TITLE:
Sporadic Hemiplegic Migraine Case Study and Classification Using Machine Learning
AUTHORS:
Anusha Reddy, Ajit Reddy
KEYWORDS:
Migraine, Familial Hemiplegic Migraine, Sporadic Hemiplegic Migraine, Decision Tree Classifier, Classification and Regression Tree, Random Forest Classifier
JOURNAL NAME:
World Journal of Neuroscience,
Vol.15 No.4,
November
7,
2025
ABSTRACT: Background: Sporadic hemiplegic migraine (SHM) is a rare and complex type of migraine that is characterized by temporary paralysis or weakness (hemiplegia) on one side of the body, which is estimated to affect a tiny percentage of the population and is often misdiagnosed by physicians, and is generally treated as a headache. The hemiplegia in SHM is typically reversible and usually lasts for the duration of the migraine attack, but it can cause significant concern due to its neurological nature. Methods: Since the number of patients suffering from this type of migraine is rare, not enough data has been collected, making diagnosis extremely difficult even for the best physicians. This often leads to misdiagnoses and delayed treatment for the patient. In this paper, we present a case study of a patient with sporadic hemiplegic migraine who has gone undiagnosed for many years by physicians. So, we discuss using Machine Learning classifiers to diagnose migraines of different types and, very specifically, the rare types of migraine, such as sporadic hemiplegic migraine, with a high degree of certainty. Results: We present the use of Machine Learning classifiers, such as Decision Tree (DT) and Random Forest (RF), to classify different types of migraine, specifically for diagnosing sporadic hemiplegic migraine. The classifiers’ results, trained on the dataset, are verified for their performance across the overall dataset for all types of migraines, specifically for sporadic hemiplegic migraines. Given the dataset, these classifiers indicate that a high degree of precision and accuracy is achievable with Decision Trees and Random Forests for sporadic hemiplegic migraines. We find that the accuracy and precision obtained by tree-based machine learning classifiers would aid in patients’ diagnoses. Conclusion: In this study, we present classifiers that promise to aid everyday physicians. We evaluate the classifiers’ performance using classification metrics. With these classifiers, different types of migraines, specifically sporadic hemiplegic migraine, can be classified with high accuracy and reliability so that physicians can make timely clinical diagnoses of patients.