VFAST - Virtual Foundation for Advancement of Science and Technology (Pakistan)
Not a member yet
1255 research outputs found
Sort by
Revolutionizing Healthcare with Smarter AI: In-depth Exploration of Advancements, Challenges, and Future Directions
Full text linkArtificial intelligence (AI) is the main branch of computer science that permits advanced machines to interpret and analyze complex healthcare data elaborating the recent challenges in the medical field of study. The current state of AI applications in healthcare is examined in this systematic literature review, with an emphasis on the technology\u27s accomplishments, difficulties, and potential. The wide breadth of AI technologies used in healthcare settings, such as robots, computer vision, machine learning, and natural language processing, is highlighted in this review through an extensive analysis of peer-reviewed publications. It talks about how customized medicine, predictive analytics, illness detection, and treatment planning are just a few of the areas of healthcare delivery that AI-driven technologies are transforming. According to research by investment bank Goldman Sachs, 300 million full-time employees could be replaced by artificial intelligence (AI). In the US and Europe, it might replace 25% of labor duties, but it might also lead to an increase in productivity and the creation of new jobs. Additionally, it might eventually result in a 7% rise in the global annual value of products and services produced. Additionally, the paper projects that approximately 25% of all employment might be performed totally by AI and that two-thirds of jobs in the U.S. and Europe "are exposed to some degree of AI automation. "The most likely groups to be impacted by workforce automation are educated white-collar workers making up to $80,000 annually, according to research from OpenAI and the University of Pennsylvania. According to a McKinsey Global Institute study, developments in digitalization, robots, and artificial intelligence may require at least 14% of workers worldwide to change jobs by 2030.
 
Generating synthetic data in biomedical imaging by designing GANs
Full text linkRecent advances in deep learning techniques have made medical analysis available with enhanced accuracy and efficiency, where brain tumor classification is automatically identified in an influential role. Hence, one of the synthesized approaches of an innovative idea to use GANs in this paper development is the synthesis of T1-weighted and post-contrast ischemic stroke brain MRIs to increase performance in the classification of the mentioned diseases according to deep learning. This paper, therefore, has the following objective: to evaluate the efficiency of GAN-generated images in learning deep in the transfer learning models and the performance in both tumor and non-tumor brain images. We use the two main architectures of GAN in our process: Vanilla and Deep Convolutional GAN (DCGAN). Details of the three major deep transfer learning models below portray the Convolutional Neural Network (CNN), MobileNetV2, and ResNet152v2. This learned weight would become a pre-trained representation of the models combined with the augmented dataset for feature extraction and classification purposes. I.e., where transfer learning is applied in the models, it is way more accessible for those architectures of the neural network to tap into the knowledge learned by the former from large-scale datasets and adapt it for tasks at hand on classifying brain tumors. Concerning training and validation, Python programming language integrated with the Keras deep learning framework was employed to implement the indicated operations. In terms of training, GPU processing power was available to allow the model to learn faster. In this regard, this was incorporated with the GPU processing by using the NVIDIA GeForce RTX 2060 GPU. Both Vanilla GAN and DCGAN have counterparts when generating images
Database Security Empowered with Independent Field Encryption
