ENE, RAPHAEL TUKWASICHUKWUOBI (2025) DEVELOPMENT OF A MACHINE LEARNING MODEL FOR PROSTRATE CANCER RISK PREDICTION IN AFRICAN MEN. Other thesis, GODFREY OKOYE UNIVERSITY, ENUGU STATE..
![[img]](https://eprints.gouni.edu.ng/style/images/fileicons/text.png) |
Text
REPO239.docx Download (2MB) |
Abstract
Prostate cancer remains to be one of the major causes of cancer morbidity and mortality globally. Due to the latter, the early and accurate detection is indeed significant in the event that we wish to advance treatment and patient outcome. This study presents a novel method of predicting the risk of prostate cancer by using pandas, the YOLOv11 deep learning model and developed using images of MRI scan. The results were trained and validated on a mature segmentation dataset which made it learn on highlights and pinpoint the cancerous area more accurately. YOLOv11 did well in evaluation. At an IoU cutoff of 0.5 it achieved a mean Average Precision (mAP) of 93.7%. Similarly, it posted a precision of 91.2 percent as well as a recall of 89.6 which are good scores as well. As a way of making the system simple to apply in the actual world we constructed a web-based version. In this version, users can be able to upload MRI scans and the detection results will appear instantly with well marked zones of the prostate. It is made simple and easy to use in clinical setting. Overall, the findings indicate that the model based on YOLOv11 is precise, quick and simple to operate. It can actually assist radiologists in providing them with rapid and trustworthy information when carrying out detection in prostate cancer which would get them to make a more informed and swift decision.
| Item Type: | Thesis (Other) |
|---|---|
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science |
| Divisions: | Faculty of Engineering, Science and Mathematics > School of Electronics and Computer Science |
| Depositing User: | Uchenna Eneogwe |
| Date Deposited: | 29 May 2026 13:55 |
| Last Modified: | 29 May 2026 15:33 |
| URI: | http://eprints.gouni.edu.ng/id/eprint/5680 |
Actions (login required)
 |
View Item |
