A Comparison of YOLOv5 and YOLOv8 in the Context of Mobile UI Detection
International Conference on Mobile Web and Intelligent Information Systems, 2023
Our Publications
Total: 69 publications
Featured Research
A Comparative Study on Distal Radius Fracture Detection: YOLOv8 and YOLOv11 Versus Faster R-CNN
International Conference on Mobile Web and Intelligent Information Systems, 2025
Distal radius fractures (DRFs) are among the most frequently encountered fractures in clinical practice, requiring accurate and timely diagnosis for effective treatment. Thus far, deep learning-based object detection models have shown promise in automating fracture detection, however the potential of newer architectures, such as YOLOv8 and YOLOv11, remains largely unexplored in medical imaging. This study systematically evaluates YOLOv8m, YOLOv8l, YOLOv11m, and YOLOv11l, comparing their performance against the commonly utilized Faster R-CNN model in the literature for DRF detection. A diverse dataset of wrist radiographs, including publicly available and real-world clinical images, is used for training and evaluation. The models are assessed based on precision, recall, F1-score, mean average precision (mAP), and inference speed. The results indicate that YOLOv11l outperforms all other models, achieving the highest precision of 96.5%, recall of 95.7%, and F1-score of 96.1% on the validation set, while Faster R-CNN demonstrated the worst performance. The findings set a benchmark for selecting the most effective deep learning model for automated fracture detection, which enables the integration of AI-assisted diagnosis into clinical workflows and medical education.
Deep Learning-Based Segmentation of Hysteroscopic Images for Early Detection of Endometrial Cancer
2025
Endometrial cancer (EC) is one of the most prevalent gynecological malignancies, and fast, accurate diagnosis is critical for effective treatment. The current diagnostic workflow relies heavily on hysteroscopy followed by histopathological analysis, a process that is often time-consuming and prone to subjectivity. Accordingly, this study proposes a deep learning-based diagnostic support system designed to assist in post-hysteroscopic evaluation through automated image segmentation. The proposed solution incorporates three state-of-the-art architectures, namely EfficientNet, YOLOv11x, and Mask R-CNN, trained and tested on a dataset of hysteroscopic images annotated by experienced gynecologists. Comparative performance metric results show that Efficient-Net achieves precision of 92.45%, recall of 93.17%, F1-score of 92.79%, and mAP50 of 97.36%, outperforming other models across all metrics. Furthermore, an application is developed for potential clinical workflow integration of the models. It supports both image and video input, with features such as blur detection and confidence scoring. Usability testing with medical students indicates high user satisfaction and confirms the system's clinical potential. Overall, the findings support the use of segmentation-based AI tools to reduce diagnostic delays and improve objectivity in endometrial cancer detection.
AI Assisted Customer Review Sentiment Analysis and Department Classification Tool
2024 9th International Conference on Computer Science and Engineering (UBMK), 2024
Sentiment analysis is a crucial technique in electronic commerce as it enables businesses to understand customer opinions and emotions about products or services. By analyzing customer reviews, a company can enhance a product or service, improve customer satisfaction, adjust marketing strategies, and even increase sales by addressing negative comments. Accordingly, this study proposes an artificial i ntelligence integrated Python application to perform sentiment analysis on customer reviews to determine whether the review is positive, neutral, or negative. Moreover, a department classification m odel is also proposed to match each customer review to its potential department, such as logistics, sales, etcetera. For evaluation purposes, a benchmark study is utilized. The proposed model achieves an accuracy of 93.7% while deciding the sentiment of a particular review. On the other hand, the department classification model achieves an accuracy of 93%. Furthermore, a user-friendly Python application is developed to make the models more accessible and easy to use.
AI-Enhanced Endometrial Cancer Diagnosis System
2024 9th International Conference on Computer Science and Engineering (UBMK), 2024
Endometrial cancer is the most prevalent type of uterine cancer, and early diagnosis is important for effective treatment. Traditional methods like biopsy and histopathological examination, are often time-consuming and prone to human error. Accordingly, this study proposes, implements and evaluates an Artificial Intelligence (AI)-based prototype utilizing deep learning algorithms for endometrial cancer diagnosis, which would increase the accuracy and the speed of cancer detection by acting as an assistant to the physicians. The proposed system leverages advanced object detection models, making use of multiple You Only Look Once (YOLO) versions, trained on a comprehensive dataset of hysteroscopy images annotated by physicians. Specialized AI models demonstrate high precision and recall rates, significantly enhancing diagnostic capabilities. Moreover, the developed system features a user-friendly interface that enable clinicians to easily upload images and employ existing trained models to detect endometrial abnormalities. After training the models, the results show that the YOLOv9c model achieves the highest mAP of 0.906 at IoU=0.5 and the highest precision of 0.894, and the YOLOv8s model achieves the best recall of 0.906. Also, expert evaluations confirm t he system's reliability, noting accuracy of models ranging from 94% to 98% and marking a substantial advancement in endometrial cancer diagnostics.
Enhanced Therapeutic Engagement: A Gamified Arduino Glove System for Hand Rehabilitation
2024 Innovations in Intelligent Systems and Applications Conference (ASYU), 2024
This study integrates gamification with advanced sensor technology to enhance hand rehabilitation, focusing on fine and gross motor movements. The system uses an Arduinopowered glove with potentiometers and IMUs (Inertial Measurement Unit) to accurately record finger and hand motions. Our approach stands out by incorporating an initial assessment of hand function deficits and joint mobility limitations, allowing the rehabilitation process to be tailored to individual needs. Unlike similar systems, our method specifically targets both fine motor skills (small, precise movements) and gross motor skills (larger, broader movements) through customized games, addressing the unique challenges posed by post-stroke or orthopedic trauma. The pilot study, tested in healthy individuals, showed an increase in System Usability Scale (SUS) scores from 57.87 to 69.25, indicating improved user engagement. A strong correlation was found between the glove's measurements and manual goniometer readings for the MCP (Metacarpophalangeal) index finger (r = 0.712, p < 0.001), with no significant difference in measurements (t = 1.83, p = 0.076). These findings highlight the system's potential for accurate rehabilitation, though further calibration and clinical validation are needed.
Web Camera-Based Gamified Rehabilitation System for Improving Fine Motor Skills: Pilot Study
2024
Hand function loss caused by conditions such as stroke, hand surgery, or neurological disorders significantly affects quality of life. Rehabilitation methods are essential for restoring these functions, and technology offers innovative, efficient, and accessible solutions. This study proposes a novel gamified system focusing on fine and gross motor skill assessment and improvement. The system uses the OpenCV hand tracking module with standard webcam to capture hand joint coordinates. These coordinates are utilized to calculate ROM angles and animate a virtual hand model in three Unity-based games designed for fine motor skill development. The system allows real-time interaction, accurate progress tracking, and seamless user engagement. A pilot study was conducted to evaluate usability and feasibility with 40 healthy individuals (aged 19–25 years) in a controlled environment. Additionally, a ROM angle calculation test was performed on four healthy participants (with a mean age of 23), comparing the system's measurements with traditional goniometer readings. The evaluation included joints across all fingers of the hand: distal interphalangeal (DIP), proximal interphalangeal (PIP) joints for all fingers, as well as the interphalangeal (IP) joint for the thumb. To ensure consistency, participants gripped predefined cylindrical objects of standard sizes during both measurements, reducing variability. Results showed strong correlations between the system's measurements and goniometer readings, with r-values ranging from 0.585 to 0.998. The PIP joints exhibited the strongest correlations, including 0.998 (p = 0.002) for the ring finger and 0.997 (p = 0.003) for the pinky finger. DIP joints showed high correlations such as 0.929 (p = 0.071) for the index finger and 0.883 (p = 0.117) for the ring finger. The thumb IP joint also showed a robust correlation of 0.934 (p = 0.066). These results suggest the system provides measurements comparable to traditional goniometers, particularly for PIP joints. Furthermore, user experience was also assessed using the System Usability Scale (SUS), which yielded an average score of 82.69, indicating high usability and accessibility, even for first-time users. The system's low latency (under 2 milliseconds) ensures a smooth user experience, essential for maintaining motivation during gamified rehabilitation activities. In conclusion, the proposed system is a reliable tool for hand rehabilitation, offering accurate and real-time ROM measurements and demonstrating high user usability. While some correlations did not achieve statistical significance due to the small sample size, our results show that the proposed solution can be a promising solution for home-based rehabilitation.
A Comparison of YOLOv5 and YOLOv8 in the Context of Mobile UI Detection
International Conference on Mobile Web and Intelligent Information Systems, 2023
With ever increasing technological capabilities, nowadays artificially intelligent systems solve mathematical equations, write poems and songs. However, to this day, many companies who build native mobile applications replicate their work and increase their workload by implementing the same user interface for each one of their target mobile platforms. Accordingly, this study aims to design, train and test a GUI element recognition model by utilizing the latest, state-of-the-art YOLOv8 and Roboflow Object Detection (Fast) algorithm, which then can be used to implement a multi-platform user interface generator. For evaluation purposes, a study in the same domain is set as a benchmark so that the newly obtained results can be interpreted meaningfully and put into a perspective. Accordingly, the results showed that the newly proposed YOLOv8s and YOLOv8n models have performed with a 3.32% and 1.62% better mAP respectively, than the benchmark study. On the other hand, the Roboflow Object Detection (Fast) model performed with a 1.08% lower mAP than the YOLOv5s benchmark study.
