sensing environment that range from the design of the RIS elements to various RIS-aided sensing architectures, to optimization and deployment issues. Furthermore, open issues and challenges associated to unlock the potential of RISs for the implementation of the ISAC framework are addressed. Finally, performance evaluation results from both analytical studies and experimentation are presented, for representative ambient sensing use cases with emphasis to human posture recognition and assistive health care.
==> Abstract (PDF)In this work, we present the development and evaluation of a computer-aided diagnostics system designed to enhance lung cancer screening. We employ advanced image analysis techniques, specifically 3D Convolutional Neural Networks (CNNs), to facilitate the segmentation and classification of pulmonary nodules in low-dose CT (Computed Tomography) images.
The early detection of lung cancer is of paramount importance, and this work aims to contribute by automating image analysis to assist radiologists. Our investigation extensively explores the capabilities of 3D CNNs in nodule segmentation and classification. We utilize a publicly available dataset of low-dose CT images to train our model, enabling it to discern intricate spatial features necessary for precise nodule segmentation. Furthermore, we meticulously fine-tune the classification capabilities to distinguish benign from malignant nodules.
The results demonstrate significant promise, as the integration of 3D CNNs improves nodule segmentation, yielding accurate 3D predictions. Moreover, our classification performance advances the accurate identification of malignant nodules, thereby assisting in making informed clinical decisions.
In summary, this work constitutes a valuable contribution to the field of computer-aided diagnostics. By introducing a novel approach to lung cancer screening utilizing 3D CNNs, our findings underscore the potential of the developed system as an essential tool for radiologists. This system has the potential to streamline the detection and classification of pulmonary nodules in low-dose CT images, ultimately aiding in the early detection of lung cancer.
==> Abstract (PDF)Industry 4.0. covers the set of technologies that are allowing the leap to the digital and connected industry. Large companies already have innovation departments that allow them to integrate digitization, but small and medium-sized companies often do not have the resources to tackle such a major change and need the support of external agents to help them. The Digital Transformation in Advanced Manufacturing-DTAM project wanted to meet this skills gap. The DTAM initiative has defined an educational curriculum to train EU technicians to deploy and manage digital tools in Smart Manufacturing. The curriculum focuses on deploying and managing digitalisation technology in the Advanced Manufacturing sector.
The DTAM project will run until the end of 2023 under the coordination of Politeknika Txorierri (Spain) with a budget of almost 1 million euros linked to the European Union Erasmus+ programme. The DTAM consortium brings together 10 organisations from Spain, the Netherlands, Italy, Greece and Bulgaria representing different stakeholders sharing a common vision: vocational education and training providers, higher education institutions, digital transformation experts and sectoral representatives are collaboratively designing, testing, refining and exploiting an integral curriculum in digital transformation competence for mid to high level EU technicians, available in English and all 5 partner languages.
In this session, we will present the developed online modules on Cybersecurity, Sensorica, Machine Learning, Big Data and Soft Skills and their connected IoT labs for hands-on practical training.
==> Abstract (PDF)In this paper, we present a novel software-based face Presentation Attack Detection (PAD) method named ”Apple in My Eyes (AIME)” using screen display as a challenge and corneal specular reflections as a response for authenticating the liveness against presentation. To detect face liveness, AIME creates multiple image patterns on the authentication screen as a challenge, then captures meaningful corneal specular reflection responses from user’s eyes using the front camera, and analyzes the reflective pattern images using various lightweight Machine Learning (ML) techniques under a subsecond level delay (200 ms).
We demonstrate that AIME can detect various attacks, including digital images displayed on the phone or tablet, printed paper images, 2D paper masks, videos, 3D silicon masks, and 3D facial models using VR. AIME liveness detection can be applied for various contactless biometric authentication accurately and efficiently without any costly extra sensors.
DeepFake has accomplished notable advancement with the AI-leveraged production and manipulation techniques of fictitious human facial images. Despite many benign and fun applications, the generated fake images can negatively influence the authenticity of online information by originating deception, manipulation, persecution, and seduction, defying societal quality and human rights, which becomes critical security and privacy threat in social networks. Hence, real-time DeepFake detection and limitation technologies on the mobile platform are essential to building a controlled, harmless DeepFake ecosystem. This paper presents a real-time, cloudless, lightweight mobile app for human visual DeepFake detection using machine learning technologies named MobiDeep (Mobile DeepFake Detection through Machine Learning-based Corneal-Specular Backscattering).
