The increasing adoption of Internet of Things (IoT) devices in smart homes has transformed the way we interact with technology in our everyday lives, enabling greater convenience, automation, and efficiency. However, the proliferation of these devices has also created new security challenges, as they often serve as gateways for cyber-attacks targeting home networks. Routers play a crucial role in managing IoT device connectivity and security, yet many smart home networks are vulnerable due to weak router firmware, out dated software, and insecure configurations.
2.
Advanced Cybersecurity Measures for Embedded Systems in Critical InfrastructureSami Fayyad Ahmed, Syed Mustafa
As the integration of embedded systems into critical infrastructure grows, the need for robust cybersecurity measures becomes increasingly urgent. Embedded systems are pivotal in managing essential services such as electricity grids, transportation networks, and healthcare systems. However, these systems are often vulnerable to cyberattacks due to their specialized nature, resource constraints, and long lifecycle. This research explores advanced cybersecurity techniques tailored for embedded systems within critical infrastructure, focusing on securing both hardware and software components.
3.
The Role of Robotics Process Automation in Optimizing Data Analytics and Decision MakingSammer Ahmed A. S.
Robotic Process Automation (RPA) is rapidly transforming the way organizations approach data analytics and decision-making. By automating repetitive and rule-based tasks, RPA enhances data accuracy, reduces operational inefficiencies, and accelerates the availability of actionable insights. In the context of data analytics, RPA plays a pivotal role in automating data collection, integration, and preprocessing, enabling organizations to unlock valuable insights in real time. Furthermore, RPA contributes significantly to decision-making processes by streamlining workflows, reducing human errors, and providing decision-makers with accurate, timely information.
4.
Integrating RPA with Artificial Intelligence for Predictive Analytics in Business AutomationPraveen Srinivasan
This research explores the integration of Robotic Process Automation (RPA) with Artificial Intelligence (AI) for predictive analytics in business automation. As businesses increasingly adopt automation technologies to enhance efficiency, the synergy between RPA and AI presents a powerful opportunity to optimize decision-making and improve operational performance. RPA is designed to automate repetitive, rule-based tasks, while AI provides advanced capabilities such as machine learning and data analysis to predict future outcomes and inform business strategies. The study investigates how these technologies can work together to streamline processes, reduce costs, and enhance customer experiences.
5.
Strategies for Optimizing Database Storage Efficiency Using Bigfile Shrink Tablespaces in OracleMustafa Karim Ahmed
Efficient storage management is a critical factor in maintaining high-performance and cost-effective Oracle databases, particularly as data volumes continue to grow. The use of Bigfile tablespaces in Oracle offers a scalable approach to managing large datasets, but optimizing their storage efficiency remains a complex challenge. This research explores the potential of using Bigfile shrink tablespaces as a strategy for improving storage utilization in Oracle databases. The paper provides an in-depth analysis of Bigfile tablespaces, examining their advantages, how the SHRINK command can be applied to reclaim unused space, and the associated risks and best practices.
6.
Automating Oracle Database Performance Tuning with Robotics Process Automation (RPA)Rithik Rahmann
In today’s data-driven world, Oracle Database plays a pivotal role in supporting enterprise applications and managing large volumes of critical information. However, ensuring optimal performance in Oracle Database environments presents challenges due to complex configurations, high workloads, and the need for continuous monitoring. Traditional performance tuning techniques are often labor-intensive, prone to human error, and require extensive expertise. This research explores the integration of Robotics Process Automation (RPA) to automate the performance tuning of Oracle Databases, focusing on repetitive tasks such as query optimization, index management, and resource allocation.
7.
Securing Backup Data across Hybrid Cloud Environments: Encryption, RBAC, and Disaster Recovery Best PracticesSrinivas Mandal
Securing backup data across hybrid cloud environments presents significant challenges and requires a comprehensive approach to ensure the integrity, confidentiality, and availability of critical business information. Hybrid cloud solutions, which combine both public and private cloud infrastructures, offer flexibility but also introduce security risks, particularly when managing backup data. This research explores best practices for securing backup data in hybrid cloud environments, focusing on three core strategies: encryption, role-based access control (RBAC), and disaster recovery planning.
8.
Cloud-Based Database Management: Enhancing Security and Scalability Using RPA and AutomationKarthik Vigneswaran
Cloud-based database management has become a pivotal aspect of modern IT infrastructure, offering businesses flexibility, cost-efficiency, and scalability. However, ensuring security and scalability within these systems remains a significant challenge. This research explores the integration of Robotic Process Automation (RPA) and automation technologies to enhance the security and scalability of cloud-based databases. By automating routine tasks such as data backup, patch management, and threat detection, RPA can reduce human error, enhance response times to security threats, and optimize the allocation of resources.
9.
Cybersecurity Risks and Mitigations in Home Network Routers: Lessons from Firmware AnalysisArmand de Lambilly
The increasing reliance on home network routers for internet connectivity has made them prime targets for cyberattacks. These devices, which are often the first line of defense against external threats, can harbor significant vulnerabilities due to insecure firmware, weak default configurations, and lack of timely updates. This research investigates the cybersecurity risks associated with home network routers, with a particular focus on the analysis of their firmware. By extracting and analyzing the firmware of several popular consumer routers, this study identifies common security flaws, such as insecure default credentials, backdoors, and outdated software versions.
10.
Developing an Effective Disaster Recovery Plan Using RPA and Cloud Technologies for Database ManagementFirozi Ahmed, Ashraf Rahman
Disaster recovery planning is essential for organizations to ensure business continuity in the face of disruptions, such as data loss, system failures, or natural disasters. Traditional disaster recovery methods often require significant human intervention and can be time-consuming and error-prone. This research explores the integration of Robotic Process Automation (RPA) and Cloud Technologies to develop a more efficient and automated disaster recovery plan for database management.
The advent of Artificial Intelligence (AI) has paved the way for significant advancements in the field of precision agriculture. This research explores the integration of AI-driven technologies in optimizing crop yield prediction and resource management. Precision agriculture leverages AI algorithms, machine learning models, and sensor networks to collect and analyze data from various sources such as satellite imagery, drones, IoT sensors, and environmental data. This data-driven approach helps farmers make informed decisions on crop production, irrigation, pest control, and resource utilization, thereby improving yield outcomes and reducing wastage of valuable resources like water and fertilizers.
12.
Deep Learning for Plant Disease Detection and Early Warning SystemsMohamed Kasim
The global agricultural sector faces substantial challenges due to plant diseases, which threaten crop yield and food security. Traditional methods of disease detection, such as manual inspection and laboratory testing, are often slow, labor-intensive, and inadequate for large-scale applications. This research explores the potential of deep learning techniques, particularly Convolutional Neural Networks (CNNs), in the early detection of plant diseases. By leveraging large-scale image datasets of plant diseases, we aim to develop a deep learning-based model capable of accurately classifying plant diseases from images of plant leaves.
13.
AI-Based Autonomous Farming Equipment for Smart Harvesting and Field MonitoringAbdul Ajees Rahuman
The integration of AI-based autonomous systems into agriculture is revolutionizing farming practices by enhancing the efficiency of both harvesting and field monitoring. This research investigates the application of AI-driven technologies in autonomous farming equipment, focusing on smart harvesting and precision field monitoring.