A Smartphone Application Designed To Analyze Foot Deformity: A Descriptive Pilot Research (Correlation Study)
2023
<strong>Introduction:</strong> An X-ray, CT scan, or foot analysis machines are important diagnostic tools for foot deformity. In the literature, several studies have investigated their effectiveness for correct diagnose. However, these methods cannot be used in a remote manner and patients have to spend considerable amount of time and money to make physical clinical visits. <strong>Objective: </strong>We aimed to develop a low-cost, contactless system using the smartphone application to remotely evaluate foot deformity and to investigate the correlation between the smartphone application and pedographic analysis. <strong>Method:</strong> 14 individuals (28 feet) with foot deformities were included in this study. We developed a smartphone application called ‘ArdAyak’ to evaluate the foot deformities remotely. Additionally, we collected pedographic analysis reports of patients by SIDAS custom foot analysis machine in a clinical setting to investigate the correlation with ‘ArdAyak’ application. <strong>Results:</strong> According to pedographic analysis, the percentage of 1st degree pes planus was 36, the percentage of pes cavus was found to be 29. Additionally, the Pearson Correlation Coefficient showed moderate correlation between the pedographic analysis and ArdAyak app (r=.468, 95% confidence interval [CI]= (.07-.86), p<0.05). <strong>Conclusion:</strong> The smartphone app ‘‘Ardayak’’ may have the potential to be a convenient, easy-to-use, and feasible tool for the assessment of foot deformities.
The effectiveness of “Dijital Steps” web based telerehabilitation system for patient with hindfoot pain: A randomised controlled trial
2023
Objective The aim of the study was to evaluate and compare effective therapeutic options for hindfoot pain, develop and investigate the effectiveness of tele-rehabilitation systems, and ensure patients perform their exercises and preventive measures regularly and accurately, while monitoring results. Methods Hindfoot pain (HP) patients (N= 77 with 120 feet) were admitted to this study and divided into two pathologies; Plantar Fasciitis and Achilles Tendinopathy. Patients in each pathology were randomized into three different rehabilitation …
User Acceptance of Contact Tracing Apps: A Study During the Covid-19 Pandemic
International Conference on Machine Learning, IoT and Big Data, 2023
The unpredicted coronavirus outbreak, termed COVID-19, has placed numerous governments worldwide in a difficult position. The scarcity of resources to tackle the outbreak, combined with the fear of overburdening the healthcare system, has forced most countries into a state of lockdown. Many governments have shown great interest in digital contact tracing applications that can help automate the demanding task of tracking newly infected individuals’ recent contacts. However, these apps have created a great deal of discussion, especially regarding their technology, architecture, and the adoption rate needed. This study aims to contribute to an increased understanding of the acceptance of this technology in the Norwegian population. Based on the unified theory of acceptance and use of technology (UTAUT) model, our research model incorporates the following five constructs: performance expectancy, effort expectancy, social influence, facilitating conditions, and privacy consideration. A survey was distributed amongst the Norwegian population, and the results were obtained from a sample of 258 respondents. The results from this study indicate that performance expectancy has the most significant impact on the intention to use a contact tracing application. Privacy considerations are also important, followed by effort expectancy and social influence. Facilitating conditions were found to be much less important.
GUI Element Detection from Mobile UI Images Using YOLOv5
Mobile Web and Intelligent Information Systems: 18th International Conference, MobiWIS 2022, Rome, Italy, August 22–24, 2022, Proceedings, 2022
In mobile application development, building a consistent user interface (UI) might be a costly and time-consuming process. This is especially the case if an organization has a separate team for each mobile platform such as iOS and Android. In this regard, the companies that choose the native mobile app development path end up going through do-overs as the UI work done on one platform needs to be repeated for other platforms too. One of the tedious parts of UI design tasks is creating a graphical user interface (GUI). There are numerous tools and prototypes in the literature that aim to create feasible GUI automation solutions to speed up this process and reduce the labor workload. However, as the technologies evolve and improve new versions of existing algorithms are created and offered. Accordingly, this study aims to employ the latest version of YOLO, which is YOLOv5, to create a custom object detection model that recognizes GUI elements in a given UI image. In order to benchmark the newly trained YOLOv5 GUI element detection model, existing work from the literature and their data set is considered and used for comparison purposes. Therefore, this study makes use of 450 UI samples of the VINS dataset for testing, a similar amount for validation and the rest for model training. Then the findings of this work are compared with another study that has used the SSD algorithm and VINS dataset to train, validate and test its model, which showed that proposed algorithm outperformed SSD’s mean average precision (mAP) by 15.69%.
Indoor Navigation Using Hybrid Personal Pedestrian Dead Reckoning (Hybrid P-PDR)
2022 IEEE International Conference on Internet of Things and Intelligence Systems (IoTaIS), 2022
The proliferation of smart devices has dramatically changed how people live their daily lives. Today, on top of their initial communicator role, smart devices act as guides, companions, and aids. For a long time, people have been using navigation systems and mobile phones as navigators in their cars. Indeed, there have been interests in implementing similar indoor navigation systems using technologies such as Wi-Fi, Bluetooth, and ultra-wideband. However, the proposed indoor navigation solutions were either too expensive to implement and maintain, or not accurate enough for a wider acceptance. Accordingly, this paper proposes a hybrid pedestrian dead reckoning (PDR) for indoor navigation, which utilizes the built-in sensors of smart devices. As part of this study, the authors implement three approaches to pedestrian dead-reckoning namely PDR, Personal PDR, and Hybrid P-PDR-and evaluate in a real-world setting. The findings of the the evaluation shows that the Hybrid P-PDR approach, which harnesses the user's walking pattern and signals from low-energy beacons, can navigate users in an indoor environment with a minimum of 0.77 and maximum of 1.35-meter average distance error.
LOCATION PROOFING USING VIDEO SIMILARITY
2022
With the increasing availability of mobile devices and technological improvements, location-based services have become a vital part of our everyday lives. However, a layer of checks and verifications might be required to identify the user’s declared location authenticity, since it is possible to bypass GPS or any other indoor location detection solutions. Existing location proofing solutions mainly propose techniques requiring some sort of an infrastructure, such as access points and/or beacons, or a co-located prover, witnesses and verifier formation to prove the presence of the user. This paper proposes a location proofing solution using the video similarity technique, which is based on the comparison of visual similarities in video pairs to determine the surrounding environment without any infrastructural overhead. The indoor location test results of our prototype indicate that it can achieve a verification accuracy of 97.05% on average in 9.67 seconds.
Location Proving with a Single Witness: A hybrid Prover-Witness and TOTP-based Solution
2022 Innovations in Intelligent Systems and Applications Conference (ASYU), 2022
With the increasing number of mobile and Internet of Things (IoT) devices, Location-Based Services (LBS) became an important part of our life. These applications provide value added services to users within a specific area or vicinity, which as a result may motivate users to declare counterfeit locations and benefit from them. Accordingly, this study proposes a beacon based Prover-Witness (P-W) technique to create a hardened solution for location proving with increased privacy and security in mind. Hence, the proposed solution incorporates P-W technique with time-based one time password (TOTP) mechanism to perform with only one witness in the environment. A prototype of the proposed solution is implemented and tested in a real-world environment by using one prover and one witness. The findings of the evaluation showed that the prototype managed to improve the security and privacy of the location proving process without increasing the required minimum number of witnesses.
On User Experience in The Internet of Things
2022 IEEE International Conference on Internet of Things and Intelligence Systems (IoTaIS), 2022
The IoT (Internet-of-Things) era is in full swing, with the transformation of all everyday objects extending the internet to communicate. The need to focus on UX (user experience) design of IoT services stems from the complex sequence of interactions between users from physical systems to virtual ones. This study aims to contribute to the definition of new IoT UX standards that go beyond the standard website and mobile guides that exist today. In attaining a better knowledge in the domain of UX and IoT, we focus on learning how UX designers are approaching IoT solutions and how user interactions are evolving with the use of IoT systems. Accordingly a study exploring the perspectives on the topic encompassing qualitative interviews with five UX designers and questionnaires administered to 65 IoT users was undertaken. Based on this, a series of guidelines are drafted and reviewed by UX designers to gain validity of the results produced thus creating a positive impact for not only users of IoT services but the designers behind the scenes creating the experiences.