MobiDeep stems from a hypothesis that the existing DeepFake creation methods, including replacement, editing, and synthesis, lack the ensemble with the reflective objects. Focusing on the most reflective area of a human face, corneal-specular backscatter images of eyes, we seek the similarity and consistency with multiple surrounding environment features, including color components, shapes, and textures. We have implemented a crossplatform mobile application to evaluate the performance using various input parameters and lightweight Deep Neural Network (DNN) architectures. The empirical results show that MobiDeep achieves high accuracy (98.7%) and rapid detection speed (less than 200 ms) in detecting sophisticated DeepFake images within a subsecond.
Blended Mobility Project methodology is devoted to create and manage international multidisciplinary teams of students who will collaborate in order to develop a solution for an engineering problem. These teams are set up for a semester with the purpose of developing and presenting a prototype or a proof of concept for a given challenge.
Blended Mobility Project in the academic year 2022/23, completed its 13th edition which started in the academic year 2009/10. Altogether HIEs from 10 countries were participating including Portugal, Belgium, Germany, the UK, Greece, Italy, Kurdistan - Iraq, Slovenia, France and Lithuania. Information Technology companies provide real project proposals. With this professional involvement, students got a context which is international, multicultural, multidisciplinary and professional.
The preparation of the course starts at the beginning of the first semester in October. At this stage the teachers collect challenges from companies and select the most interesting with regards to its pedagogical potential. Initially 6 projects were selected. The selected challenges are presented to final-year-undergraduate or master students. Each project is implemented by a team of about 10 students which are mainly from Information Technology and Software Engineering disciplines but students from other fields of study, such as: Business Development, Management, Electronics, IT & Design participate. Applicants are selected based on a set of criteria defined by each partner university and the teams for each challenge are setup. In 2022/23 edition, 52 students were involved actively, as well as 17 teachers from the 12 participating HEIs. Allocating 10-11 students per team 5 projects were chosen to be implemented. This process was concluded by the end of January and the first face-to-face meeting that runs at the beginning of the second semester, took place at Orleans, France, 13-17 February, 2023. At this first face-to-face meeting, students get to know each other, the company offering the challenge and its details. The challenge is provided to the students by the company but no specifications are given concerning the solution to develop; that is the students task. Students have to interact and cooperate during the semester in order to agree on the necessary specifications and on how to integrate all the elements of the solution from a technical, marketing and business point of view. The first face-to-face meeting runs for five working days during which students design a first draft of the solution for the challenge at hand organize themselves to work as a team during the semester and assign responsibilities to each team member according to the number of ECTS credits they get for their work. At the end of the week the envisaged solution by each team is discussed with the company, the teachers and the students so all agree on a definite proposal. After this first meeting, students work at their home institution working at a distance through online groupware platforms. At the end of the project all modules are integrated and the fully operational system, a unique product, is presented by the students as a team. The second and final meeting took place in Porto, Portugal 19-23 June 2023, students get together face-to-face again to finalize their solution, their final presentation and to discuss the delivered product with the client company and the teachers. The team as a whole must guarantee that all parts integrate well to produce a unique solution for the problem and present the full solution to the project jury. The project jury was composed by a teacher from each partner institution and a representative from the client company.
In total there were 5 projects implemented in the academic year 2022/23 and 2 of them will be presented where HMU students participated. The first project was suggested by MEFA, a German company based in Trier. The students of team MEFA were tasked with developing an online platform for real estate marketing, which essentially functions as a way for companies to promote real estate projects to the worldwide housing market. There are various online marketplaces for real estate currently in operation, however this platform stands out by having a user-friendly design and providing a separate, customizable front-end page for each company's profile and real estate projects, thus allowing companies to maintain their branding while using MEFA's website. The platform provides a dynamic environment for clients to submit their projects, each provided with a personalized webpage for showcasing the projects they’ve been working on and a specialized administrative page was developed for managing and approving client submissions, while offering an admin dashboard for efficient project management. The webpage design closely followed the branding aesthetics of MEFA, parent company, and thus achieving uniformity. The development was carried out on Node.js, React, MongoDB, and Express while communication was managed by tools like Slack for messaging, Github for version management, and Google Meet for virtual meetings. The team consisted of ten members from different nationalities and was characterized by its diversity and uniqueness. Disregarding the language and physical barriers, teamwork was achieved and developed within the agile Scrum framework, comprising Front-end, Back-end, and Marketing subgroups. The Marketing team focused on research around the topic and what innovative ideas could be implemented while the Front-end and Back-end teams worked closely together on the development of the platform and thus adding functionality to product. This unified and collective effort lead to the creation of MEFA real estate platform, which was then delivered to the client for the initialization of its operation. The project was a learning experience for all involved, regarding both the development tools that were used and also working as a team in a remote-working environment. For many of the members, it was their first time using the tools that were decided upon, such as React, Express and Git/GitHub and considering the scope of the project, everyone certainly learned many new skills and gained plenty of valuable experience from the project.
