Migraine headache (MH) classification using machine learning methods with data augmentation.

Area of Science:

• Neuroscience
• Medical Informatics
• Artificial Intelligence

Background:

• Migraine is a complex neurovascular disease with diagnostic challenges due to subjective pain measures.
• Accurate migraine diagnosis is crucial given its significant impact on brain, body, and overall function.
• Limited medical resources and awareness in some regions necessitate advanced diagnostic tools.

Purpose of the Study:

• To leverage machine learning (ML) algorithms for accurate prediction and classification of various migraine types.
• To evaluate the performance of different ML models, including deep learning, in migraine diagnosis.
• To address the need for improved diagnostic specificity in headache conditions.

Main Methods:

• Utilized a publicly available dataset for training ML models.
• Applied data augmentation techniques to enhance model robustness.
• Compared the performance of Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Random Forest (RF), Decision Tree (DST), and Deep Neural Networks (DNN).

Main Results:

• Deep Neural Networks (DNN) achieved the highest accuracy at 99.66%.
• Other models also demonstrated strong performance: Random Forest (98.50%), K-Nearest Neighbors (97.10%), Support Vector Machine (94.60%), and Decision Tree (88.20%) with data augmentation.
• All models showed improved performance with data augmentation for classifying seven types of migraine.

Conclusions:

• Deep learning and other ML algorithms show significant promise for accurate migraine classification.
• AI-powered tools can enhance diagnostic capabilities, particularly in resource-limited settings.
• The study highlights the transformative potential of artificial intelligence in improving migraine diagnosis and patient care.