Full text linkEnsuring robust database security presents numerous challenges, necessitating a meticulously designed encryption model. Conventional single-key encryption systems like AES (Advanced Encryption Standard) and RSA (Rivest-Shamir-Adleman) are effective but frequently lack the ability to offer detailed protection and finely manage security at a granular level. These systems typically encrypt data at the file or table level, leaving individual cells vulnerable to unauthorized access once the primary key is compromised. To overcome these challenges, we suggest an Independent Field Encryption (IFE) approach. Our method encrypts databases at the cell level, offering a highly granular security mechanism where each cell is encrypted using a unique key. Leveraging the principles of modular arithmetic and Fermat’s Little Theorem for modular reduction, IFE provides superior security by minimizing the risk of data breaches through cell-specific encryption. This approach differs from industry standards by ensuring that the impact remains isolated and minimal even if one encryption key is compromised. The security strength of our proposed technique is analogous to the difficulty of reducing a high modular exponent, ensuring robust protection against cryptographic attacks. Moreover, our algorithm operates efficiently in logarithmic time, ensuring scalability and maintaining ciphertext size without enlargement. This innovative approach balances robust protection with computational efficiency, setting a new standard for database security by addressing the shortcomings of current industry-standard algorithms. Through this novel approach, IFE offers a compelling solution for enhancing database security, ensuring that each piece of data is individually protected and secure from a wide range of threats
Exploring Long-Range Order in Diblock Copolymers through Cell Dynamic Simulations
Full text linkSoft materials have played an important role in the development of nanotechnology over the past decade. Diblock copolymer systems in these soft materials have opened up new avenues of research, introducing discoveries in experimental and theoretical research in the bulk and melt states. To this end, computer programming has advanced the simulation of soft materials through mathematical models that have enabled the prediction of novel ordered structures and morphologies from simulations on long-range order. Using this approach proved to be cost-effective and time-efficient. There are many mathematical models for predicting novel morphologies in diblock copolymer systems by computer simulation. Still, cell dynamic simulation (CDS) stands out for its efficiency and robustness in achieving long-range order. This paper presents a cell dynamic simulation model for predicting simulation results by examining flow, deformation and phase transitions within diblock copolymer systems in curvilinear coordinate systems. The paper insight into the interpretation, understanding, scope, and application of the partial differential equations involved in the model by presenting a block diagram of the CDS model with a modified algorithm. A numerically consistent CDS numerical scheme is developed. Laplacian is involved in the CDS model based on curvilinear geometries to solve regular and irregular system boundaries. Also, self-assembly, phase separation mechanism, predicted results and applications in diblock copolymer systems are highlighted. Finally, the results of the CDS model are also presented for comparison with other models
A Collaborative Learning Technique for Improved Email Security
Full text linkIn the present era of common email use, the constant challenge of distinguishing between emails that are genuine and spam necessitates the adoption of complex approaches. This study evaluates a Random Forest and Naive Bayes ensemble\u27s performance in handling the difficult problem of email classification by using a voting classifier. The research uses important preprocessing techniques, such as feature selection and data integrity checks in addition to machine learning models, to ensure the validity of the analysis using real email data. Training and evaluating the collaborative learning model—a hybrid of Random Forest and Naive Bayes—focuses on key performance indicators including accuracy and classification reports. Robust techniques are used to address common problems with email data, such as missing values. In particular, our Collaborative Voting Classifier demonstrates its effectiveness as a powerful tool that enhances overall model performance by providing an equitable means of email classification. The results offer a thorough examination of memory, accuracy, and precision together with an understandable illustration made possible by confusion matrices. In this study, we assess the effectiveness of a number of classification algorithms on a particular dataset, including our proposed Voting Classifier, K-Nearest Neighbors, Gaussian Naive Bayes, and Random Forest. With considerable precision (99\%), recall (96\%), and F1-Score (95\%), the proposed Voting Classifier performs exceptionally well overall, with high accuracy (95.9\%). This study offers a thorough viewpoint for real-world classification task applications, giving insightful information about the relative advantages and disadvantages of different methods.