14.
Leveraging Machine Learning for Climate-Smart Agriculture: Enhancing Productivity, Resilience, and Sustainability in the Face of Climate ChangeMohamed Imran
Climate change poses significant challenges to global agriculture, threatening food security and environmental sustainability. Climate-Smart Agriculture (CSA) aims to address these challenges by promoting practices that increase productivity, build resilience to climate impacts, and reduce greenhouse gas emissions. Machine learning (ML) has emerged as a transformative tool in supporting CSA, offering data-driven solutions for improved decision-making and resource management.
15.
IoT and AI for Smart Irrigation Systems: Enhancing Water Usage Efficiency in AgricultureHasan Harun
The agricultural sector faces significant challenges in managing water resources effectively, particularly with increasing concerns over water scarcity and climate change. Traditional irrigation systems are often inefficient, leading to overuse or waste of water. This research explores the integration of Internet of Things (IoT) and Artificial Intelligence (AI) technologies in the development of smart irrigation systems that optimize water usage while ensuring the health and productivity of crops.
16.
Cloud Migration Strategies for Mainframe Modernization: A Comparative Study of AWS, Azure, and GCPShalman Khon
Mainframe systems have long been the backbone of many large enterprises, handling critical workloads, transactions, and data processing. However, the growing demand for agility, cost efficiency, and innovation has driven many organizations to consider cloud migration as part of their mainframe modernization strategy. This research explores cloud migration strategies for mainframe systems, focusing on a comparative study of three major cloud providers: Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP).
17.
A Comparative Analysis of Cloud Migration Frameworks for Mainframe Modernization: AWS vs Azure vs GCPPraveen Srinivasan
Mainframe systems have been the backbone of many large enterprises for decades, but the rapidly evolving IT landscape, coupled with the need for greater scalability, flexibility, and cost-efficiency, is driving organizations to modernize their legacy systems. Cloud migration is a key strategy for transforming mainframe applications to align with contemporary business demands.
18.
Cost-Benefit Analysis of Cloud Migration for Mainframe Modernization: A Study of AWS, Azure, and GCPAbdul Rahuman
The increasing need for enterprises to modernize legacy mainframe systems has prompted many organizations to consider cloud platforms as a viable solution for migration. This research examines the cost-benefit landscape of migrating mainframe workloads to the three leading cloud platforms: Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP).
19.
Evaluating the Performance and Scalability of Mainframe Workloads Post-Migration to Cloud: AWS vs Azure vs GCP Shaik Rasool
The migration of mainframe workloads to the cloud is becoming a critical strategy for enterprises aiming to modernize their IT infrastructure. As organizations move away from traditional mainframes, the choice of cloud platform becomes paramount, with major providers such as Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP) offering distinct approaches for migration and workload optimization.
20.
Enhancing Cloud Backup Security with Multi-Factor Authentication, RBAC, and Encryption Protocols V. Mohanadas, V.Yazhini, A.Vasuki
Cloud computing has revolutionized data management by offering scalable, flexible, and cost-effective storage solutions. However, with the rise in data breaches and cyber-attacks, securing cloud-based backups has become a paramount concern for businesses and individuals alike. This research explores the integration of Multi-Factor Authentication (MFA), Role-Based Access Control (RBAC), and encryption protocols to enhance the security of cloud backup systems.
21.
AI-Powered Backup Security: Detecting Anomalies, Optimizing Encryption, and Enforcing RBAC PoliciesS.Vennila, K. G. Subiksha
In today’s digital landscape, securing backup systems is crucial to protecting sensitive data from breaches, corruption, or loss. While traditional security measures such as encryption and Role-Based Access Control (RBAC) have served as foundational elements in backup security, the increasing complexity of cyber threats necessitates more advanced techniques.
22.
Securing Backup and Recovery Systems for Regulatory Compliance with Encryption, MFA, and Audit TrailsV.Mohanadas, P.Vasanth
Backup and recovery systems play a critical role in ensuring business continuity, yet securing these systems in compliance with stringent regulatory frameworks remains a significant challenge. This research examines the role of encryption, multi-factor authentication (MFA), and audit trails in securing backup and recovery systems to meet regulatory compliance standards such as GDPR, HIPAA, and PCI-DSS.
23.
Building a Resilient Backup Infrastructure: Combining Data Redundancy, Encryption, and RBAC for Maximum Protection. M.Jeevanantham, B.Veerapandian
In today's digital landscape, where data is a critical asset for businesses, building a resilient backup infrastructure is paramount to ensuring data availability and protection. This research investigates the synergy between three essential elements—data redundancy, encryption, and Role-Based Access Control (RBAC)—to enhance backup system resilience.
24.
Leveraging Blockchain for Backup Integrity: Enhancing with MFA and Role-Based Access Controls B.K.Hemalatha, M.Suganthan, S.Afrosekhan
The integrity of backup systems is a critical concern in data protection strategies. Traditional backup methods are vulnerable to unauthorized access, corruption, and manipulation, which poses significant risks to organizations. This research explores the potential of leveraging blockchain technology to enhance the integrity and security of backup systems.
25.
Data Loss Prevention in Backup Systems: Integrating RBAC, Encryption, and AI-Powered Threat DetectionR.Vigneswaran, M.Ajay
In today's increasingly digital world, the protection of data from loss, theft, or corruption has become a critical concern for organizations. Backup systems, essential for data recovery, are vulnerable to various threats, including accidental deletion, cyberattacks, and insider breaches.
26.
Securing Backup Data Across Hybrid Cloud Environments: Encryption, RBAC, and Disaster Recovery Best Practices C.I.Vimalarani, N.Bhuvaneshwari
With the growing adoption of hybrid cloud environments, organizations face increased challenges in securing backup data across diverse infrastructure platforms. Hybrid clouds combine on-premises IT systems with public and private cloud services, offering flexibility but also raising concerns over data security and integrity.
27.
The Role of MFA and RBAC in Protecting Cloud-Based Backup SystemsV.Anushyadevi, V.Sriharini
As organizations increasingly migrate their critical data to the cloud, ensuring the security of cloud-based backup systems has become paramount. Cloud backups provide a robust solution for disaster recovery and data preservation, but they also introduce significant risks, including unauthorized access, data breaches, and data loss. Two essential security mechanisms—Multi-Factor Authentication (MFA) and Role-Based Access Control (RBAC)—play a crucial role in mitigating these risks.
28.
Securing Data Backup for Remote Work: RBAC, Encryption, and Multi-Factor AuthenticationS.Sangeetha, S.Swarnapriya
With the rise of remote work, the need for robust data security measures has become more critical than ever, particularly when it comes to safeguarding data backups. This research explores three key security mechanisms—Role-Based Access Control (RBAC), encryption, and Multi-Factor Authentication (MFA)—that are essential for protecting data backup systems in remote work environments.
29.
The Evolution of Backup Security: Integrating Zero Trust, Encryption, and Access Control for Data Integrity V. Merlin Freeda, C. Indra
With the increasing frequency and sophistication of cyber-attacks, ensuring the security of backup data has become a critical concern for organizations worldwide. Traditional backup security methods, often reliant on perimeter defenses and basic encryption, are no longer sufficient in the face of advanced persistent threats, insider risks, and evolving regulatory requirements.