The second project was introduced by YEAT, a Belgian startup company. The YEATapp is a mobile application that send immediate notifications whenever a highly recommended restaurant has a last-minute table available. The challenge was to create a Dashboard for the Admin and a Web version for concierge services. Regarding the Admin Dashboard, it simplifies oversight for administrators, handling user management, settings adjustments, and ensuring smooth operations while the Web version catering to concierge services, it offers clients a platform to request assistance and enjoy exceptional service. The main focus is on the effectiveness and the provision of a fantastic overall experience for all involved. The team was divided into four subgroups: Development, Human Resources, Marketing-Business and Design. The Development team, who was responsible for the creation of the Dashboard for the Admin and the Web version for concierge services, was divided into 2 sub-teams, the Front-end and the Back end. In the process of creating the dashboard and web version, we employed the technology stack consisting of ReactJS for the front-end, Laravel for the back-end and MySQL as the database management system. The Human Resources team was inquired to formulate and create a career plan, to search for employer branding strategies and to deal with recruitment. Regarding the Marketing-Business team, it was engaged in analyzing the database, developing a marketing plan and conducting research on potential expansion into new countries. In conclusion, the design team's primary focus was the design aspects of the Web Version and documentation. Moreover, they undertook the responsibility of managing the Miro Board and executing the rebranding of the application while it is noteworthy to mention that Trello was selected as the tool for team organization and management. Additionally, it is important to highlight that communication within the team was achieved through diverse platforms including Google Meets, Discord, and WhatsApp.
==> Abstract (PDF)Colorectal cancer remains a significant public health concern, and effective screening is fundamental in early detection. In response to the complexity of screening decisions and the need to tailor screening approaches to individual patients, our project focuses on developing a shared decision-making tool to guide healthcare providers and patients, principally from vulnerable population in choosing the most appropriate screening method
Utilizing the PLCO dataset, we constructed a LGBM model. This model supplemented with SHAP enables us to not only calculate probabilities but also gain valuable insights into a patient's individual risk of developing colorectal cancer.
Given the sensitive nature of the vulnerable population we serve, we meticulously designed our feature set to be straightforward and easily answerable by patients, prioritizing accessibility and effectiveness for those with limited health literacy and to fit in with the current practices of healthcare professionals.
Our feature set includes essentially non-exam-based variables such as age, sex, weight, height, BMI, alcohol intake (current and during ages 40-54), smoking history, smoking quantity, diabetes, hypertension, and heart problems. These variables not only simplify the decision-making process but also align with our commitment to cost-effective and timely decision-making.
In assessing the performance our model achieved an AUROC score of 0.716, signifying its ability to discriminate between patients at different levels of colorectal cancer risk. This outcome underscores the potential of our tool to provide valuable insights into the personalized screening decision-making process.
In summary, our project seeks to bridge the gap in colorectal cancer screening by offering a user-friendly shared decision-making tool with its simplicity, robust performance, and focus on accessibility for a vulnerable population, this tool holds promise in enhancing patient engagement and promoting informed decision-making for more effective colorectal cancer screening strategies.
==> Abstract (PDF)The field of artificial intelligence AI deals with theoretical issues and applications that support the ability of machines/computers to perform intelligent actions and tasks, those that are equal to human ability. Applications in the field of AI are complex and require careful planning and constant consideration of both the capabilities of the computational and technological infrastructure on the one hand and aspects of usability and the application itself. In every area of our lives, intelligent systems are integrated that leverage the capabilities of processing information of various types for the sake of streamlining processes, personalization and user experience.