A General Framework for Teaching Software Engineering and Improving Collaboration Skills in Multidisciplinary Teams
2021
International Engineering Accreditation (IEA) organization measure and improve the quality of the engineering education in universities. This paper presents a framework for Software Engineering Education in line with the IEA criteria. A multidisciplinary project between computer engineers and designers is presented aiming to improve students’ understanding of software processes and increase their communication skills in multidisciplinary teams. Computer engineering students follow specific software engineering processes, attend meetings and communicate via web-based project management tools. Their evaluation is made according to the IEA outcomes. As the results show, a multidisciplinary project structured according to specific IEA outcomes ensures students’ skills in software processes and give them an overall satisfaction. The methods and procedures for developing collaboration skills in multidisciplinary projects are necessary. The creation of a profession-learning seminar improved student’s communication skills thus showing that additional methods towards this direction need to be further designed and tested.
Multienvironment performance comparison of robot‐assisted indoor location estimation system
2021
Summary Many indoor location systems utilizing the fingerprinting technique require frequent signal map generation due to the dynamic properties of indoor environments. This time‐consuming task can be freed from human involvement by using various techniques. Thus, a robot‐aided signal map creator is implemented and its positioning performance as well as its mapping time is compared with a widely used human facilitated approach. For testing purposes, a low‐cost, low‐power, not off‐the‐shelf signal data collection robot is implemented. Its signal collection speed and location estimation precision is then compared with the manual method both in a small and large environment. The findings indicate that in the small environment, robot‐aided approach performs with 268‐cm error‐rate 70% of the time. On the other hand, the improved robot‐aided approach in large environment saves 52 min of labor‐intense legwork and achieves 303.5‐cm mean error‐rate.
3-D Pain Drawings–-Mobile Data Collection Using a PDA
2008
A large number of the adult population suffers from some kind of back pain during their lifetime. Part of the process of diagnosing and treating such back pain is for a clinician to collect information as to the type and location of the pain that is being suffered. Traditional approaches to gathering and visualizing this pain data have relied on simple 2-D representations of the human body, where different types of sensation are recorded with various monochrome symbols. Although patients have been shown to prefer such drawings to traditional questionnaires, these pain drawings can be limited in their ability to accurately record pain. The work described in this paper proposes an alternative that uses a 3-D representation of the human body, which can be marked in color to visualize and record the pain data. This study has shown that the new approach is a promising development in this area of medical practice and has been positively received by patients and clinicians alike.
A context-aware Tour Guide: User implications
2007
Although examples of tourist guides abound, the role of context aware feedback in such systems is an issue that has been insufficiently explored. Given the potential importance of such feedback, this paper investigates, from a usability perspective, two tour guide systems developed for Brunel University: one with context-aware user feedback and the other without. An empirical study was undertaken in which each of the applications was assessed through the prism of three usability measurements: efficiency, effectiveness, and satisfaction. Incorporating the participant feedback gathered as a result, the paper compares the use of the two applications in order to determine the impact of real-time feedback with respect to user location. Efficiency, understood as the time taken by a participant to successfully complete a task, was found to be significantly affected by the use of context-aware functionality. Effectiveness, understood as the amount of information a participant assimilated from the application, was shown not to be impacted by the provision of context-aware feedback, even though average experiment duration was found to be significantly shorter in this case. Lastly, participants' subjective satisfaction when using context-aware functionality was shown to be significantly higher than when using the non-context aware application.
A Jini-based solution for electronic prescriptions
2006
In most countries today, handwritten, paper-based medical prescriptions are the norm. While efforts have been made in the past and are being made at present to migrate toward electronic dispensation of prescriptions, these have generally omitted to incorporate ubiquitous computing technology in their proposed solutions. In this paper, we focus on this issue and describe a Jini-based prototypical solution for electronic prescriptions, which allows for their wireless transmission to in-range pharmacies and the augmentation of the service levels rendered to the user, with, for instance, information about queue lengths and estimated waiting times being provided to the patients. Clinical and user evaluation revealed that there were high levels of agreement as regards the prototype's effectiveness, ease of use, and usefulness.
A ubiquitous approach for visualizing back pain data
Computational Science and Its Applications–ICCSA 2005: International Conference, Singapore, May 9-12, 2005, Proceedings, Part II 5, 2005
We describe a wireless enabled solution for the vizualisation of back pain data. Our approach uses pain drawings to record spatial location and type of pain and enables data collection with appropriate time stamping, thus providing a means for the seldom-recorded (but often attested) time-varying nature of pain, with consequential impact on monitoring the effectiveness of patient treatment regimes. Moreover, since the implementation platform of our solution is that of a Personal Digital Assistant, data collection takes place ubiquitously, providing back pain sufferers with mobility problems (such as wheelchair users) with a convenient means of logging their pain data and of seamlessly uploading it to a hospital server using WiFi technology. Stakeholder results show that our approach is generally perceived to be an easy to use and convenient solution to the challenges of anywhere/anytime data collection.
An interdisciplinary approach in software engineering: Engineers meet designers [Yazılım Mühendisliğinde Disiplinler Arası Bir Yaklaşım: Mühendisler Tasarımcılar ile Buluşuyor]
Bu çalışmada, Yeditepe Üniversitesi Bilgisayar Mühendisliği ile Görsel İletişim ve Tasarım Bölümleri lisans öğrencileri arasındaki ortak projeyle Yazılım Mühendisliği alanında yürütülen disiplinler arası bir işbirliğinden kaynaklanan zorlukları inceliyoruz. İki bölüm arasındaki bu işbirliği iki yıl üst üste gerçekleştirilmiştir. Her yıl her iki bölümden öğrenciler de dahil olmak üzere belli sayıda grup oluşturulmuştur. Her grup, hem tasarım hem de bilgisayar programlama becerileri gerektiren bir oyun tasarlamayı hedeflemektedir. Sonuçlar, bu işbirliğinin hem öğrenciler hem de eğitmenler için faydalı olduğunu ortaya koymaktadır. Öğrencilerin bakış açısına göre, farklı geçmişlerden gelen insanlarla kurulan iletişim, onları gerçek yaşam durumlarına hazırlamaktadır. Ek olarak, bu deneyim, eğitmenlerin, öğrencilerin karşılaştıkları zorlukları yeterince anlamalarına olanak tanımaktadır ve bunun sonucu olarak, ders kalitesinin sürekli iyileştirilmesi için bir gösterge niteliğinde geri bildirim olarak kullanılmaktadır.
Automating Mobile UI Generation: A Study on Linux Mobile UI Prototyping Using YOLOv8
2025 Innovations in Intelligent Systems and Applications Conference (ASYU), 2025
As mobile device use becomes ubiquitous and smartphones are now an essential part of our daily routine, a vast number of mobile applications have been developed to simplify everyday tasks. In parallel, the variety of devices and operating systems (OS) has expanded significantly. This growing diversity has led to a sharp increase in the number of available applications and the need to create and maintain consistent user interfaces across multiple platforms. Consequently, designing graphical user interfaces (GUIs) for mobile operating systems such as iOS, Android, and Linux Mobile, which are repetitive, has become a time-consuming and costly task. This study aims to implement and evaluate an automated GUI generation tool that utilizes a VINS dataset-trained YOLOv8 object detection model to extract interface components from Android/iOS screenshots and generate GTK-compatible code for Linux mobile platforms, where crossplatform GUI development is particularly limited compared to mainstream operating systems like iOS and Android. In practice, evaluation of the prototype on a Linux Mobile operating system-postmarketOS- shows that the proposed YOLOv8-based method helps achieve faster prototyping, improves development process, provides UI consistency across platforms and reduces the overall development time. Consequently, the trained model (YOLOv8m) achieved an overall mAP of 97.1 % and over 99 % class-wide accuracy for critical elements such as InputField and Text.
Blockchain Interoperability for Future Telecoms
2025
The inherent characteristics of blockchain, including immutability, self-execution, and the removal of intermediaries, consistently generate increasing interest in its applications within the telecom sector, making it an exciting area for investment. This literature review aims to explore a promising research area known as blockchain interoperability. Interoperability seeks to connect two or more independent blockchains to effectively exchange information. Through leveraging the interoperability features of blockchain, independent telecom networks can seamlessly share information with other mobile, fixed, and next-generation networks. This results in improved security and efficiency, cost savings, and an enhanced customer experience. This study reviews highly cited research papers in the literature to assess blockchain’s relevance to telecom use cases for interoperability. Additionally, it presents prominent interoperability solutions and identifies essential requirements for the successful implementation of blockchain interoperability in the telecom sector. The findings highlight key research gaps and future directions for the adoption of blockchain in telecommunications, particularly for the forthcoming sixth generation (6G).
Analyzing Bitcoin Price Trends: News Sentiment and Financial Metrics
Joint International Conference on AI, Big Data and Blockchain, 2024
Bitcoin (BTC) is a digital currency that is gaining popularity day by day, capturing the attention of investors, analysts, and researchers due to its volatile nature and the potential for high returns. However, the prices of cryptocurrencies like BTC are susceptible to significant fluctuations and risks. As a result, the prediction of Bitcoin price movements is increasingly crucial for both individuals and organizations. Hence, in this study, two datasets are constructed where the first dataset only includes news headlines and sentiment scores while the second one incorporates additional financial metrics. Both of the datasets are trained using Logistic Regression (LR), Random Forest (RF) and Support Vector Machine (SVM) models. Accordingly, results yield that SVM model performs the best for the sentiment only dataset, whereas RF model performs the best for the dataset with financial metrics. Furthermore, the model with the financial indicators performs better classification.