30.
Enhancing Backup Resilience with Data Redundancy, Encryption, and Role-Based Access ControlU. Sakthi Meena
In an era where data integrity, security, and availability are critical to organizational operations, ensuring backup resilience has become an essential aspect of IT infrastructure management. This research explores the intersection of three foundational components of secure and resilient backup systems: data redundancy, encryption, and role-based access control (RBAC).
31.
Advanced Threat Detection in Backup Systems: AI, Encryption, and Role-Based Access ControlKarthick R
With the growing complexity and frequency of cyber-attacks, backup systems have become a prime target for malicious actors aiming to compromise critical data. This research explores advanced threat detection mechanisms in backup systems by focusing on the integration of Artificial Intelligence (AI), encryption, and Role-Based Access Control (RBAC) as critical layers of defense.
32.
Integrating Two-Factor Authentication and RBAC for Securing Backup and Recovery WorkflowsAkash R
The rapid increase in cyber threats has made securing backup and recovery workflows a critical component of an organization's overall cybersecurity strategy. Traditional security measures, such as passwords, are no longer sufficient to protect sensitive backup data from unauthorized access. Two-Factor Authentication (2FA) and Role-Based Access Control (RBAC) are two powerful security mechanisms that, when integrated, offer enhanced protection against data breaches.
33.
Backup System Vulnerabilities: Assessing Risks and Implementing Encryption, MFA, and RBACGokulakrishnan K
Backup systems are a critical component of modern data protection strategies, yet they are frequently targeted by cybercriminals and vulnerable to various security risks. As organizations increasingly rely on backup data for business continuity, it is essential to assess and mitigate the vulnerabilities inherent in these systems. This research examines the common security threats faced by backup systems, including ransomware, insider threats, and unauthorized access.
34.
Enhancing Data Integrity in Backup Systems Using Blockchain, MFA, and Encryption TechniquesRagavendra M K
Data integrity is a critical concern in the management and storage of backup data, as it ensures that information remains accurate, consistent, and unaltered over time. With the increasing complexity of cyber threats and the volume of data being stored, traditional backup systems face significant challenges in ensuring the authenticity and security of backed-up data.
35.
Combining Disaster Recovery, Backup Security, and Compliance: Best Practices for Encryption, RBAC, and MFA Dharshini M, Kalai Arase M
In today’s increasingly complex digital landscape, ensuring the security of disaster recovery (DR) systems and backup data has become paramount for organizations aiming to protect their critical assets and maintain business continuity. This research explores the integration of key security practices—encryption, Role-Based Access Control (RBAC), and Multi-Factor Authentication (MFA)—into disaster recovery and backup security frameworks.
36.
AI-Driven Backup Solutions: Detecting Security Threats and Enhancing Encryption and RBAC PoliciesMeena B J, Lamikaa S
In the evolving landscape of cybersecurity, backup systems play a critical role in ensuring the integrity and availability of data. However, as cyber threats continue to grow in sophistication, traditional backup security mechanisms are increasingly inadequate. This research explores the integration of artificial intelligence (AI) into backup solutions, focusing on its potential to detect security threats and enhance encryption and Role-Based Access Control (RBAC) policies.
37.
Multi-Layered Security for Backup and Recovery: Leveraging RBAC, Encryption, and MFAK Karthikeyan
In today’s digital age, data protection has become a critical aspect of any organization’s cybersecurity strategy. Backup and recovery systems, often targeted by cybercriminals, require robust security mechanisms to ensure that sensitive information is not compromised. This research explores the integration of three fundamental security components—Role-Based Access Control (RBAC), encryption, and Multi-Factor Authentication (MFA)—to create a multi-layered defense system for backup and recovery solutions.
38.
Backup Solutions for Large Enterprises: Combining Role-Based Access Control, Encryption, and Disaster Recovery Planning M Vijayakumar
In the modern enterprise environment, the protection of critical data is paramount, as large organizations face increasing risks of data breaches, cyber-attacks, and system failures. Backup solutions are essential in safeguarding against data loss, but their effectiveness is deeply influenced by the integration of security measures such as Role-Based Access Control (RBAC), encryption, and comprehensive disaster recovery planning.
39.
Securing Backup Systems in the Age of IoT: RBAC, Encryption, and Multi-Factor AuthenticationK.Bonisha
With the rapid expansion of the Internet of Things (IoT), securing backup systems has become an increasingly complex challenge. Backup systems are essential for protecting critical data, but their security is often compromised due to the growing number of connected devices, vulnerability to cyberattacks, and the inherent risks of large-scale data storage.
40.
Managing Backup Security in Distributed Systems: Encryption, RBAC, and Secure Storage PracticesK Karthikeyan , M Vijayakumar
In modern distributed systems, securing backup data is a critical component of ensuring system resilience, data integrity, and compliance with legal and regulatory standards. This research focuses on the three key pillars of backup security: encryption, Role-Based Access Control (RBAC), and secure storage practices. Encryption is vital in protecting backup data from unauthorized access, with techniques such as full-disk and file-level encryption being explored.
41.
Creating a Secure Backup Strategy for SaaS Providers: Encryption, RBAC, and Disaster Recovery FrameworksK.Bonisha
This research explores the crucial aspects of creating a secure backup strategy for Software as a Service (SaaS) providers, focusing on encryption, Role-Based Access Control (RBAC), and disaster recovery frameworks. As SaaS models continue to grow, the security and integrity of backup data become vital components in protecting against data loss, unauthorized access, and service disruptions.
42.
Multi-Tenant Backup Security: Using Role-Based Access Control, Encryption, and MFA for Shared Environments U.Ranjani ,R.Gokulnath
In the age of cloud computing and multi-tenant environments, ensuring the security of backup data is paramount. Multi-tenant systems are increasingly popular, allowing organizations to share infrastructure resources; however, this introduces unique challenges in safeguarding sensitive data.
43.
Integrating Backup and Disaster Recovery: Securing Data with MFA, RBAC, and End-to-End EncryptionR.Rajasathish
As the digital landscape continues to evolve, businesses are increasingly relying on Backup and Disaster Recovery (BDR) systems to protect critical data. However, the security of these systems is often overlooked, leaving sensitive information vulnerable to cyber threats. This research investigates the integration of Multi-Factor Authentication (MFA), Role-Based Access Control (RBAC), and End-to-End Encryption (E2E) as key security measures to enhance the resilience of BDR solutions.
44.
Data Integrity in Backup Systems: Using Blockchain, Encryption, and RBAC for Enhanced SecurityA.Suriyaprakash
The rapid growth of data has increased the need for reliable backup systems to ensure data integrity and protection against loss or unauthorized alteration. Traditional methods for securing backups often fall short in providing robust protection against sophisticated attacks, tampering, and unauthorized access. This research explores the integration of cutting-edge technologies—Blockchain, Encryption, and Role-Based Access Control (RBAC)—to enhance the security and integrity of backup systems.
45.
