Migraine: Causes, Symptoms, and Management. | ||||
| Damanhour Journal of Intelligent Systems and Informatics | ||||
| Volume 2, Issue 1 - Serial Number 20250200, June 2025 PDF (1000.83 K) | ||||
| Document Type: Scientific | ||||
| DOI: 10.21608/djis.2025.395659.1012 | ||||
 View on SCiNiTO | ||||
| Authors | ||||
M.A. Sayedelahl 1; Reem Shereif  2
| ||||
| 1Faculty of Computers and Information, Damanhur University, Egypt. | ||||
| 2Faculty of Computing and Bioinformatics, Banha University, Egypt. | ||||
| Abstract | ||||
| Migraine is a multifaceted neurological condition, and its diagnosis now relies on subjective clinical evaluations because of the absence of valid biomarkers. Recent developments in artificial intelligence (AI) and machine learning (ML) offer hopeful prospects for the improvement of diagnostic precision by automatic evaluation of patient data. This review discusses the application of AI, ML, and deep learning (DL) techniques in migraine subtype classification and diagnosis. We conducted a systematic literature search and identified 130+ studies from 1988-2024. We rigorously filtered the studies and chose 30 high-quality papers based on predefined inclusion criteria. We only included studies based on clinical data, neuroimaging, or patient-reported outcomes to train prediction models. Primary findings demonstrate that supervised machine learning algorithms (i.e., SVM, random forests) and deep neural networks can effectively classify migraine subtypes with more than 85% accuracy in some studies. Natural language processing (NLP) techniques are particularly promising for identifying diagnostic patterns from free-text clinical records. However, challenges remain regarding dataset heterogeneity and model generalizability | ||||
| Keywords | ||||
| headache; throbbing; unilateral pain; photophobia; phonophobia | ||||
|
Statistics Article View: 116 PDF Download: 159 |
||||
