CCIS

Title of research project	Brief	Authors	Branch
An efficient deep learning model for aircraft safety by bird strike prevention technology at Oman civil aviation/armed airport authorities	The paper introduces a deep learning model aimed at detecting and reducing bird strikes near airports to enhance aircraft safety. Using a large, diverse bird image dataset and advanced image augmentation, the model is trained to handle various flight scenarios. It employs a spatiotemporal convolutional neural network with spatial attention and dynamic temporal processing to accurately detect birds in flight. This approach improves real-time detection reliability under changing environmental conditions. The goal is to support more effective bird strike prevention technologies in aviation.	Dr.Najiba Said Hamed Al-Zadjali, Dr.B. Sudaravadivazhagan, Dr.Charles Savarimuthu, Emanuel O. Rances	UTAS-Al Musanna
Classification and Early Detection of Organ-Specific Autoimmune Induced Liver Disease and Conceivable Solutions using Machine Learning and Deep Learning Techniques	The paper addresses the challenge of diagnosing Autoimmune Liver Disease (AiLD), a chronic condition that can lead to liver cirrhosis or cancer if untreated. Due to overlapping symptoms with other autoimmune disorders, early detection is difficult. The study proposes a hybrid model combining supervised machine learning (using patient history and lab data) and deep learning (using liver images from the LKS dataset) to predict AiLD. This integrated system aims to improve diagnostic accuracy, reduce the need for invasive tests, and offer a reliable second opinion for clinicians.	Dr. Ben George Ephrem ,Dr. Susan Teresa, Dr. Gnana Rajesh, Ms. Jeba Roseline, Dr. Abraham Varghese, Dr. Giri Ramdoss, Dr. Samuel Giftson	UTAS- Muscat
Voice-enabled Assistive System for Visually impaired People to Detect the Object in Dark Environment by using Thermal Image Processing	Visually impaired individuals face challenges in daily activities, often leading to physical and psychological impacts. Assistive technologies have improved their confidence and social interaction. While tools like reading apps and magnifiers exist, indoor object detection remains underdeveloped. Traditional RGB and IR cameras have limitations based on lighting, whereas thermal cameras detect objects through heat radiation. This research highlights the potential of thermal imaging for effective indoor navigation support for the blind and visually impaired (BVI), enhancing their safety and independence.	Dr. Mary Amirtha Sagayee, Dr. Rolando Jr Lontok, Dr. Ahmed Al Siyabi	UTAS Nizwa
Control of Security Breach of Optimization Technique Using Machine Learning	This study focuses on reducing the growing power consumption of IP networks using both hardware and software solutions. While changing transmission media can save energy, combining dynamic power states with traffic matrix analysis offers greater efficiency. A lightweight heuristic model is also proposed for energy-efficient traffic management. Machine learning techniques are used to analyze population data trends, enhancing performance. To prevent data breaches, the study introduces the S-GRIDA optimization technique, which achieved 98.4% accuracy, proving its effectiveness in securing network operations.	Vimal Kumar Stephen, Ramesh Palanisamy, Zaina Rashid Salim Al-Jufaili, Mallika Banu Basheer Ahmed	UTAS-IBRA
Adaptive learning process with AI environment using RoboLearnPI	This study explores the use of AI-powered adaptive learning systems to personalize education based on individual student needs and preferences. Technologies like machine learning, computer vision, and NLP enhance engagement and learning outcomes. The project proposes a smart robot built with Raspberry Pi to deliver a more interactive and tailored educational experience. Acting as a bridge between digital tools and physical interaction, the robot supports immersive learning in adaptive environments, helping teachers achieve improved student performance.	Dr.Mathivanan V and Dr. Ramesh Palaniswmy	UTAS-IBRA
Smart Sustainable Green UTAS Campus	This research focuses on identifying challenges in educational institutions lacking smart infrastructure and explores existing smart devices that can enhance resource utilization. It proposes the design and implementation of a smart classroom using modern technologies to improve efficiency in higher education. A system architecture will be developed based on institutional needs and available smart tools. The study includes a comparison between traditional and smart systems in terms of automation, utilization, and efficiency. It also extends to other campus areas like parking and libraries, aiming to build a fully integrated Smart Campus through detailed functional and non-functional requirement analysis.	Dr. Balaji Dhanasekaran (PI), Mr. Senthilkumar Natarajan, Dr. Abhishek Dubey, Dr. Rhouma (Co-PI, Resigned)	UTAS-Salalah
Analyzing and Developing an Automated Allergen Detection System for Food Ingredients through Predictive Modeling	This research aims to enhance food safety by developing an automated system for accurate allergen detection in food ingredients. Using advanced machine learning and predictive modeling techniques, the system analyzes large datasets to identify patterns linked to allergenic compounds. The findings demonstrate that the automated system outperforms manual methods, significantly improving detection accuracy and reliability. This approach offers valuable insights for food manufacturers and regulatory bodies, helping to prevent allergic reactions. The study presents a novel integration of data science and public health to address a major gap in current food safety practices.	Dr.Rajesh Natarajan, Dr A Saleem Raja, Dr Sujatha Krishna, Dr Pradeepa Ganesan, Ms Myra, Mr Syed Ibrahim	UTAS-Shinas
Improving the Motion Capture Data Using Artificial Intelligence Methods	This research aims to improve the reliability of motion capture data from wearable sensors by developing AI-based methods to impute data lost due to connectivity issues or obstructions. Using Perception Neuron 3 sensors, the study captures human motion and introduces artificial gaps to simulate real-world data loss. Multiple AI models—including ML (e.g., k-NN, MLP), DL (e.g., autoencoders, Echo State Networks), and GANs (e.g., MisGAN)—are compared against traditional imputation methods. Model performance is evaluated using RMSE and DTW metrics. The project seeks to identify the most accurate and robust technique for real-world applications in animation, healthcare, and biomechanics.	Dr. Ahmed Salah, Dr. Ahmed Alrawahi, Dr. Khalil Al Ghafri	UTAS-Ibri
Exploring the Nexus of Microplastics Contamination in Foods, Environmental Health, and AI-Driven Solutions for Enhanced Food Security	Microplastics pose a significant threat to human health and the environment, contaminating various food products and requiring urgent, sustainable solutions. It is crucial to detect and analyze these pollutants without harming natural resources. This research project aims to tackle this challenge by utilizing advanced technologies and Artificial Intelligence to identify and characterize microplastics in food and environmental sources effectively.	Dr. Mohamed Abbas-Co PI, Dr. Syed Shahul Hameed -Member	UTAS-Sur
Unraveling Cardiovascular Disease: Leveraging Artificial Intelligence(AI) and Explainable-AI for Precision Cardiovascular Disease Prediction	Cardiovascular Disease (CVD) encompasses various heart and blood vessel disorders such as coronary artery disease, hypertension, heart failure, and stroke. It is a leading global cause of death, with the WHO reporting approximately 17.9 million CVD-related deaths annually, accounting for about 31% of total global deaths. Recognizing its impact, the Ministry of Health in Oman prioritizes CVD and metabolic syndrome research. This highlights the need to utilize population-specific data in Oman to better assess and manage CVD risk among its citizens.	Faizal Hajamohideen, Karthikeyan Subramanian, Viswan Vimbi, Mrs.Alya Salim AL-Harthi	UTAS-Suhar

Contact UTAS

About

Study