Multimodal Deep Learning for Early Detection of Colorectal Polyps Using Colonoscopy and Histopathological Images Kind Martin
Colorectal polyps are precursors to colorectal cancer, and their early detection can significantly reduce mortality rates. Traditional detection methods such as colonoscopy have limitations in accuracy and efficiency. In this research, we propose a novel multimodal deep learning framework for the early detection of colorectal polyps, leveraging both colonoscopy images and histopathological slides. Our approach combines the strengths of convolutional neural networks (CNNs) to analyze and integrate these complementary data sources. We investigate different fusion strategies to merge features from both imaging modalities, aiming to improve the detection performance.
46.
Transfer Learning in Colorectal Cancer Polyp Detection: Leveraging Pretrained Models from General Image Recognition to Enhance Colorectal ImagingJoel Tuvey
The detection of colorectal cancer (CRC) at an early stage significantly improves patient outcomes, making the identification of colorectal polyps a critical task in clinical practice. While traditional diagnostic methods like colonoscopy rely heavily on human expertise, automated systems leveraging deep learning techniques have shown promise in assisting with polyp detection. However, training deep learning models for medical imaging typically requires large, labeled datasets, which are often scarce in medical domains. This study explores the application of transfer learning, a technique that utilizes pretrained models from general image recognition tasks, to enhance polyp detection in colorectal cancer screening.
47.
Explainable AI for Polyp Detection: Interpreting Deep Learning Models in Colorectal Cancer DiagnosisKamal Mundey
Colorectal cancer (CRC) remains one of the leading causes of cancer-related deaths globally, with early detection playing a pivotal role in improving patient prognosis. Polyps, which can develop into cancer, are often detected through colonoscopy or other medical imaging techniques. Recent advancements in deep learning have shown promising results in automating polyp detection, offering the potential to aid healthcare professionals in identifying at-risk patients. However, the lack of interpretability in deep learning models presents a significant challenge for their widespread adoption in clinical settings. This research focuses on developing an explainable AI framework for polyp detection, aimed at enhancing the transparency and trustworthiness of deep learning models in the context of colorectal cancer diagnosis.
48.
Enhancing Liver Tumor Segmentation Using Multi-Scale Attention Mechanisms in TransUNet for CT ImagingDr. Salim Javed
Liver tumor segmentation in CT imaging remains a crucial yet challenging task for accurate diagnosis and treatment planning. Traditional deep learning models, such as U-Net and its variants, have demonstrated success in medical image segmentation, but they often struggle with complex tumor shapes, variations in size, and the presence of noise. This research presents a novel approach that enhances liver tumor segmentation by incorporating a multi-scale attention mechanism into the Transformer-based TransUNet architecture. The proposed method leverages the strengths of both Convolutional Neural Networks (CNNs) for feature extraction and Transformer-based attention mechanisms for capturing long-range dependencies and context information at multiple scales.
49.
Cross-Modality Fusion of CT and MRI for Liver Tumor Detection Using a TransUNet-Based NetworkDr. Marianne Mureithi
The early and accurate detection of liver tumors is critical for timely intervention and treatment. Multi-modality imaging, such as computed tomography (CT) and magnetic resonance imaging (MRI), offers complementary information that can enhance tumor detection. However, combining CT and MRI images for improved tumor segmentation remains a challenging task due to their inherent differences in image characteristics, resolution, and tissue contrast. In this study, we propose a novel cross-modality fusion approach for liver tumor detection using a Transformer-based U-Net (TransUNet) network. The TransUNet model is specifically designed to handle multi-modality data by leveraging transformer attention mechanisms to capture long-range dependencies and enhance the fusion of CT and MRI images.
50.
Few-Shot Learning for Liver Tumor Segmentation from CT Scans Using TransUNet with Meta-LearningKabir Umar
Accurate liver tumor segmentation in CT scans is critical for effective diagnosis and treatment planning in hepatocellular carcinoma (HCC). However, the availability of annotated datasets is often limited, making it difficult to train robust deep learning models. This research addresses the challenge of liver tumor segmentation in a few-shot learning scenario by proposing a novel approach that integrates the TransUNet architecture with meta-learning techniques. TransUNet combines the strengths of the Transformer model for capturing long-range dependencies and the U-Net for efficient segmentation in medical images.
51.
Explainability and Interpretability of Liver Tumor Segmentation Models Using TransUNet and Saliency MappingAmar Kabir
Liver tumor segmentation is a crucial task in medical image analysis, enabling clinicians to accurately assess and treat liver cancer. While deep learning models like TransUNet have shown significant promise in achieving high segmentation accuracy, the "black-box" nature of these models limits their interpretability and trust in clinical practice. This research investigates the explainability and interpretability of TransUNet-based liver tumor segmentation models by employing saliency mapping techniques. We explore methods such as Grad-CAM and Integrated Gradients to visualize and understand the decision-making process of the model. Our experimental evaluation, conducted on publicly available liver tumor datasets, demonstrates that saliency maps can effectively highlight key tumor regions that the model focuses on during segmentation.
52.
Improving Face Recognition Accuracy in Low-Resolution Images Using Deep Learning TechniquesFaruku Umar Ambursa
Face recognition technology has witnessed significant advances in recent years, but its performance degrades when applied to low-resolution images due to the loss of crucial facial features. This research explores how deep learning techniques, particularly Convolutional Neural Networks (CNNs), Super-Resolution Generative Adversarial Networks (SRGANs), and transfer learning, can be leveraged to improve face recognition accuracy in low-resolution environments. We propose a multi-stage framework that includes preprocessing for image enhancement and a custom face recognition network designed to handle degraded visual information. Experiments were conducted on several publicly available datasets, and the results demonstrate a notable improvement in accuracy compared to traditional face recognition methods, particularly when combined with image super-resolution techniques
53.
Self-Supervised Learning for Anomaly Detection in Medical Imaging DataArthur Jr.
Self-supervised learning (SSL) has emerged as a promising approach in the field of machine learning, particularly for applications involving large-scale data with limited labeled samples. In medical imaging, where acquiring labeled data can be costly and time-consuming, SSL offers a way to leverage unlabeled data for training robust models. This research investigates the use of SSL for anomaly detection in medical imaging, a critical task for identifying rare or unusual patterns such as tumors, fractures, or abnormalities. Traditional supervised methods often face challenges in handling imbalanced datasets and generalizing across diverse cases.
54.
Real-Time Object Detection and Tracking in Video Streams Using YOLO and Deep Reinforcement LearningJane Wanjiku Gitau
Object detection and tracking in video streams have become essential for a variety of real-time applications, including autonomous driving, surveillance systems, and robotics. In this research, we propose an innovative framework that combines the strengths of YOLO (You Only Look Once) for high-speed object detection with Deep Reinforcement Learning (DRL) for adaptive and robust object tracking. While YOLO excels in providing fast and accurate detection, it often struggles with handling object interactions, occlusions, and scale variations in dynamic environments. To address these challenges, we leverage DRL, which allows for continuous learning and adaptation in tracking agents, thereby enhancing the system’s ability to handle complex scenarios.