In light of last year Generative AI revolution, we are experiencing a critical change educational system both in teaching AI to undergraduate and graduate students and also in adding AI tools as part of the ongoing tutoring. In todays’ world the student is facing the challenge of designing an AI system with an emphasis on human-machine interfaces, language processing, and data analysis for the benefit of streamlining and leveraging applications.
On the other hand, as teachers we need to make sure that the AI tools and applications that exits are used properly to enhance creative and critical design and not to “manufacture” synthetic results. What should be “allowed” or “prohibited”, which ethical core need we adapt. What is our responsibility in teaching about the advantages and restrictions of the new technology to encourage “Responsible AI”?
Estonian IT College (EIC) was established in 2000 when there was bold linear growth of IT related workplaces but universities were too slow to increase number of graduates. Private sector, who saw real business obstacles ahead was ready to contribute eagerly and as a result new educational institution was defined as joint venture of state, private sector and two largest universities in Estonia. Needless to say that EIC changed a picture and initialized real competition between HE institutions. In following 15 years the situation changed gradually as universities recovered and started to show up more flexibility in all study and research processes. The cost per graduate at EIC was in par with universities and the number of graduates per year still lagged initial expectations. Owner’s decision in 2016 was to merge EIC with Tallinn University of Technology (TalTech) to simplify ministry level control and because EIC building was already in TalTech campus. The transition was expected to last three years.
There is old saying - moving is one third of burning and similarly in case of merging, changes were inevitable.
All students did become students of TalTech, all study applied HE study programs become bachelor study programs of School of IT, enrolment, quality assurance, financing, real estate management etc. were taken over by university support units. Re-definition and re-learning of EIC processes were required. Differences in processes, study regulations, information flow in organization, etc. caused natural tension among teaching and support staff members.
IT College is part of School of IT of TalTech along with its 5 (classical) institutes. When institutes are responsible for balanced research and teaching, the IT College was initially intended for educating first level of bachelor studies only. The difference disappeared in 2022 when the first research group of IT College was formed followed recruiting of post-doctoral student, teaching track professors, setting up cooperation network inside and out of university (as well abroad), research projects, doctoral students and research publications.
Today’s IT College is responsible of near half of study load at School of IT, it still is using highly ranked specialists in IT filed outside of university, runs Cisco Network Academy and is teaching students (is responsible of courses) at all three higher education levels (bachelor, master and doctoral). IT College is home of university high-performance computation centre (HPC, administrative and developing activities) whereas HPC specialists are contribution in teaching and supervising of graduate thesis. IT College has active research group in IT Didactics as well centre of IT Didactics which is responsible of training teachers in university and gymnasiums.
In conclusion the merge was done in right time and turned out to be successful. Reputation of "old" EIC is still in background, graduates of EIC have reached to high position in companies, state institutions and university itself, and all this plays a role in shaping the choices of those enrolling to university.
Myocardial Perfusion Single Photon Emission Computed Tomography (SPECT) is a non-invasive diagnostic imaging technique that utilizes radioactive isotopes to create images of the heart muscle and assess blood flow within the heart. By injecting a small amount of radioactive tracer into the bloodstream, the SPECT scan provides valuable information about the heart's functioning and helps doctors diagnose conditions such as coronary artery disease. This procedure plays a crucial role in guiding treatment decisions and monitoring the effectiveness of interventions for heart-related issues. Deep Learning (DL) is a rapidly evolving field of Artificial Intelligence that has demonstrated remarkable success in image analysis and pattern recognition tasks. Its underlying neural network structures, inspired by the human brain, consist of layers of interconnected nodes that process data and progressively learn representations. The significance of this research study lies in the ability to provide timely and accurate diagnosis of coronary artery disease, enabling appropriate disease management by classifying the patients to healthy and non-healthy. The use of DL algorithms for image analysis has the potential to enhance the effectiveness and accuracy of diagnosis, reducing the need for invasive or time-consuming diagnostic procedures, while improving the quality of life for patients. Potential findings will have a significant impact on the field of cardiovascular medicine and will pave the way for further research in this area.