Client-Server-Based Mobile Application for Skin Cancer Diagnosis Using Convolutional Neural Networks
2024 Innovations in Intelligent Systems and Applications Conference (ASYU), 2024
Skin cancer is the most prevalent cancer worldwide. Similarly, melanoma is one of the most common types of skin cancer that is highly prone to metastasis, which complicates the treatment greatly. Therefore, early detection of skin cancer, especially melanoma, is of utmost importance. However, the current widely used approach to diagnose melanoma relies on visual examination by dermatologists, which can be subjective and can lead to neglected early signs of malignancy. Considering the time-consuming nature of this process, it is a costly task to have every single mole checked by a professional. This study develops and evaluates a client-server-based mobile application designed to detect and classify moles as benign or malignant using a comprehensive dataset and two machine learning models, facilitating the early detection of malignant moles. Evaluations yield that the model accurately detects 93% of malignant moles from real-time videos taken with mobile device cameras and images selected from the gallery.
Brain Tumor Detection and Localization with YOLOv8
2023 8th International Conference on Computer Science and Engineering (UBMK), 2023
Brain tumor detection plays a crucial role in the early diagnosis as well as treatment planning of neuro-oncological conditions. Accurate localization and identification of brain tumors using magnetic resonance imaging (MRI) images are essential for guiding medical interventions. In this paper, a comprehensive approach for brain tumor detection using the BR35h dataset and the YOLOv8 algorithm is proposed. BR35h dataset consists 800 of magnetic resonance images (MRI), with tumor perimeters annotated. Hence, this study's goal is to employ the latest version of YOLO, which is YOLOv8, to create a model that locates and detects brain tumors in a given MRI image. Through various evaluations, the suggested model achieves high performance with a mean average precision (mAP) of 97.6%. Additionally, the evaluation of loss values is discussed, showcasing the model's progress in detecting and localizing brain tumors. The outcomes highlight the potential of the model as a valuable tool for accurate and efficient brain tumor detection, which may help physicians make more informed decisions efficiently.
Remote Pre-Diagnosis of Pes Planus and Pes Cavus Using Arch Index
Avrupa Bilim ve Teknoloji Dergisi, 2021
According to the literature, people with foot deformities report poor quality of life and nearly one-third of the population has some type of foot deformity. Of all the deformities, Pes Planus, caused by the loss of the medial longitudinal arch of the foot, and pes cavus, caused by having an abnormally high plantar longitudinal arch, are the ones that negatively influence the productivity of society most. In the light of the above, this study proposes a novel mobile pre-diagnosis system for pes planus and pes cavus that is utilizing conventional deformity identification methods accepted in the literature through a mobile phone app by harnesing image processing and deep neural networks. As part of the study, a prototype is implemented and tested using 34 participants - 22 (64.71%) males and 12 (35.29%) females - with an average age of 24.06. In order to benchmark our prototype, an orthopedic specialist was asked to identify the key decision making points, which was then used to calculate the deformity type, on a set of foot images collected from participants. Then the same images were fed to the prototype with the objective of identifying the key points and calculating the deformity type via the help of image processing and deep learning algorithms. The comparison of the results showed that specialist’s and prototypes findings were in 91.80% match, which indicated an overall success
Turkish Traffic Sign Recognition: Comparison of Training Step Numbers and Lighting Conditions
Avrupa Bilim ve Teknoloji Dergisi, 2021
Yollardaki araç sayısının her geçen gün artmasıyla birlikte trafik işaretleri her geçen gün daha da önem kazanmaktadır. Trafik işaretleri basit ve anlaşılması kolay olmasına rağmen, sıkışık trafikte sürücüler bunları gözden kaçırabilir. Milisaniyelerin bile kazaları önlemede büyük fark yarattığını göz önünde bulundurarak, sürücüye trafik işaretleri konusunda yardımcı olacak bir sistemin olmasının büyük bir fayda sağlayacağı oldukça açıktır. Bunun için bir trafik işareti tanıma sisteminin geliştirilmesi gerekmektedir. Bu makalede, Daha Hızlı R-CNN algoritması kullanılarak bir Türk trafik işareti tespit ve tanıma sisteminin geliştirilmesi amaçlanmaktadır. Önerilen çözüm, TensorFlow çerçevesi ile nesne algılama modelini eğitmek için Daha Hızlı R-CNN Inception-v2-COCO'yu kullanır. Modelin eğitilmesi için 54 sınıf ve 10842 adet Türk trafik işareti görüntüsünü içeren yeni bir veri seti oluşturulmuştur. Modelin eğitimi sırasıyla 51.217 ve 200.000 eğitim adım numaraları ile iki kez gerçekleştirilir. Daha sonra bu iki model kullanılarak gündüz ve gece çekilen 10 adet Türk trafik işareti görüntüsü tespit edilmeye çalışılmıştır. Sonuçlar, önerilen modellerin 51.217 eğitim adımıyla eğitildiğinde ortalama hassasiyetin %67,2 ve ortalama hatırlamanın %78,3 olduğunu göstermektedir; Öte yandan, model 200.000 eğitim adımıyla eğitildiğinde ortalama hassasiyet %76'ya ve ortalama hatırlamanın da %82,8'e yükselir.
Detecting defected crops: precision agriculture using haar classifiers and UAV
Mobile Web and Intelligent Information Systems: 16th International Conference, MobiWIS 2019, Istanbul, Turkey, August 26–28, 2019, Proceedings 16, 2019
According to recent studies, the world’s population has doubled since 1960. Furthermore, some projections indicate that the world’s population could reach more than ten billion in the next half of this century. As the world is getting increasingly crowded, the ever-growing need for resources is rising. It appears that depletion of natural resources will be three times more than current rates by the mid-century. People would not only consume more resources but also will need more agricultural produce for their everyday life. Hence, in order to meet the ever-increasing demand for farming products, yield should be maximized using top-end technologies. Precision agriculture is the application of technologies and methods to obtain data driven crop management of the farmland. In the middle of the 1980s, precision farming techniques initially were used for soil analysis using sensors and evolved to advanced applications that makes use of satellites, handheld devices and aerial vehicles. Drones commonly referred as unmanned aerial vehicles (UAVs) and have been extensively adopted in precision farming. Consequently, in the last two decades, 80 to 90% of the precision farming operations employed UAVs. Accordingly, this paper proposes a prototype UAV based solution, which can be used to hover over tomato fields, collect visual data and process them to establish meaningful information that can used by the farmers to maximize their crop. Furthermore, the findings of the proposed system showed that this was viable solution and identified the defected tomatoes with the success rate of 90%.
Evaluation of crosshair-aided gyroscope gamepad controller
Mobile Web and Intelligent Information Systems: 16th International Conference, MobiWIS 2019, Istanbul, Turkey, August 26–28, 2019, Proceedings 16, 2019
With an ever-increasing number of technological tools and gadgets in our life, people have become familiar with multiple kinds of user interaction interfaces and devices. Day in, day out people interact with various devices that follow different user interaction paradigms, such as when they are using a mobile phone, smart TV or a gaming console. Although these devices have various forms among themselves, the interaction method used to control them can make a significant difference in usability. Gaming consoles have become a huge area of the computer entertainment business in years. With every new generation of gaming consoles, the technologies behind it improve dramatically. Most of the time, the improvements are about the graphics and interaction devices. It can be clearly said that even though the graphics have improved in the last two generations of gaming consoles, the interaction paradigms and approaches were not up to the users’ expectations. This is especially the case for a first-person shooter and real-time strategy (RTS) games. Accordingly, bearing in mind the above as a motivation, the aim of this work is to develop a prototype game controller that will improve the usability and gameplay experience of the first-person shooter games.
Indoor Mapping and Positioning using Augmented Reality
2019 7th International Conference on Future Internet of Things and Cloud (FiCloud), 2019
Location-based services are becoming an important part of life and there is an increasing demand for indoor positioning. Combination of GPS and mobile cellular networks has solved the problem for outdoor environments. However, the same level of precision has not been achieved for indoor location estimation. The problem of locating an indoor environment has been studied only recently. Much research contributed to the innovative concept of an indoor positioning system. Considering the cost and the effort involved in the existing location estimation approaches, Augmented Reality (AR) based positioning method is one of the most promising methods to determine the location of a mobile device. Accordingly, in this paper, we propose, implement, and evaluate an AR-based location estimation smartphone application that can be used indoors. The evaluation results show indicate that the proposed application can estimate the location in small areas with an error-rate of 0.31 meters and in large areas with error-rate of 7.8 meters.
