NNAMANI, CHIZOBAM JOSHUA (2026) EXPERT DIAGNOSTIC SYSTEM WITH MACHINE LEARNING TREATMENT RECOMMENDATION SUPPORT FOR RURAL HEALTH FACILITY. Other thesis, GODFREY OKOYE UNIVERSITY, ENUGU,.
![[img]](https://eprints.gouni.edu.ng/style/images/fileicons/text.png) |
Text
TREATMENT RECOMMENDATION SUPPORT FOR RURAL HEALTH FACILITY.pdf Download (2MB) |
Abstract
Rural healthcare facilities frequently struggle with a lack of skilled staff, limited gear, slow diagnosis, and limited specialist help. Because of these issues, it's tough to tackle typical diseases early in places with few resources. So, we created an Expert Diagnostic System that suggests treatments using ML for rural areas. RuralAI was built as a web-based tool for diagnosing patients. It was built using Python Flask along with HTML, CSS, and JavaScript. For its AI side, we went with the XGBoost machine learning method, which works well for sorting through classification problems. Essentially, the system sorts patient cases by looking at their symptoms and key vitals like temperature, heart rate, breathing rate, oxygen levels, and blood sugar. It also factors in age and gender when deciding what to recommend. The system lets healthcare workers sign up, log in, input patient symptoms and vital signs, and get some helpful diagnostic suggestions. It also offers guidance on triage and treatment options, along with explanations for its recommendations. This focuses on common health issues in rural areas like malaria, typhoid fever, pneumonia, diabetes, and acute respiratory infections. Experts tested the system with various standards, and it performed really well – accuracy at 91.3%, precision at 90.7%, recall at 91.0%, F1-score at 90.8%, and an AUC-ROC of 0.94. So, it can indeed offer valuable initial diagnosis help for healthcare workers in rural places. Still, it's meant to assist in clinical decision-making, not replace the judgment of professionals.
| Item Type: | Thesis (Other) |
|---|---|
| Subjects: | Q Science > Q Science (General) |
| Divisions: | Faculty of Computing And Information Technology (FACIT) |
| Depositing User: | Nnenna Ayo |
| Date Deposited: | 24 Jul 2026 15:37 |
| Last Modified: | 24 Jul 2026 15:37 |
| URI: | http://eprints.gouni.edu.ng/id/eprint/6018 |
Actions (login required)
 |
View Item |