Order Structured Graphs of Cyclic Groups and their Classification
Full text linkLet with a cyclic group of order be an order structured graph. The group will be assumed as the vertex set of the graph and an edge between vertices will be built on the basis of a defined relation via order structure. Certain graphical parameters such as independence ratio, clique number, domination number, and separability are discussed. Some characterizations are proposed and proved by incorporating the defined relation. It is further proved that can never be a hamiltonian graph. Lastly, It is shown that is isomorphic to $\Gamma^{o}(C_{p})
Relationship between Digital Literacy Skills, Attitude towards Teaching Profession and Academic Motivation among Prospective Teachers
Full text linkAbstract
The objective of the study was to identify the correlation between the digital literacy skills, academic motivation, and attitudes towards the teaching profession among prospective teachers. The present study employed a survey research methodology and a correlational technique. The study\u27s sample consisted of 120 prospective teachers who held Bachelor of Science (BS) and Bachelor of Education (B.Ed) degrees. Data was acquired through convenient sampling technique. The dimensions of interest in this study were assessed using the Attitude toward Teaching Profession Scale (Kahramanolu, 2018), Digital Literacy Skills Scale (Üstünda et al., 2017), and Academic Motivation Scales (Karagüven, 2012). Data were analyzed through SPSS 26. The Pearson correlation revealed a statistically significant positive association between future teachers’ self-engagement and their digital literacy skills, attitude towards teaching, and academic motivation. The linear regression analysis revealed a distinct correlation between academic motivation, attitudes towards teaching, and digital literacy skills. Gender inequalities were apparent. The report suggests engaging education authorities and adopting steps for teacher training. Implementing strategies to enhance the academic motivation and digital proficiency of prospective educators.
Key concepts: Digital literacy, Attitude towards the teaching profession, Academic motivation, prospective teachers
Semi-Automated Approach for Evaluation of Software Defect Management Process using ML Approach
Full text linkEvaluation of the software development process is crucial for enhancing software production and product quality inside a company. Traditional methods that rely on manual qualitative evaluations (such as artifact inspection) are flawed because they are (i) time-consuming, (ii) hampered by authority limits, and (iii) frequently subjective. This research introduces a unique machine learning-based semi-automated method for software process assessment to get over these constraints. We specifically frame the issue as a sequence classification challenge that can be resolved using machine learning methods. We develop a new quantitative indicator to impartially assess the effectiveness and quality of a software process based on the framework. We use it to assess the defect management procedure used in four actual industrial software projects in order to verify the effectiveness of our methodology. Our empirical findings demonstrate the effectiveness and potential of our technique in offering a reliable, quantitative assessment of software process
A Comparative Analysis of Machine Learning Algorithms for Online Signature Recognition
Full text linkBiometrics recognition plays a vital role in modern human recognition and verification systems. An extensive latest research by the research community has rendered the field of biometrics inevitable for real-life applications. This research study focuses on online signature recognition. The research study is performed to identify if an online signature is genuine or forged. A novel online signature dataset, based on 1000 online signatures, has been collected from 200 participants, wherein every participant provided 5 instances of the online signature. An Android-based mobile application was developed to collect the online signature data. Moreover, a data augmentation technique was used to increase the training samples of the online signature dataset. Some common features such as the width and height of the signature, x and y coordinate values, pressure, pen ups and pen downs, total duration of the signature, etc. were extracted. The dataset has been trained and tested using machine-learning techniques. The performance of the five existing classifiers on the newly collected database has been compared. The classifiers used for training and testing included a Support Vector Machine (SVM), a Random Forest Classifier (RFC), a variant of RFC called an Extra Tree Classifier (ETC), a Decision Tree Classifier, and K-Nearest Neighbors. The performance of each classifier was evaluated in terms of precision score, recall score, and f-1 score. The RFC, and ETC classifiers gave an overall classification accuracy of 96%
Efficient Real-Time Detection of Plant Leaf Diseases Using YOLOv8 and Raspberry Pi
Full text linkThe utilization of deep learning-based models for automatic plant leaf disease detection has been established for many years. Such methods have been successfully integrated in the agriculture domain, aiding the swift and accurate identification of various diseases. However, the unavailability of annotated data, the variability of systems, and the lack of an efficient model for real-time use remain unresolved challenges. The goal of this work was to develop a deep learning-based model for crop disease detection and recognition system for real-field scenarios. For this, we trained lightweight versions of the YOLOv5, YOLOv7, YOLOv8 and compared their detection performance. Experiments were carried out on a self-collected dataset containing 3136 real-field images of apples ( healthy and diseased ) and 567 images of PlantDoc dataset. Results revealed that the prediction accuracy of YOLOv8 was superior to others on AdamW optimizer. The results were further validated by deploying it on Raspberry Pi 4