RoboMapper: An Automated Signal Mapping Robot for RSSI Fingerprinting
2019 7th International Conference on Future Internet of Things and Cloud (FiCloud), 2019
With the ever-increasing numbers of mobile devices, location-based services became a crucial part of mobile development. Many indoor location detection systems are developed to solve positioning problem where satellite-based solutions prone to failure. Among many proposed solutions, fingerprinting technique proved to be the most reliable approach for indoor location. However, it comes with a cost; it entails a time-consuming learning phase which should be repeated many times during the system's life time to preserve system accuracy. Thus, we propose an automated signal mapping robot called RoboMapper to alleviate time-consuming nature of the learning phase of fingerprinting technique. With the help of its accurate distance keeping mechanisms, RoboMapper can construct the signal map of the environment so that the created map can be used for user positioning. Our findings indicate that using RoboMapper 2.68-meter positioning accuracy with 70% probability can be achieved.
Mobile computing: prototype development of a context dependent location tracking system
Proceedings of the 10th International Conference on Management of Digital EcoSystems, 2018
The aim of this paper is to identify whether the development of a context-dependent location tracking prototype would be possible with technology already present in modern smartphones. To do this, a market gap analysis of the current competitors, and how the final product would differentiate from the competitors are given. Furthermore, the chosen technologies and frameworks are described briefly followed by a technical presentation of the prototype and how it works. Finally, the concept and future work are discussed. As a result, the conclusion is that the prototype presented in this paper shows that it is possible to develop a context-dependent location tracking system with technology present in modern smartphones and thus providing proof-of-concept.
Server-Based Indoor Location Detection System
Mobile Web and Intelligent Information Systems: 15th International Conference, MobiWIS 2018, Barcelona, Spain, August 6-8, 2018, Proceedings 15, 2018
With the advancement of technology and telecommunication services, data consumption rates are increasing ever since. People for long have started using applications with the help of contextual information to improve their user experience. Thus, providing a cross-platform location service to further enrich such applications has become a necessity. For this purpose, numerous client-based indoor location systems on mobile devices are developed to perform this task. Nevertheless, most of the time these systems suffer from elimination of features from operating systems for security purposes. Indeed, with the current security trends, to ensure the privacy of mobile users, mobile operating system designers are progressively eliminating certain low-level features such as reading RSSI and introducing randomized MAC addresses. Thus, in this study, the authors propose, design and implement a server-based indoor positioning system to eliminate platform dependency and to provide the location detection in wide range of devices. The designed server-based system is scalable and platform independent; hence can run on virtually any family of smart device. Furthermore, the evaluation findings indicate that the proposed system performs in acceptable accuracy to client-based systems compared to more complex and costly implementations.
Duco-hybrid indoor navigation
Mobile Web and Intelligent Information Systems: 13th International Conference, MobiWIS 2016, Vienna, Austria, August 22-24, 2016, Proceedings 13, 2016
This paper proposes an application especially designed for indoor navigation, Duco. A hybrid approach at trying to find a solution to the problem of indoor navigation by mainly utilising pedestrian dead-reckoning (PDR) along with the aid of iOS wireless location determination systems to aid the process. Using merely the digital accelerometer and compass sensors of modern smartphones, PDR can reflect location changes in real-time with high-precision while retaining battery life at maximum. An algorithm is utilised to analyse the data from these noisy sensors to enable high success rate of detecting step count. Duco also makes use of wireless location determination systems to retrieve the initial location where PDR falls short or iBeacons to get around problematic places inside an indoor venue like stairs, elevators or signal dead-zones.
inLoc: Location-aware Emergency Evacuation Assistant
16th IEEE International Conference on Computer and Information Technology (CIT 2016), 2016
Healthcare services and research centre reports estimate that by 2050, Americans aged 65 or older will reach the 89 million mark; this is more than double the number of older adults in the United States in 2010. In an aging society, it is important to home care elderly and maintain their daily needs and provide safe evacuation alternatives in the case emergency. Hence, this paper aims to implement and evaluate a smart and location-aware indoor emergency evacuation system for elderly users. The prototype system employs multiple Bluetooth Low Energy and iBeacon sensors and a smartphone. The smartphone utilizes the beacon signals to estimate its location within the building and uses the sensor data transmitted by the iBeacons to identify the level of threat (if any) to the elderly user and guide him/her to a safe zone in the case of fire, earthquake etc. Our prototype was tested in three different locations utilizing two different types of sensors. The results show that our system can detect a threat and guide the elderly user with minimum 85.47 centimeters and maximum 239.8 centimeters distance error.
InLoc: Location-aware emergency evacuation assistant
Computer and Information Technology (CIT), 2016 IEEE International Conference on, 2016
Healthcare services and research centre reports estimate that by 2050, Americans aged 65 or older will reach the 89 million mark, this is more than double the number of older adults in the United States in 2010. In an aging society, it is important to home care elderly and maintain their daily needs and provide safe evacuation alternatives in the case emergency. Hence, this paper aims to implement and evaluate a smart and location-aware indoor emergency evacuation system for elderly users. The prototype system employs multiple Bluetooth Low Energy and iBeacon sensors and a smartphone. The smartphone utilizes the beacon signals to estimate its location within the building and uses the sensor data transmitted by the iBeacons to identify the level of threat (if any) to the elderly user and guide him/her to a safe zone in the case of fire, earthquake etc. Our prototype was tested in three different locations utilizing two different types of sensors. The results show that our system can detect a threat and guide the elderly user with minimum 85.47 centimeters and maximum 239.8 centimeters distance error.
Bina İçi Yer Belirleme Saha Çalışması: Geniş Mekânda Yer İşaretleme Yöntemiyle Konum Tespitinde Sinyal Çıkış Güçlerinin Etkisinin Karşılaştırılması, XX
Türkiye’de İnternet Konferansı, İSTANBUL, 2015
Mobil teknolojinin ilerlemesiyle uygulamaların sunduğu servisleri zenginleştirmek için konum temelli bilgiye duyulan ihtiyaç giderek artmaktadır. Dış mekânda GPS gibi kendine yer edinmiş ve yüksek kesinlikli teknolojiler hali hazırda kullanılmaktadır. Fakat iç mekânda konum belirleme için halen bir standart yöntem ortaya çıkamamıştır. Bu çalışmada kablosuz ağdan (WLAN) yararlanılarak Alınan Sinyal Gücü Göstergesi (RSSI) verileriyle yer işaretleme yöntemi kullanılmış ve geniş iç mekânda konum belirleme sistemi gerçekleştirilmiştir. Bunun için iç mekâna özel kablosuz erişim noktaları (AP) yerleştirilmiştir ve mekândaki daha önceden belirli noktalarda bu erişim noktaları sinyal seviyesi kaydedilmiştir. Konum belirleme sırasında önceden kaydedilmiş sinyal seviyeleriyle kablosuz bağlantı özelliği bulunan bir tabletin anlık sinyal seviyesi karşılaştırılarak konumu tespit edilmeye çalışılmıştır. Bunun yanı sıra kablosuz erişim noktalarının sinyal çıkış güçleri düşük ve yüksek olarak iki farklı şekilde yer işaretleme yapılarak konumlandırma hata paylarına olan etkisi karşılaştırılmıştır.
Bina İçi Yer Belirlemede Sinyal Çıkış Güçlerinin Etkisinin Karşılaştırılması
inet-TR'15: Türkiye'de İnternet Konferansı, 2015
Mobil teknolojinin ilerlemesiyle uygulamaların sunduğu servisleri zenginleştirmek için konum temelli bilgiye duyulan ihtiyaç giderek artmaktadır. Dış mekânda GPS gibi kendine yer edinmiş ve yüksek kesinlikli teknolojiler hali hazırda kullanılmaktadır. Fakat iç mekânda konum belirleme için halen bir standart yöntem ortaya çıkamamıştır. Bu çalışmada kablosuz ağdan (WLAN) yararlanılarak Alınan Sinyal Gücü Göstergesi (RSSI) verileriyle yer işaretleme yöntemi kullanılmış ve geniş iç mekânda konum belirleme sistemi gerçekleştirilmiştir. Bunun için iç mekâna özel kablosuz erişim noktaları (AP) yerleştirilmiştir ve mekândaki daha önceden belirli noktalarda bu erişim noktaları sinyal seviyesi kaydedilmiştir. Konum belirleme sırasında önceden kaydedilmiş sinyal seviyeleriyle kablosuz bağlantı özelliği bulunan bir tabletin anlık sinyal seviyesi karşılaştırılarak konumu tespit edilmeye çalışılmıştır. Bunun yanı sıra kablosuz erişim noktalarının sinyal çıkış güçleri düşük ve yüksek olarak iki farklı şekilde yer işaretleme yapılarak konumlandırma hata paylarına olan etkisi karşılaştırılmıştır.