Securing IoT Devices in Smart Homes: Best Practices for Router Firmware ProtectionSarah Michelle Christy
Download
The increasing adoption of Internet of Things (IoT) devices in smart homes has transformed the way we interact with technology in our everyday lives, enabling greater convenience, automation, and efficiency. However, the proliferation of these devices has also created new security challenges, as they often serve as gateways for cyber-attacks targeting home networks. Routers play a crucial role in managing IoT device connectivity and security, yet many smart home networks are vulnerable due to weak router firmware, out dated software, and insecure configurations.
Advanced Cybersecurity Measures for Embedded Systems in Critical InfrastructureSami Fayyad Ahmed, Syed Mustafa
Download
As the integration of embedded systems into critical infrastructure grows, the need for robust cybersecurity measures becomes increasingly urgent. Embedded systems are pivotal in managing essential services such as electricity grids, transportation networks, and healthcare systems. However, these systems are often vulnerable to cyberattacks due to their specialized nature, resource constraints, and long lifecycle. This research explores advanced cybersecurity techniques tailored for embedded systems within critical infrastructure, focusing on securing both hardware and software components.
The Role of Robotics Process Automation in Optimizing Data Analytics and Decision MakingSammer Ahmed A. S.
Download
Robotic Process Automation (RPA) is rapidly transforming the way organizations approach data analytics and decision-making. By automating repetitive and rule-based tasks, RPA enhances data accuracy, reduces operational inefficiencies, and accelerates the availability of actionable insights. In the context of data analytics, RPA plays a pivotal role in automating data collection, integration, and preprocessing, enabling organizations to unlock valuable insights in real time. Furthermore, RPA contributes significantly to decision-making processes by streamlining workflows, reducing human errors, and providing decision-makers with accurate, timely information.
Integrating RPA with Artificial Intelligence for Predictive Analytics in Business AutomationPraveen Srinivasan
Download
This research explores the integration of Robotic Process Automation (RPA) with Artificial Intelligence (AI) for predictive analytics in business automation. As businesses increasingly adopt automation technologies to enhance efficiency, the synergy between RPA and AI presents a powerful opportunity to optimize decision-making and improve operational performance. RPA is designed to automate repetitive, rule-based tasks, while AI provides advanced capabilities such as machine learning and data analysis to predict future outcomes and inform business strategies. The study investigates how these technologies can work together to streamline processes, reduce costs, and enhance customer experiences.
Strategies for Optimizing Database Storage Efficiency Using Bigfile Shrink Tablespaces in OracleMustafa Karim Ahmed
Download
Efficient storage management is a critical factor in maintaining high-performance and cost-effective Oracle databases, particularly as data volumes continue to grow. The use of Bigfile tablespaces in Oracle offers a scalable approach to managing large datasets, but optimizing their storage efficiency remains a complex challenge. This research explores the potential of using Bigfile shrink tablespaces as a strategy for improving storage utilization in Oracle databases. The paper provides an in-depth analysis of Bigfile tablespaces, examining their advantages, how the SHRINK command can be applied to reclaim unused space, and the associated risks and best practices.
Automating Oracle Database Performance Tuning with Robotics Process Automation (RPA)Rithik Rahmann
Download
In today’s data-driven world, Oracle Database plays a pivotal role in supporting enterprise applications and managing large volumes of critical information. However, ensuring optimal performance in Oracle Database environments presents challenges due to complex configurations, high workloads, and the need for continuous monitoring. Traditional performance tuning techniques are often labor-intensive, prone to human error, and require extensive expertise. This research explores the integration of Robotics Process Automation (RPA) to automate the performance tuning of Oracle Databases, focusing on repetitive tasks such as query optimization, index management, and resource allocation.
Securing Backup Data across Hybrid Cloud Environments: Encryption, RBAC, and Disaster Recovery Best PracticesSrinivas
Mandal
Download
Securing backup data across hybrid cloud environments presents significant challenges and requires a comprehensive approach to ensure the integrity, confidentiality, and availability of critical business information. Hybrid cloud solutions, which combine both public and private cloud infrastructures, offer flexibility but also introduce security risks, particularly when managing backup data. This research explores best practices for securing backup data in hybrid cloud environments, focusing on three core strategies: encryption, role-based access control (RBAC), and disaster recovery planning.
Cloud-Based Database Management: Enhancing Security and Scalability Using RPA and AutomationKarthik Vigneswaran
Download
Cloud-based database management has become a pivotal aspect of modern IT infrastructure, offering businesses flexibility, cost-efficiency, and scalability. However, ensuring security and scalability within these systems remains a significant challenge. This research explores the integration of Robotic Process Automation (RPA) and automation technologies to enhance the security and scalability of cloud-based databases. By automating routine tasks such as data backup, patch management, and threat detection, RPA can reduce human error, enhance response times to security threats, and optimize the allocation of resources.
Cybersecurity Risks and Mitigations in Home Network Routers: Lessons from Firmware AnalysisArmand de Lambilly
Download
The increasing reliance on home network routers for internet connectivity has made them prime targets for cyberattacks. These devices, which are often the first line of defense against external threats, can harbor significant vulnerabilities due to insecure firmware, weak default configurations, and lack of timely updates. This research investigates the cybersecurity risks associated with home network routers, with a particular focus on the analysis of their firmware. By extracting and analyzing the firmware of several popular consumer routers, this study identifies common security flaws, such as insecure default credentials, backdoors, and outdated software versions.
Developing an Effective Disaster Recovery Plan Using RPA and Cloud Technologies for Database ManagementFirozi Ahmed,
Ashraf Rahman
Download
Disaster recovery planning is essential for organizations to ensure business continuity in the face of disruptions, such as data loss, system failures, or natural disasters. Traditional disaster recovery methods often require significant human intervention and can be time-consuming and error-prone. This research explores the integration of Robotic Process Automation (RPA) and Cloud Technologies to develop a more efficient and automated disaster recovery plan for database management.
AI-Driven Precision Agriculture: Optimizing Crop Yield Prediction and Resource ManagementAbdullah J
Download
The advent of Artificial Intelligence (AI) has paved the way for significant advancements in the field of precision agriculture. This research explores the integration of AI-driven technologies in optimizing crop yield prediction and resource management. Precision agriculture leverages AI algorithms, machine learning models, and sensor networks to collect and analyze data from various sources such as satellite imagery, drones, IoT sensors, and environmental data. This data-driven approach helps farmers make informed decisions on crop production, irrigation, pest control, and resource utilization, thereby improving yield outcomes and reducing wastage of valuable resources like water and fertilizers.
Deep Learning for Plant Disease Detection and Early Warning SystemsMohamed Kasim
Download
The global agricultural sector faces substantial challenges due to plant diseases, which threaten crop yield and food security. Traditional methods of disease detection, such as manual inspection and laboratory testing, are often slow, labor-intensive, and inadequate for large-scale applications. This research explores the potential of deep learning techniques, particularly Convolutional Neural Networks (CNNs), in the early detection of plant diseases. By leveraging large-scale image datasets of plant diseases, we aim to develop a deep learning-based model capable of accurately classifying plant diseases from images of plant leaves.
