ONYEKA-OKELU, OLISAEMEKA FRANCISCO (2025) AI-BASED LUNGS DISEASES DETECTION SYSTEM USING MACHING LEARNING ALGORITHM (AMLs). Other thesis, GODFREY OKOYE UNIVERSITY, ENUGU.
![[img]](https://eprints.gouni.edu.ng/style/images/fileicons/text.png) |
Text
REPO194.docx Download (1MB) |
Abstract
The present project introduces the concept of lung disease diagnosis framework based on Artificial Intelligence (AI) technology with the help of powerful machine learning algorithms. The system is driven by the fact that there is a rising demand to increase access to timely and accurate lung conditions diagnosis particularly in low resource settings, there is a need to diagnose diseases like pneumonia, chronic obstructive pulmonary disease (COPD) and tuberculosis using chest X-ray images. With the help of Convolutional Neural Networks (CNNs), the system assesses medical images to detect patterns of diseases very accurately and with little human interference. The paper examines current AI models and defines their flaws that include the specificity to individual problems and the use of unbalanced data. The proposed set of solutions uses the combination of image and clinical data to improve the reliability of the diagnosis. It will consist of a user-friendly web interface, which was developed using Next.js and the web application is connected to a Flask REST API, where the trained models will be hosted. The validated models had a prediction rate of more than 89% and this good result shows that the models can be used to support the work of healthcare providers and provide them with more reliable and faster results. Effective treatment requires the identification of gaps in low-resource areas through diagnosis; thus, the project will enable the project to fill these gaps by providing an accessible and scalable instrument, which permits early intervention and improved patient outcomes. With regard to the perspective, in the future, a real-world clinical validation, expanded data-sets to generalize the current finding, explainable AI properties, and the development of a mobile app to expand reach will be recommended.
| 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 13:48 |
| Last Modified: | 19 Jun 2026 13:48 |
| URI: | http://eprints.gouni.edu.ng/id/eprint/5841 |
Actions (login required)
 |
View Item |