Indoor location detection with a RSS-based short term memory technique (KNN-STM)
Pervasive Computing and Communications Workshops (PERCOM Workshops), 2012 IEEE International Conference on, 2012
The interaction between devices and users has changed dramatically with the advances in mobile technologies. User friendly devices and services are offered by utilizing smart sensing capabilities and using context, location and motion sensor data. However, indoor location sensing is mostly achieved by measuring radio signal (WiFi, Bluetooth, GSM etc.) strength and nearest neighbor identification. The algorithm that is most commonly used for Received Signal Strength (RSS) based location detection is the K Nearest Neighbor (KNN). KNN algorithm identifies an estimate location using the K nearest neighboring points. Accordingly, in this paper, we aim to improve the KNN algorithm by integrating a short term memory (STM) where past signal strength readings are stored. Considering the limited movement capabilities of a mobile user in an indoor environment, user's previous locations can be taken into consideration to derive his/her current position. Hence, in the proposed approach, the signal strength readings are refined with the historical data prior to comparison with the environment's radio map. Our evaluation results indicate that the performance of enhanced KNN outperforms KNN algorithm.
Indoor location detection with a RSS-based short term memory technique (KNN-STM)
2012 IEEE International Conference on Pervasive Computing and Communications Workshops (PERCOM Workshops), 2012
The interaction between devices and users has changed dramatically with the advances in mobile technologies. User friendly devices and services are offered by utilizing smart sensing capabilities and using context, location and motion sensor data. However, indoor location sensing is mostly achieved by measuring radio signal (WiFi, Bluetooth, GSM etc.) strength and nearest neighbor identification. The algorithm that is most commonly used for Received Signal Strength (RSS) based location detection is the K Nearest Neighbor (KNN). KNN algorithm identifies an estimate location using the K nearest neighboring points. Accordingly, in this paper, we aim to improve the KNN algorithm by integrating a short term memory (STM) where past signal strength readings are stored. Considering the limited movement capabilities of a mobile user in an indoor environment, user's previous locations can be taken into consideration to derive his/her current position. Hence, in the proposed approach, the signal strength readings are refined with the historical data prior to comparison with the environment's radio map. Our evaluation results indicate that the performance of enhanced KNN outperforms KNN algorithm.
Improving RSS-Based Indoor Positioning Algorithm via K-Means Clustering
Wireless Conference 2011-Sustainable Wireless Technologies (European Wireless), 11th European, 2011
Advances in mobile technologies and devices has changed the way users interact with devices and other users. These new interaction methods and services are offered by the help of intelligent sensing capabilities, using context, location and motion sensors. However, indoor location sensing is mostly achieved by utilizing radio signal (Wi-Fi, Bluetooth, GSM etc.) and nearest neighbor identification. The most common algorithm adopted for Received Signal Strength (RSS)-based location sensing is K Nearest Neighbor (KNN), which calculates K nearest neighboring points to mobile users (MUs). Accordingly, in this paper, we aim to improve the KNN algorithm by enhancing the neighboring point selection by applying k-means clustering approach. In the proposed method, k-means clustering algorithm groups nearest neighbors according to their distance to mobile user. Then the closest group to the mobile user is used to calculate the MU's location. The evaluation results indicate that the performance of clustered KNN is closely tied to the number of clusters, number of neighbors to be clustered and the initiation of the center points in k-mean algorithm.
Back Pain Data Collection Using Scalable Vector Graphics and Geographical Information Systems
Visualizing the Semantic Web: XML-Based Internet and Information Visualization, 2006
According to a Department of Health survey, in Britain back pain affects 40% of the adult population, 5% of which have to take time off to recover (Boucher, 1999). This causes a large strain on the health system, with some 40% of back pain sufferers consulting a GP for help and 10% seeking alternative medicine therapy (Boucher, 1999). Due to the large number of people affected, back pain alone cost industry£ 9090 million in 1997/8 (Frank and De Souza, 2000), with between 90 and 100 million days of sickness and invalidity benefit paid out per year for back pain complaints (Frank and De Souza, 2000; Main, 1983; Papageorgiou et al., 1995). Back pain is not confined to the UK alone, but is a worldwide problem: in the United States, for instance, 19% of all workers’ compensation claims are made with regard to back pain. Although this is a lot less than the percentage of people affected by back pain in the UK, it should be noted that in the United States not all workers are covered by insurance and not all workers will make a claim for back pain (Jefferson and McGrath, 1996). Any improvement in the way that patients with back pain can be analyzed should therefore be viewed as one potentially capable of significantly saving both benefit expenditure and lost person-hours. The problem with back pain is that “there exist no standardised clinical tests or investigations by which all people with low back pain can be evaluated”(Papageorgiou et al., 1995). Nor will there ever be, as different people have different pain thresholds and will be affected differently. It is also difficult for medical personnel to know what has caused the back pain, as there are potentially many different causes behind it (Frank and De Souza, 2000).
Correlation between Measurements Using a Universal Goniometer and a Smartphone Application for Ankle Range of Motion
Background: Goniometers are commonly used to measure ankle range of motion (ROM).Recently, clinicians and physiotherapists can measure the ankle-ROM using smartphone applications.However, these measurement methods cannot be done remotely and body integration is required.For this reason, our study aim is to develop a smartphone application that can measure ankle-ROM remotely and to investigate the its correlation with the universal goniometer.Methods: twenty-two of healthy university students with 44 feet were recruited in the study.DijiA application was developed for Android smartphone to measure ankle dorsiflexion and plantar flexion ROM remotely.44 feet were evaluated by both universal goniometer and DijiA application.After completion of testing, all of the participant were filled out the system usability scale (SUS) survey for measuring usability of application. Results:The variation in Pearson Correlation Coefficient Between the Universal Goniometer and DijiA Smartphone App result showed that there was a moderate positive relationship Between the Universal Goniometer and DijiA ( r=.323 for DF, r=.435 for PF).Conclusions: Smartphone-based ankle ROM measurement with ''DijiA''app can be used to assess active ROM of the ankle joint without weight-bearing.The result of the study showed that usability of DijiA app is satisfactory and above the standards.
Improving RSS-Based Indoor Positioning Algorithm via K-Means Clustering
17th European Wireless 2011 - Sustainable Wireless Technologies, 2011
Advances in mobile technologies and devices has changed the way users interact with devices and other users. These new interaction methods and services are offered by the help of intelligent sensing capabilities, using context, location and motion sensors. However, indoor location sensing is mostly achieved by utilizing radio signal (Wi-Fi, Bluetooth, GSM etc.) and nearest neighbor identification. The most common algorithm adopted for Received Signal Strength (RSS)-based location sensing is K Nearest Neighbor (KNN), which calculates K nearest neighboring points to mobile users (MUs). Accordingly, in this paper, we aim to improve the KNN algorithm by enhancing the neighboring point selection by applying k-means clustering approach. In the proposed method, k-means clustering algorithm groups nearest neighbors according to their distance to mobile user. Then the closest group to the mobile user is used to calculate the MU's location. The evaluation results indicate that the performance of clustered KNN is closely tied to the number of clusters, number of neighbors to be clustered and the initiation of the center points in k-mean algorithm.
Location-aware patient record access: Patient identification using fingerprinting technique
2011 IEEE International Symposium on a World of Wireless, Mobile and Multimedia Networks, 2011
In the last decade, penetration of mobile technologies has changed our daily lives and the way we interact with each other and the rest of the world. With the latest wireless and mobile technologies, today, mobile users can find out about the congestion levels of motorways and route their trip accordingly. They can access public transport timetables on-the-move and have the data tailored dynamically based on their location. In this work, not only to improve lifestyles but increase life standards, we aim to combine the paradigms above with healthcare and hospital patient records systems. Accordingly, this paper describes a location-aware electronic health record system that can sense the location of the physician by utilizing fingerprinting technique, and retrieve the relevant patient's medical data on to the physician's mobile device. Furthermore, the system also enables medical personnel to transcribe post-it-like, audio-notes, and facilitate communication among physicians on other shifts by posting location-based notes. The prototype system's location precision and usability evaluation results indicate that the proposed system is conceived as easy to use, accurate, and efficient tool.
Location-aware patient record access: Patient identification using fingerprinting technique
2011 IEEE International Symposium on a World of Wireless, Mobile and Multimedia Networks (WoWMoM), 2011
In the last decade, penetration of mobile technologies has changed our daily lives and the way we interact with each other and the rest of the world. With the latest wireless and mobile technologies, today, mobile users can find out about the congestion levels of motorways and route their trip accordingly. They can access public transport timetables on-the-move and have the data tailored dynamically based on their location. In this work, not only to improve lifestyles but increase life standards, we aim to combine the paradigms above with healthcare and hospital patient records systems. Accordingly, this paper describes a location-aware electronic health record system that can sense the location of the physician by utilizing fingerprinting technique, and retrieve the relevant patient's medical data on to the physician's mobile device. Furthermore, the system also enables medical personnel to transcribe post-it-like, audio-notes, and facilitate communication among physicians on other shifts by posting location-based notes. The prototype system's location precision and usability evaluation results indicate that the proposed system is conceived as easy to use, accurate, and efficient tool.