AI-Based Autonomous Farming Equipment for Smart Harvesting and Field MonitoringAbdul Ajees Rahuman
Download
The integration of AI-based autonomous systems into agriculture is revolutionizing farming practices by enhancing the efficiency of both harvesting and field monitoring. This research investigates the application of AI-driven technologies in autonomous farming equipment, focusing on smart harvesting and precision field monitoring.
Leveraging Machine Learning for Climate-Smart Agriculture: Enhancing Productivity, Resilience, and Sustainability in
the Face of Climate ChangeMohamed Imran
Download
Climate change poses significant challenges to global agriculture, threatening food security and environmental sustainability. Climate-Smart Agriculture (CSA) aims to address these challenges by promoting practices that increase productivity, build resilience to climate impacts, and reduce greenhouse gas emissions. Machine learning (ML) has emerged as a transformative tool in supporting CSA, offering data-driven solutions for improved decision-making and resource management.
IoT and AI for Smart Irrigation Systems: Enhancing Water Usage Efficiency in AgricultureHasan Harun
Download
The agricultural sector faces significant challenges in managing water resources effectively, particularly with increasing concerns over water scarcity and climate change. Traditional irrigation systems are often inefficient, leading to overuse or waste of water. This research explores the integration of Internet of Things (IoT) and Artificial Intelligence (AI) technologies in the development of smart irrigation systems that optimize water usage while ensuring the health and productivity of crops.
Cloud Migration Strategies for Mainframe Modernization: A Comparative Study of AWS, Azure, and GCPShalman Khon
Download
Mainframe systems have long been the backbone of many large enterprises, handling critical workloads, transactions, and data processing. However, the growing demand for agility, cost efficiency, and innovation has driven many organizations to consider cloud migration as part of their mainframe modernization strategy. This research explores cloud migration strategies for mainframe systems, focusing on a comparative study of three major cloud providers: Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP).
A Comparative Analysis of Cloud Migration Frameworks for Mainframe Modernization: AWS vs Azure vs GCPPraveen Srinivasan
Download
Mainframe systems have been the backbone of many large enterprises for decades, but the rapidly evolving IT landscape, coupled with the need for greater scalability, flexibility, and cost-efficiency, is driving organizations to modernize their legacy systems. Cloud migration is a key strategy for transforming mainframe applications to align with contemporary business demands.
Cost-Benefit Analysis of Cloud Migration for Mainframe Modernization: A Study of AWS, Azure, and GCPAbdul Rahuman
Download
The increasing need for enterprises to modernize legacy mainframe systems has prompted many organizations to consider cloud platforms as a viable solution for migration. This research examines the cost-benefit landscape of migrating mainframe workloads to the three leading cloud platforms: Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP).
Evaluating the Performance and Scalability of Mainframe Workloads Post-Migration to Cloud: AWS vs Azure vs GCP
Shaik Rasool
Download
The migration of mainframe workloads to the cloud is becoming a critical strategy for enterprises aiming to modernize their IT infrastructure. As organizations move away from traditional mainframes, the choice of cloud platform becomes paramount, with major providers such as Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP) offering distinct approaches for migration and workload optimization.
Enhancing Cloud Backup Security with Multi-Factor Authentication, RBAC, and Encryption Protocols
V. Mohanadas, V.Yazhini, A.Vasuki
Download
Cloud computing has revolutionized data management by offering scalable, flexible, and cost-effective storage solutions. However, with the rise in data breaches and cyber-attacks, securing cloud-based backups has become a paramount concern for businesses and individuals alike. This research explores the integration of Multi-Factor Authentication (MFA), Role-Based Access Control (RBAC), and encryption protocols to enhance the security of cloud backup systems.
AI-Powered Backup Security: Detecting Anomalies, Optimizing Encryption, and Enforcing RBAC PoliciesS.Vennila, K. G. Subiksha
Download
In today’s digital landscape, securing backup systems is crucial to protecting sensitive data from breaches, corruption, or loss. While traditional security measures such as encryption and Role-Based Access Control (RBAC) have served as foundational elements in backup security, the increasing complexity of cyber threats necessitates more advanced techniques.
Securing Backup and Recovery Systems for Regulatory Compliance with Encryption, MFA, and Audit TrailsV.Mohanadas,
P.Vasanth
Download
Backup and recovery systems play a critical role in ensuring business continuity, yet securing these systems in compliance with stringent regulatory frameworks remains a significant challenge. This research examines the role of encryption, multi-factor authentication (MFA), and audit trails in securing backup and recovery systems to meet regulatory compliance standards such as GDPR, HIPAA, and PCI-DSS.
Building a Resilient Backup Infrastructure: Combining Data Redundancy, Encryption, and RBAC for Maximum Protection.
M.Jeevanantham, B.Veerapandian
Download
In today's digital landscape, where data is a critical asset for businesses, building a resilient backup infrastructure is paramount to ensuring data availability and protection. This research investigates the synergy between three essential elements—data redundancy, encryption, and Role-Based Access Control (RBAC)—to enhance backup system resilience.
Leveraging Blockchain for Backup Integrity: Enhancing with MFA and Role-Based Access Controls
B.K.Hemalatha, M.Suganthan, S.Afrosekhan
Download
The integrity of backup systems is a critical concern in data protection strategies. Traditional backup methods are vulnerable to unauthorized access, corruption, and manipulation, which poses significant risks to organizations. This research explores the potential of leveraging blockchain technology to enhance the integrity and security of backup systems.
Data Loss Prevention in Backup Systems: Integrating RBAC, Encryption, and AI-Powered Threat DetectionR.Vigneswaran, M.Ajay
Download
In today's increasingly digital world, the protection of data from loss, theft, or corruption has become a critical concern for organizations. Backup systems, essential for data recovery, are vulnerable to various threats, including accidental deletion, cyberattacks, and insider breaches.
Securing Backup Data Across Hybrid Cloud Environments: Encryption, RBAC, and Disaster Recovery Best Practices
C.I.Vimalarani, N.Bhuvaneshwari
Download
With the growing adoption of hybrid cloud environments, organizations face increased challenges in securing backup data across diverse infrastructure platforms. Hybrid clouds combine on-premises IT systems with public and private cloud services, offering flexibility but also raising concerns over data security and integrity.
The Role of MFA and RBAC in Protecting Cloud-Based Backup SystemsV.Anushyadevi, V.Sriharini
Download
As organizations increasingly migrate their critical data to the cloud, ensuring the security of cloud-based backup systems has become paramount. Cloud backups provide a robust solution for disaster recovery and data preservation, but they also introduce significant risks, including unauthorized access, data breaches, and data loss. Two essential security mechanisms—Multi-Factor Authentication (MFA) and Role-Based Access Control (RBAC)—play a crucial role in mitigating these risks.
Securing Data Backup for Remote Work: RBAC, Encryption, and Multi-Factor AuthenticationS.Sangeetha, S.Swarnapriya
Download
With the rise of remote work, the need for robust data security measures has become more critical than ever, particularly when it comes to safeguarding data backups. This research explores three key security mechanisms—Role-Based Access Control (RBAC), encryption, and Multi-Factor Authentication (MFA)—that are essential for protecting data backup systems in remote work environments.
