EZEUDU, VINCENT IKECHUKWU (2025) INTEGRATED WEB APP FOR 2D TUMOR VISUALIZATION, SEGMENTATION, AND GROWTH PREDICTION. Other thesis, GODFREY OKOYE UNIVERSITY, ENUGU.
![[img]](https://eprints.gouni.edu.ng/style/images/fileicons/text.png) |
Text
REPO220.docx Download (3MB) |
Abstract
This study presents the development of an integrated web-based application for 2D tumor visualization, segmentation, and growth prediction using deep learning and interactive web technologies. The system is designed to assist medical professionals by automating brain tumor analysis from MRI scans, providing both accurate detection and predictive analytics for improved clinical decision-making. The application combines a user-friendly Streamlit interface for uploading and viewing medical images, a YOLOv8 instance segmentation model for tumor detection, and WebGL/Three.js for 2D rendering of tumor structures. The methodology involved training the YOLOv8 model on a labeled MRI dataset, implementing cloud-based deployment, and integrating a growth prediction feature using time-series modeling techniques. The results demonstrate high accuracy in tumor segmentation, with the added benefit of time-lapse visualizations showing potential tumor progression. Evaluation metrics, including Dice similarity and Intersection-over-Union (IoU), validate the system's performance. The web app also supports linking to electronic patient records, enabling personalized treatment planning and remote access. This study addresses critical challenges in medical imaging by offering a scalable, interactive, and AI-powered solution. The system's potential impact includes enhanced diagnostic efficiency, better patient communication, and broader access to advanced medical imaging tools, particularly in low-resource settings.
| Item Type: | Thesis (Other) |
|---|---|
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science |
| Divisions: | Faculty of Natural and Applied Sciences |
| Depositing User: | Uchenna Eneogwe |
| Date Deposited: | 19 Jun 2026 17:04 |
| Last Modified: | 19 Jun 2026 17:04 |
| URI: | http://eprints.gouni.edu.ng/id/eprint/5859 |
Actions (login required)
 |
View Item |
