DCU-Final-Year-Projects-Booklet-2025

35 9. Parallelising LUT-Based Topological Thinning for Colon Centreline Extraction This project investigates the parallelisation of a topological thinning algorithm for centreline extraction in virtual colonoscopy. Centreline extraction is a key step for navigating tubular structures in medical imaging, used for patient diagnosis. The project implements and accelerates an existing Lookup Table (LUT)-based thinning algorithmoptimisation to improve computational efficiency. A region-growing segmentation algorithm is developed to extract the colon from volumetric CT data before thinning is applied. GPU acceleration using OpenCL is utilised to enhance performance. The project evaluates the speed, accuracy, and scalability of the parallel vs sequential implementations. Other contributions to the method are also examined in the project. Student Programme Electronic and Computer Engineering (Year 5) Project Area Image/Video Processing, Software Development, GPU-Accelerated Computing Project Technology Python, OpenCL Student Name(s) Patriks Vitols Jegurs Email patriks.vitolsjegurs2@mail.dcu.ie Supervisor Dr Robert Sadleir 10. Addressing the Shine Through Artefact in 2DTextureMapping-Based Volume Rendering In volume rendering, there is an artefact called “shine-through” that occurs when the 2D planes of a rendered image are parallel with the viewing angle of the camera, reducing image clarity. Texture mapping volume rendering is utilised in the areas of medical imagining and computer game graphics. This is one of the main techniques used in the visualisation of CT scans. Addressing this issue will enhance the accuracy and reliability of CT scans and improve the visualisation of volumetric objects. Through a combination of research and creative problem-solving, this project will explore various solutions to mitigate shine-through, compare their effectiveness and assess the solutions in terms of performance for volume-based rendering. Student Programme Electronic and Computer Engineering (Year 4) Project Area 3-DModelling, Graphics, Software Development Project Technology HTML5, JavaScript, OpenGL Student Name(s) HughManley Email hugh.manley2@mail.dcu.ie Supervisor Dr Robert Sadleir 11. Road Pal This project investigates the development of amobile app, Road Pal, designed to enhance road safety by leveraging real-time tracking, machine learning, andmapping APIs. The appmonitors vehicle speed, compares it to speed limits, and provides alerts when limits are exceeded. Using the device’s camera, it detects road signs, such as no-entry zones, and warns drivers of restricted areas. Hands-free voice notifications ensure minimal distractions, promoting safer driving. This solution addresses the growing need for road safety tools, particularly in regions with frequent changes in speed limits and restricted zones. By combining technologies such as TensorFlow Lite and GoogleMaps API, this project offers drivers real-time assistance to reduce traffic violations and improve compliance. Student Programme Computer Applications Project Area Android, Artificial Intelligence, Computer Vision, GPS/GIS, Mobile App Project Technology Python, Machine Learning, Flutter & Android Studio Student Name(s) AsimAbdullah Email asim.abdullah6@mail.dcu.ie Supervisor Dr Michael Scriney