Migraine Classification Using Machine Learning and Deep Learning in Low-Resource Healthcare Settings
Abstract
Migraine is a neurological condition that significantly impairs quality of life, with diagnostic challenges, particularly in resource-limited settings where specialised tools and expertise are lacking. While Artificial Intelligence (AI) models for migraine classification have been explored in standard healthcare contexts, limited research focuses their applicability in low-resource environments. To address this, we evaluated the efficacy of Machine Learning and Deep Learning models namely Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Decision Tree (DT), Random Forest (RF), and TabNet for migraine classification, with a specific focus on computational efficiency and interpretability. Among these models, RF emerged as the effective model, achieving 95.8% accuracy, precision, recall, and F1 score, whereas TabNet exhibited slightly lower performance with respective scores of 91.1%, 91.8%, 91.1%, and 90.7%. In addition, RF demonstrated enhanced computational efficiency, with a training time of 0.9s and memory usage of 0.14 MB, compared to TabNet's 10.8s and higher memory usage. Furthermore, SHapley Additive exPlanation (SHAP) analysis supported RF’s interpretability, and we propose RF as a cost- effective, AI-driven diagnostic tool for migraine classification, improving access to healthcare in resource-limited regions.
Recommended Citation
Ashokan, Anithamol; Rehman, Ikram Ur; and Saadati, Parisa, "Migraine Classification Using Machine Learning and Deep Learning in Low-Resource Healthcare Settings" (2025). UK Academy for Information Systems Conference Proceedings 2025. 5.
https://aisel.aisnet.org/ukais2025/5