SICKLE CELL CRISES PREDICTION SYSTEM
Abstract
This project presents the design and development of, a wearable health monitoring
system aimed at predicting and managing sickle cell crises. The system integrates
real-time monitoring of vital signs such as SpO₂, body temperature, and hydration
status, alongside demographic risk factors like age and gender. A machine learning
model processes these parameters to predict crisis onset with high accuracy,
providing early warnings to patients and caregivers.
The system's architecture includes a wearable device that collects physiological data,
a mobile application for real-time monitoring and alerts, and a web interface for
caregiver oversight. A key feature is the predictive model, built using an ensemble
approach combining Logistic Regression, Random Forest, and XGBoost, with a
focus on stability and generalizability. The model was trained using a dataset of
sickle cell patients, and its performance was evaluated through metrics such as ROC
AUC, F1 Score, precision, and recall, yielding promising results in both
classification and real-time prediction tasks.
Full-Text Access Notice
In accordance with the NERD Policy on promoting peer-reviewed publication, public access to the full text of a project, thesis or dissertation is restricted for three years, allowing the author and supervisors sufficient time to pursue peer-reviewed publication.
During this period, researchers with legitimate academic or research purposes may request authorisation directly from the author to enable NERD to release the indexed full texts of the work using the form below.