Satellite-based delivery of educational content to geographically isolated communities: a service based approach
2009
Enabling learning for members of geographically isolated communities presents benefits in terms of promoting regional development and cost savings for governments and companies. However, notwithstanding recent advances in e-Learning, from both technological and pedagogical perspectives, there are very few, if any, recognised methodologies for user-led design of satellite-based e-learning infrastructures. In this paper, we present a methodology for designing a satellite and wireless based network infrastructure and learning services to support distance learning for such isolated communities. This methodology entails (a) the involvement of community members in the development of targeted learning services from an early stage, and (b) a service-oriented approach to learning solution deployment. Results show, that, while the technological premises of distance learning can be accommodated by hybrid satellite/wireless infrastructures, this has to be complemented with (a) high-quality audio–visual educational material, and (b) the opportunity for community members to interact with other community members either as groups (common-room oriented scenarios) or individuals (home-based scenarios), thus providing an impetus for learner engagement in both formal and informal activities.
Enhancing the Multimedia Tour Guide Experience
Handbook of Research on Wireless Multimedia: Quality of Service and Solutions: Quality of Service and Solutions, 2008
This chapter describes an investigation exploring user experiences of accessing streamed multimedia content, when that content is tailored according to perceptual, device and location characteristics. It builds upon the findings of our user perception evaluations by harnessing the results together to create pre-defined profiles based on QoP requirements, device type, and location for context-aware multimedia content streaming, and, in so doing, enhance the concept of context to include perceptual requirements. In the light of the findings, we propose that multimedia transmission to mobile and wireless devices should be made based on pre-defined profiles, which contains a combination of static (perceptual, device type, CPU speed, and display specifications) and dynamic information (streamed content type location of the device/user, context of the device/user). Furthermore, we believe that using profiling technology mobile service providers can effectively manage local network traffic and cut down their bandwidth costs considerably.
Mobile information access in the real world: A story of three wireless devices
2008
The importance of the user perspective to the wireless information access experience cannot be understated: simply put, users will not indulge in devices that are perceived to be difficult to use and in technologies that do not offer quality infotainment – combined information and entertainment – content. In this paper, we investigate the impact that mobile devices have on the user wireless infotainment access experience in practice. To this end, we have undertaken an empirical study placed in a ‘real-world’ setting, in which participants undertook typical infotainment access tasks on three different wireless-enabled mobile devices: a laptop, a personal digital assistant and a head mounted display device. Results show that, with the exception of participants’ level of self-consciousness when using such devices in public environments, the user wireless information access experience is generally unaffected by device type. Location was shown, though, to be a significant factor when users engage in tasks such as listening to online music or navigation. Whilst the interaction between device and environment was found to influence entertainment-related tasks in our experiments, the informational ones were not affected. However, the interaction effects between device and user type was found to affect both types of tasks. Lastly, a user’s particular computing experience was shown to influence the perceived ease of wireless information access only in the case of online searching, irrespective of whether this is done for primarily informational purposes or entertainment ones.
Satellite-enabled educational services specification and requirements analysis based on user feedback
2008
Advanced tele-education services provision in remote geographically dispersed user communities (such as agriculture and maritime), based on the specific needs and requirements of such communities, implies significant infrastructural and broadband connectivity requirements for rich media, timely and quality-assured content delivery and interactivity. The solution to broadband access anywhere is provided by satellite-enabled communication infrastructures. This paper aims to present such satellite-based infrastructures that are capable of addressing the core requirements of rich media educational services in remote areas. The paper proceeds to examine a set of services that will realise such satellite-based distance learning systems and to assess the targeted users' interest in such services. The presented work is undertaken within the framework of the EU-funded Broadband Access Satellite Enabled Education (BASE²) project. Furthermore, requirements analysis, based on the Volere template (Robertson) and on user feedback, is undertaken.
Empowering geographically isolated communities with instant access to learning and training through satellite
2007
Enabling learning for members of geographically isolated communities such as agrarian, or maritime communities presents benefits in terms of promoting regional development and cost savings for governments and companies. We present a methodology for designing a satellite and wireless based network infrastructure and learning services to support distance learning for such isolated communities. This methodology entails (a) the involvement of community members in the development of targeted learning services from an early stage and (b) a service-oriented approach to learning solution deployment. Here this methodology is applied in the context of the European research project BASE2 (BASE2 2006), in which the following two types of geographically isolated communities are considered: agrarian and maritime.
Integrating perceptual, device and location characteristics for wireless multimedia transmission
2007 2nd International Symposium on Wireless Pervasive Computing, 2007
In this paper, we describe an investigation exploring user experiences of accessing streamed multimedia content, when that content is tailored according to perceptual, device and location characteristics. To this end, we have created pre-defined transmission profiles and stream perceptually tailored multimedia content to three different locations, each characterised by different infotainment requirements. In the light of our results, we propose that multimedia transmission to mobile and wireless devices should be made based on pre-defined profiles, which contains a combination of static (perceptual, device type, CPU speed, and display specifications) and dynamic information (streamed content type location of the device/user, context of the device/user). The evaluation of such a system showed that the users and service providers can gain from such an approach considerably, as user perceptions of quality were not detrimentally affected by QoS degradations. Consequently, service providers can utilise this information to effectively manage local network traffic and bandwidth
Interactive Services and Scenario Specifications for Satellite-enabled Education
Rural Learning for Development: Experiences from Europe, 2007
Enabling learning for members of geographically isolated communities presents benefits in terms of promoting regional development and cost savings for governments and companies. However, notwithstanding recent advances in e-Learning, from both technological and pedagogical perspectives, there are very few, if any, recognised methodologies for user-led design of satellite-based e-learning infrastructures. In this paper, we present a methodology for designing a satellite and wireless based network infrastructure and learning services to support distance learning for such isolated communities. This methodology entails (a) the involvement of community members in the development of targeted learning services from an early stage, and (b) a service-oriented approach to learning solution deployment. Results show, that, while the technological premises of distance learning can be accommodated by hybrid satellite/wireless infrastructures, this has to be complemented with (a) high-quality audio–visual educational material, and (b) the opportunity for community members to interact with other community members either as groups (common-room oriented scenarios) or individuals (home-based scenarios), thus providing an impetus for learner engagement in both formal and informal activities.
Device Impact on User Mobile Infotainment Access
School of Information Systems, Computing and Mathematics, Brunel University, Uxbridge, Middlesex, contact:{Tacha. Serif| George. Ghinea}@ brunel. ac. u k, 2006
T1: Connect to the Internet using the device provided. T2: Go to www. askjeeves. co. uk website and search for shopping centres in the area. T3: Open a shopping centre’s web page and find its interior map and identify the sports shops within. T4: Find the cheapest price for Adidas Tuscany from the local shops. T5: Search on the Internet and compare the online prices with the prices in hand. T6: Send the cheapest price available to a friend via email
Electronic prescriptions: a ubiquitous solution using Jini
Pervasive Computing and Communications Workshops, 2006. PerCom Workshops 2006. Fourth Annual IEEE International Conference on, 2006
Ubiquitous computing has thus far not touched the realm of electronic prescriptions. This paper describes a Jini-based solution for electronic prescriptions, which allows for their wireless transmission to in-range pharmacies and the augmentation of the service levels rendered to the user, with, for instance, information about queue lengths and estimated waiting times being provided to patients. Clinical evaluation of the prototype has highlighted general consensus in respect of its effectiveness, ease of use and usefulness
Location based mobile computing-A tuplespace perspective
2006
Location based or ”context aware” computing is becoming increasingly recognized as a vital part of a mobile computing environment. As a consequence, the need for location‐management middleware is widely recognized and actively researched. Location‐management is frequently offered to the application through a “location API” (e.g. JSR 179) where the mobile unit can find out its own location as coordinates or as “building, floor, room” values. It is then up to the application to map the coordinates into a set of localized variables, e.g. direction to the nearest bookshop or the local timezone. It is the opinion of the authors that a localization API should be more transparent and more integrated: The localized values should be handed to the application directly, and the API for doing so should be the same as the general storage mechanisms. Our proposed middleware for location and context management is built on top of Mobispace. Mobispace is a distributed tuplespace made for mobile units (J2me) where replication between local replicas takes place with a central server (over GPRS) or with other mobile units (using Bluetooth). Since a Bluetooth connection indicates physical proximity to another node, a set of stationary nodes may distribute locality information over Bluetooth connections, and this information may be retrieved through the ordinary tuplespace API. Besides the integration with the general framework for communication and coordination the middleware offers straightforward answers to questions like: Where is node X located? Which nodes are near me? What is the trace of node Y?