The Evolution of Backup Security: Integrating Zero Trust, Encryption, and Access Control for Data Integrity
V. Merlin Freeda, C. Indra
Download
With the increasing frequency and sophistication of cyber-attacks, ensuring the security of backup data has become a critical concern for organizations worldwide. Traditional backup security methods, often reliant on perimeter defenses and basic encryption, are no longer sufficient in the face of advanced persistent threats, insider risks, and evolving regulatory requirements.
Enhancing Backup Resilience with Data Redundancy, Encryption, and Role-Based Access ControlU. Sakthi Meena
Download
In an era where data integrity, security, and availability are critical to organizational operations, ensuring backup resilience has become an essential aspect of IT infrastructure management. This research explores the intersection of three foundational components of secure and resilient backup systems: data redundancy, encryption, and role-based access control (RBAC).
Advanced Threat Detection in Backup Systems: AI, Encryption, and Role-Based Access ControlKarthick R
Download
With the growing complexity and frequency of cyber-attacks, backup systems have become a prime target for malicious actors aiming to compromise critical data. This research explores advanced threat detection mechanisms in backup systems by focusing on the integration of Artificial Intelligence (AI), encryption, and Role-Based Access Control (RBAC) as critical layers of defense.
Integrating Two-Factor Authentication and RBAC for Securing Backup and Recovery WorkflowsAkash R
Download
The rapid increase in cyber threats has made securing backup and recovery workflows a critical component of an organization's overall cybersecurity strategy. Traditional security measures, such as passwords, are no longer sufficient to protect sensitive backup data from unauthorized access. Two-Factor Authentication (2FA) and Role-Based Access Control (RBAC) are two powerful security mechanisms that, when integrated, offer enhanced protection against data breaches.
Backup System Vulnerabilities: Assessing Risks and Implementing Encryption, MFA, and RBACGokulakrishnan K
Download
Backup systems are a critical component of modern data protection strategies, yet they are frequently targeted by cybercriminals and vulnerable to various security risks. As organizations increasingly rely on backup data for business continuity, it is essential to assess and mitigate the vulnerabilities inherent in these systems. This research examines the common security threats faced by backup systems, including ransomware, insider threats, and unauthorized access.
Enhancing Data Integrity in Backup Systems Using Blockchain, MFA, and Encryption TechniquesRagavendra M K
Download
Data integrity is a critical concern in the management and storage of backup data, as it ensures that information remains accurate, consistent, and unaltered over time. With the increasing complexity of cyber threats and the volume of data being stored, traditional backup systems face significant challenges in ensuring the authenticity and security of backed-up data.
Combining Disaster Recovery, Backup Security, and Compliance: Best Practices for Encryption, RBAC, and MFA
Dharshini M, Kalai Arase M
Download
In today’s increasingly complex digital landscape, ensuring the security of disaster recovery (DR) systems and backup data has become paramount for organizations aiming to protect their critical assets and maintain business continuity. This research explores the integration of key security practices—encryption, Role-Based Access Control (RBAC), and Multi-Factor Authentication (MFA)—into disaster recovery and backup security frameworks.
AI-Driven Backup Solutions: Detecting Security Threats and Enhancing Encryption and RBAC PoliciesMeena B J, Lamikaa S
Download
In the evolving landscape of cybersecurity, backup systems play a critical role in ensuring the integrity and availability of data. However, as cyber threats continue to grow in sophistication, traditional backup security mechanisms are increasingly inadequate. This research explores the integration of artificial intelligence (AI) into backup solutions, focusing on its potential to detect security threats and enhance encryption and Role-Based Access Control (RBAC) policies.
Multi-Layered Security for Backup and Recovery: Leveraging RBAC, Encryption, and MFAK Karthikeyan
Download
In today’s digital age, data protection has become a critical aspect of any organization’s cybersecurity strategy. Backup and recovery systems, often targeted by cybercriminals, require robust security mechanisms to ensure that sensitive information is not compromised. This research explores the integration of three fundamental security components—Role-Based Access Control (RBAC), encryption, and Multi-Factor Authentication (MFA)—to create a multi-layered defense system for backup and recovery solutions.
Backup Solutions for Large Enterprises: Combining Role-Based Access Control, Encryption, and Disaster Recovery Planning
M Vijayakumar
Download
In the modern enterprise environment, the protection of critical data is paramount, as large organizations face increasing risks of data breaches, cyber-attacks, and system failures. Backup solutions are essential in safeguarding against data loss, but their effectiveness is deeply influenced by the integration of security measures such as Role-Based Access Control (RBAC), encryption, and comprehensive disaster recovery planning.
Securing Backup Systems in the Age of IoT: RBAC, Encryption, and Multi-Factor AuthenticationK.Bonisha
Download
With the rapid expansion of the Internet of Things (IoT), securing backup systems has become an increasingly complex challenge. Backup systems are essential for protecting critical data, but their security is often compromised due to the growing number of connected devices, vulnerability to cyberattacks, and the inherent risks of large-scale data storage.
Managing Backup Security in Distributed Systems: Encryption, RBAC, and Secure Storage PracticesK Karthikeyan , M Vijayakumar
Download
In modern distributed systems, securing backup data is a critical component of ensuring system resilience, data integrity, and compliance with legal and regulatory standards. This research focuses on the three key pillars of backup security: encryption, Role-Based Access Control (RBAC), and secure storage practices. Encryption is vital in protecting backup data from unauthorized access, with techniques such as full-disk and file-level encryption being explored.
Creating a Secure Backup Strategy for SaaS Providers: Encryption, RBAC, and Disaster Recovery FrameworksK.Bonisha
Download
This research explores the crucial aspects of creating a secure backup strategy for Software as a Service (SaaS) providers, focusing on encryption, Role-Based Access Control (RBAC), and disaster recovery frameworks. As SaaS models continue to grow, the security and integrity of backup data become vital components in protecting against data loss, unauthorized access, and service disruptions.
Multi-Tenant Backup Security: Using Role-Based Access Control, Encryption, and MFA for Shared Environments
U.Ranjani ,R.Gokulnath
Download
In the age of cloud computing and multi-tenant environments, ensuring the security of backup data is paramount. Multi-tenant systems are increasingly popular, allowing organizations to share infrastructure resources; however, this introduces unique challenges in safeguarding sensitive data.
Integrating Backup and Disaster Recovery: Securing Data with MFA, RBAC, and End-to-End EncryptionR.Rajasathish
Download
As the digital landscape continues to evolve, businesses are increasingly relying on Backup and Disaster Recovery (BDR) systems to protect critical data. However, the security of these systems is often overlooked, leaving sensitive information vulnerable to cyber threats. This research investigates the integration of Multi-Factor Authentication (MFA), Role-Based Access Control (RBAC), and End-to-End Encryption (E2E) as key security measures to enhance the resilience of BDR solutions.
Data Integrity in Backup Systems: Using Blockchain, Encryption, and RBAC for Enhanced SecurityA.Suriyaprakash
Download
The rapid growth of data has increased the need for reliable backup systems to ensure data integrity and protection against loss or unauthorized alteration. Traditional methods for securing backups often fall short in providing robust protection against sophisticated attacks, tampering, and unauthorized access. This research explores the integration of cutting-edge technologies—Blockchain, Encryption, and Role-Based Access Control (RBAC)—to enhance the security and integrity of backup systems.