==> Abstract (PDF)Appropriate measures for protecting Healthcare Data are crucial when interacting with important health-related information. Data anonymization is an essential method for maintaining privacy, but it offers so many choices that the process may prove difficult, particularly for newbies. A novel app called "ShinyAnonymizer" has been designed to solve this issue. It becomes simpler as well as less costly to anonymize health data maintained in databases. The distinctive attribute of this software is its intuitive layout, allowing it easier for people of all abilities to utilize modern anonymization techniques. But it extends beyond only anonymization; it includes various strategies to data visualization and analysis. Furthermore, this study discusses hashing and encryption approaches, which improve security considerably.
In-depth descriptions of these approaches are offered to guarantee an extensive understanding of data protection strategies. The article additionally delves on an important difficulty: which approach seems the most successful to preserve the anonymity of individual data? The research findings examine anonymization, hashing, and data encryption approaches utilizing authentic medical information, displaying the benefits and drawbacks of each in real-world scenarios. The primary aspect is the fact that every technique's performance relies on its context. In acknowledging the intricate nature on life in the real world, the research emphasizes the requirement for combining approaches with specific demands as well as desired degree of information security. This research significantly improves the topic of data anonymization while at the same time offering an important addition to the wider issue of data security by offering helpful insights. It helps both individuals and organizations into taking intelligent choices for securing confidential data in an era that is growing progressively virtually.
==> Abstract (PDF)In the era of big data, our world has undergone a remarkable transformation into the digital epoch. Almost every facet of our daily existence, from photographs and texts to newspapers and books, has embraced the digital age. While these changes have brought convenience, they have also introduced significant security concerns and risks. E-books have risen in prominence due to their numerous benefits, including affordability, ease of replication, and portability. However, one major challenge associated with their widespread use is the protection of copyrights.
This approach presents the development of a web application security system designed specifically for safeguarding the copyrights of PDF e-books. Employing intelligent cryptographic algorithms and advanced AES encryption techniques, the integrity of copyright protection is fortified.The cornerstone of our approach is the AES algorihm. A cryptographic standard known for its strength and reliability. AES encryption transforms the form content into ciphertext, rendering it unreadable without the decryption key. This ensures that the sensitive intellectual property remains confidential, guarding against unauthorized access. Cryptography fortifies our solution by transforming content into a secure, unreadable format using AES.
Additionally, the implementation of steganography, specifically the Least Significant Bit (LSB) method, replacing the least significant bits of the color channels (RGB) with the corresponding bits from the encrypted data for further enhance security. This innovative approach allows for the verification of the original book purchaser, utilizing the possession of comprehensive form field data. Our methodology adopts a multi-faceted approach to safeguard intellectual property by harmonizing cryptography and steganography, we create a dynamic shield against copyright infringement. This synergy ensures both data confidentiality through encryption and covert communication via steganography.
==> Abstract (PDF)Wireless communication is being implemented in devices that are getting smaller by the day. Some need an internet connection to perform their primary task, for others this is less of a concern and rely primarily on local communication. The Internet of Things is transforming how devices interact with each other, e.g. a household setting using automations to increase comfort in daily routines. Most wireless standards rely on a separate gateway to allow interaction of these devices. This is where WiFi HaLow tries to facilitate deployment and ease of use, since a direct connection to the cloud becomes a standard feature of all equipped devices.
The WiFi HaLow 802.11ah standard is still in development, several modules are already on the market but inter-vendor communication is a work in progress. We have been building demo setups and testing out the technology to evaluate the possibilities and market-readiness for hardware and product designers The main features of this arising standard are classic IP-based communication and thus direct cloud connectivity as stated, low power consumption and security, using WPA3 as security protocol.
By building our hardware platform around a NewRacom NRC7292 based SoM, we have been able to perform power measurements and distance tests with promising results. Power consumption is clearly a work in progress, but the promised long-distance communication can be reached without major issues, with initial tests indicating a little less than 1km for single device communication using a default transmit power of 17dBm, standard omnidirectional antennas and line-of-sight. Throughput tests show that, depending on the bandwidth used, the 802.11ah standard is perfectly capable of speeds of several Mbps. This capability needs more verification to further test inter-vendor communication and check the performance and long-term stability of these connections, as well as the network stability with an increasing number of clients connected to one access point.
All in all, WiFi HaLow is a promising technology with several key features missing in other wireless standards to date. The long range and throughput capability, on top of the extra security offered by WPA3 allows end devices to communicate with the cloud directly and securely as commonly used WiFi standards are offering today but with an extra power saving feature as necessary addition for battery powered devices
==> Abstract (PDF)