Satellite-enabled Interactive Education: An Evaluation of Satellite-based Distance Learning Architectures and Network Interoperability Issues
The Institution of Engineering and Technology Seminar on Digital Video Broadcasting Over Satellite: Present and Future, 2006
Economic communities based in scattered or remote rural areas, such as agriculture and maritime, share a pronounced need for both broadband connectivity at remote geographical locations and interactive multimedia content delivering education system. Accordingly, each of these sectors also has their own special requirements with regard to the type of content needed to be delivered. Our papers reports on the networking performance and interoperability issues that raised during the evaluation two real-world distance learning …
Satellite-enabled interactive education: scenarios and systems architectures
Innovative Approaches for Learning and Knowledge Sharing: First European Conference on Technology Enhanced Learning, EC-TEL 2006 Crete, Greece, October 1-4, 2006 Proceedings 1, 2006
There are specific sectors of the economy that can benefit from satellite-based tele-education. Areas, such as maritime and agriculture, share common needs for both broadband connectivity at remote geographical areas that cannot otherwise be covered, and for innovative content for tele-education purposes. Furthermore, each area has special requirements with regard to the type of content to be delivered. In this paper we propose a set of architectural designs and case scenarios that will realise such interactive end-to-end education systems based on satellite communications. Services requirements in this setting are also identified and discussed.
The mobile information access experience-A user perspective
School of Information Systems, Computing and Mathematics, 2006
Mobile technologies, such as mobile phones, smartphones and Palmtop computers, are in an upwards trend and earliest models of such devices are already available to end-users to communicate and access multimedia content on-the-move. As a logical outcome of this development in mobile technologies and devices, content provider companies have already started investing and piloting mobile multimedia content distribution and broadcasting technologies. Nevertheless, no matter how cutting-edge technology is and no matter how stylish the mobile devices are, the ultimate success of wireless communication technologies and devices are directly associated with the user adoption and embrace of these new equipment and technologies. In this perspective, since multimedia content, for mobile or not, is ultimately produced for the education and/or enjoyment of viewers, the user's perspective concerning the presentation quality is surely of equal importance as objective Quality of Service (QoS) technical parameters, to defining distributed multimedia quality. In order to comprehensively understand user experiences whilst accessing information using mobile devices and technologies, we investigate user-mobile device interaction and look into the surrounding issues in a uniform manner by combining multiple aspects: user initial device experience (Out-of-Box Experience), mobile information access in a real-world context, device impact on user information access and perceptually tailored multimedia content impact on user information assimilation and satisfaction. Accordingly, an extensive experimental investigation has been undertaken to see how user experiences varied based on device familiarity, device type, real-world context and variable locations. The findings has shown that the overall perception, and effectively the user information access experience, is affected and improved when multimedia content is tailored according to user device type and context. Thus highlights that the future of mobile computing necessitates two-faceted research, which should combine both a user as well as a technical perspective.
Device and context influence on wireless infotainment access: a real world story
2005
User considerations are paramount when it comes to take up of technologies, and even more so in the case of mobile devices, in which the success of a particular device often depends on its novelty appeal. However, relatively little work has been undertaken exploring how day-to-day tasks are affected when mediated by such access devices. This paper reports the results of an empirical study placed in a ‘real-world’ setting, in which participants undertook typical infotainment combined information and entertainment access tasks on three different wireless-enabled mobile devices. These were a laptop, a Personal Digital Assistant and a Head Mounted Display device. Our results show that, with the exception of participants’ level of self-consciousness when using such devices in public environments, the user wireless infotainment access experience is generally unaffected by device type. Location was shown, though, to be a significant factor when users engage in tasks such as listening to online music or navigation.
HMD versus PDA: a comparative study of the user out-of-box experience
2005
The out-of-box experience (OOBE) has been identified as a significant factor contributing to user perception and acceptance of products and technologies. Whilst there has been considerable emphasis placed on formalising methodological procedures for evaluating the OOBE and on the creation of positive user experiences through appropriate interfaces and applications, relatively little work has been undertaken examining how the OOBE is impacted when the experience itself covers a range of (possibly interconnected) devices. In this paper we report the results of an empirical study which examined the OOBE when a Personal Digital Assistant (PDA) and Head Mounted Device (HMD) were configured and then connected for inter-operability purposes. Our findings show that type of device has a considerable impact on the OOBE, with the ask of interconnecting devices having a detrimental effect on the OOBE. The OOBE, however, is in main unaffected by user type and gender.
La influencia del tipo de dispositivo y del contexto en el acceso inalámbrico al infoentretenimiento: un caso real
Novática: Revista de la Asociación de Técnicos de Informática, 2005
We newly prepared para- and meta-linked alkynylpyrene oligomers and examined their photophysical properties. Oligomerization of monomeric building blocks was performed by CuI-promoted oxidative coupling reaction. The resulting oligomers mainly consist of 2-mer to 6-mer that were assigned on the basis of MALDI-TOF mass spectra, and the 2-mer, 3-mer, and 4-mer were isolated and fully characterized. From their absorption and fluorescence spectra, the para-linked oligomers were found to be somewhat pi-conjugated compared with meta-linked ones, and the fluorescence quantum yields decreased with increasing oligomer length (Phif = 0.79-0.55).
Recording of time-varying back-pain data: A wireless solution
2005
Chronic back pain is a debilitating experience for a considerable proportion of the adult population, with a significant impact on countries' economies and health systems. While there has been increasing anecdotal evidence to support the fact that for certain categories of patients (such as wheelchair users), the back pain experienced is dynamically varying with time, there is a relative scarcity of data to support and document this observation, with consequential impact upon such patients' treatment and care. Part of the reason behind this state of affairs is the relative difficulty in gathering pain measurements at precisely defined moments in time. In this paper, we describe a wireless-enabled solution that collects both questionnaire and diagrammatic, visual-based data, via a pain drawing, which overcomes such limitations, enabling seamless data collection and its upload to a hospital server using existing wireless fidelity technology. Results show that it is generally perceived to be an easy-to-use and convenient solution to the challenges of anywhere/anytime data collection.
Visualizing pain data for wheelchair users: a ubiquitous approach
2005
We describe a wireless enabled solution for the vizualisation of pain data. Our approach uses pain drawings to record spatial location and type of pain and enables data collection with appropriate time stamping, thus providing a means for the seldom-recorded (but often attested) time-varying nature of pain, with consequential impact on monitoring the effectiveness of patient treatment regimes. Moreover, since the implementation platform of our solution is that of a Personal Digital Assistant (PDA), data collection takes place ubiquitously, providing back pain sufferers with mobility problems (such as wheelchair users) with a convenient means of logging their pain data and of seamlessly uploading it to a hospital server using WiFi technology. Stakeholder results show that, notwithstanding problems related to PDA data input, our approach is generally perceived to be an easy to use and convenient solution to the challenges of anywhere/anytime data collection.
Infotainment across access devices: the perceptual impact of multimedia QoS
Proceedings of the 2004 ACM symposium on Applied computing, 2004
The introduction of multimedia on pervasive and mobile communication devices raises a number of perceptual quality issues. However, limited work has been done examining the 3-way interaction between use of equipment, user perceptual quality and quality of service. Our work measures user perceptual quality with the quality of preception (QoP) metrics which comprises levels of informational transfer (objective) and user satisfaction (subjective) when users are presented with multimedia video clips at three different frame rates, using four different display devices. Finally, our results will show that variation in frame-rate does not impact a user's level of information assimilation (IA), however, does impact a users' perception of multimedia video 'quality'.
Pervasive and standalone computing: the perceptual effects of variable multimedia quality
2004
The introduction of multimedia on pervasive and mobile communication devices raises a number of perceptual quality issues; however, limited work has been done examining the three-way interaction between use of equipment, quality of perception and quality of service. Our work measures levels of informational transfer (objective) and user satisfaction (subjective) when users are presented with multimedia video clips at three different frame rates, using four different display devices, simulating variation in participant mobility. Our …
Yazılım Mühendisliğinde Disiplinler Arası Bir Yaklaşım: Mühendisler Tasarımcılar ile Buluşuyor
Bu çalışmada, Yeditepe Üniversitesi Bilgisayar Mühendisliği ile Görsel İletişim ve Tasarım Bölümleri lisans öğrencileri arasındaki ortak projeyle Yazılım Mühendisliği alanında yürütülen disiplinler arası bir işbirliğinden kaynaklanan zorlukları inceliyoruz. İki bölüm arasındaki bu işbirliği iki yıl üst üste gerçekleştirilmiştir. Her yıl her iki bölümden öğrenciler de dahil olmak üzere belli sayıda grup oluşturulmuştur. Her grup, hem tasarım hem de bilgisayar programlama becerileri gerektiren bir oyun tasarlamayı hedeflemektedir. Sonuçlar, bu işbirliğinin hem öğrenciler hem de eğitmenler için faydalı olduğunu ortaya koymaktadır. Öğrencilerin bakış açısına göre, farklı geçmişlerden gelen insanlarla kurulan iletişim, onları gerçek yaşam durumlarına hazırlamaktadır. Ek olarak, bu deneyim, eğitmenlerin, öğrencilerin karşılaştıkları zorlukları yeterince anlamalarına olanak tanımaktadır ve bunun sonucu olarak, ders kalitesinin sürekli iyileştirilmesi için bir gösterge niteliğinde geri bildirim olarak kullanılmaktadır.