Multimodal Deep Learning for Early Detection of Colorectal Polyps Using Colonoscopy and Histopathological Images
Kind Martin
Download
Colorectal polyps are precursors to colorectal cancer, and their early detection can significantly reduce mortality rates. Traditional detection methods such as colonoscopy have limitations in accuracy and efficiency. In this research, we propose a novel multimodal deep learning framework for the early detection of colorectal polyps, leveraging both colonoscopy images and histopathological slides. Our approach combines the strengths of convolutional neural networks (CNNs) to analyze and integrate these complementary data sources. We investigate different fusion strategies to merge features from both imaging modalities, aiming to improve the detection performance.
Transfer Learning in Colorectal Cancer Polyp Detection: Leveraging Pretrained Models from General Image Recognition to
Enhance Colorectal ImagingJoel Tuvey
Download
The detection of colorectal cancer (CRC) at an early stage significantly improves patient outcomes, making the identification of colorectal polyps a critical task in clinical practice. While traditional diagnostic methods like colonoscopy rely heavily on human expertise, automated systems leveraging deep learning techniques have shown promise in assisting with polyp detection. However, training deep learning models for medical imaging typically requires large, labeled datasets, which are often scarce in medical domains. This study explores the application of transfer learning, a technique that utilizes pretrained models from general image recognition tasks, to enhance polyp detection in colorectal cancer screening.
Explainable AI for Polyp Detection: Interpreting Deep Learning Models in Colorectal Cancer DiagnosisKamal Mundey
Download
Colorectal cancer (CRC) remains one of the leading causes of cancer-related deaths globally, with early detection playing a pivotal role in improving patient prognosis. Polyps, which can develop into cancer, are often detected through colonoscopy or other medical imaging techniques. Recent advancements in deep learning have shown promising results in automating polyp detection, offering the potential to aid healthcare professionals in identifying at-risk patients. However, the lack of interpretability in deep learning models presents a significant challenge for their widespread adoption in clinical settings. This research focuses on developing an explainable AI framework for polyp detection, aimed at enhancing the transparency and trustworthiness of deep learning models in the context of colorectal cancer diagnosis.
Enhancing Liver Tumor Segmentation Using Multi-Scale Attention Mechanisms in TransUNet for CT ImagingDr. Salim Javed
Download
Liver tumor segmentation in CT imaging remains a crucial yet challenging task for accurate diagnosis and treatment planning. Traditional deep learning models, such as U-Net and its variants, have demonstrated success in medical image segmentation, but they often struggle with complex tumor shapes, variations in size, and the presence of noise. This research presents a novel approach that enhances liver tumor segmentation by incorporating a multi-scale attention mechanism into the Transformer-based TransUNet architecture. The proposed method leverages the strengths of both Convolutional Neural Networks (CNNs) for feature extraction and Transformer-based attention mechanisms for capturing long-range dependencies and context information at multiple scales.
Cross-Modality Fusion of CT and MRI for Liver Tumor Detection Using a TransUNet-Based NetworkDr. Marianne Mureithi
Download
The early and accurate detection of liver tumors is critical for timely intervention and treatment. Multi-modality imaging, such as computed tomography (CT) and magnetic resonance imaging (MRI), offers complementary information that can enhance tumor detection. However, combining CT and MRI images for improved tumor segmentation remains a challenging task due to their inherent differences in image characteristics, resolution, and tissue contrast. In this study, we propose a novel cross-modality fusion approach for liver tumor detection using a Transformer-based U-Net (TransUNet) network. The TransUNet model is specifically designed to handle multi-modality data by leveraging transformer attention mechanisms to capture long-range dependencies and enhance the fusion of CT and MRI images.
Few-Shot Learning for Liver Tumor Segmentation from CT Scans Using TransUNet with Meta-LearningKabir Umar
Download
Accurate liver tumor segmentation in CT scans is critical for effective diagnosis and treatment planning in hepatocellular carcinoma (HCC). However, the availability of annotated datasets is often limited, making it difficult to train robust deep learning models. This research addresses the challenge of liver tumor segmentation in a few-shot learning scenario by proposing a novel approach that integrates the TransUNet architecture with meta-learning techniques. TransUNet combines the strengths of the Transformer model for capturing long-range dependencies and the U-Net for efficient segmentation in medical images.
Explainability and Interpretability of Liver Tumor Segmentation Models Using TransUNet and Saliency MappingAmar Kabir
Download
Liver tumor segmentation is a crucial task in medical image analysis, enabling clinicians to accurately assess and treat liver cancer. While deep learning models like TransUNet have shown significant promise in achieving high segmentation accuracy, the "black-box" nature of these models limits their interpretability and trust in clinical practice. This research investigates the explainability and interpretability of TransUNet-based liver tumor segmentation models by employing saliency mapping techniques. We explore methods such as Grad-CAM and Integrated Gradients to visualize and understand the decision-making process of the model. Our experimental evaluation, conducted on publicly available liver tumor datasets, demonstrates that saliency maps can effectively highlight key tumor regions that the model focuses on during segmentation.
Improving Face Recognition Accuracy in Low-Resolution Images Using Deep Learning TechniquesFaruku Umar Ambursa
Download
Face recognition technology has witnessed significant advances in recent years, but its performance degrades when applied to low-resolution images due to the loss of crucial facial features. This research explores how deep learning techniques, particularly Convolutional Neural Networks (CNNs), Super-Resolution Generative Adversarial Networks (SRGANs), and transfer learning, can be leveraged to improve face recognition accuracy in low-resolution environments. We propose a multi-stage framework that includes preprocessing for image enhancement and a custom face recognition network designed to handle degraded visual information. Experiments were conducted on several publicly available datasets, and the results demonstrate a notable improvement in accuracy compared to traditional face recognition methods, particularly when combined with image super-resolution techniques
Self-Supervised Learning for Anomaly Detection in Medical Imaging DataArthur Jr.
Download
Self-supervised learning (SSL) has emerged as a promising approach in the field of machine learning, particularly for applications involving large-scale data with limited labeled samples. In medical imaging, where acquiring labeled data can be costly and time-consuming, SSL offers a way to leverage unlabeled data for training robust models. This research investigates the use of SSL for anomaly detection in medical imaging, a critical task for identifying rare or unusual patterns such as tumors, fractures, or abnormalities. Traditional supervised methods often face challenges in handling imbalanced datasets and generalizing across diverse cases.
Real-Time Object Detection and Tracking in Video Streams Using YOLO and Deep Reinforcement LearningJane Wanjiku Gitau
Download
Object detection and tracking in video streams have become essential for a variety of real-time applications, including autonomous driving, surveillance systems, and robotics. In this research, we propose an innovative framework that combines the strengths of YOLO (You Only Look Once) for high-speed object detection with Deep Reinforcement Learning (DRL) for adaptive and robust object tracking. While YOLO excels in providing fast and accurate detection, it often struggles with handling object interactions, occlusions, and scale variations in dynamic environments. To address these challenges, we leverage DRL, which allows for continuous learning and adaptation in tracking agents, thereby enhancing the system’s ability to handle complex scenarios.