Authors : Mohini Lawande

Volume/Issue : Volume 11 - 2026, Issue 5 - May

Google Scholar : https://tinyurl.com/32wjurz5

Scribd : https://tinyurl.com/47yvcab5

DOI : https://doi.org/10.38124/ijisrt/26May1446

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Abstract : Migraine is a disabling neurological condition whose symptoms, intensity, triggers, and progression vary across patients, making timely prediction difficult in routine care. This paper presents NeuroCare, a hybrid artificial intelligence framework that combines multiclass symptom-based migraine subtype classification with short-term sequential migraine risk prediction using sleep and heart rate variability data. The study draws on a structured migraine symptom dataset with 400 records and seven diagnostic categories, a sleep diary dataset containing 1,372 daily observations from 49 users, and a filtered HRV dataset with 38,913 physiological windows collected from 49 devices or users. A classical machine learning pipeline using Logistic Regression, Random Forest, and XGBoost was developed for subtype classification after preprocessing, standardization, and class balancing with SMOTE. In parallel, a 7-day LSTM framework was implemented to model temporal patterns from aligned sleep and HRV signals for next-day migraine risk estimation. The work also includes exploratory data analysis, dataset harmonization, artifact saving for deployment, and a hospital-style Tkinterbased GUI that supports patient intake, symptom capture, sequential entry, prediction review, and recommendation display. The proposed framework is intended as a practical research prototype rather than a notebook-only experiment, and it demonstrates how predictive analytics, temporal modeling, and user-centered interface design can be integrated into one clinically meaningful workflow.

Keywords : Migraine Prediction, Heart Rate Variability, Sleep Diary, LSTM, XGBoost, Clinical Decision Support, GUI, Wearable Sensors.

References :

1. Quartetti U. et al, ”Forecasting migraine attacks by managing daily lifestyle: a systematic review as a basis to develop predictive algorithms”, Pain Reports,10(2), e1247, 2025.
2. Das S. et al., ”Machine Learning-based Migraine Prediction: Analyzing Key Features and Cause-Effect Relationships for Improved Diagnosis and Management”, International Journal of Computer Applications ,Volume 187 - no.11, June 2025.
3. Amiri P. et al., ”Migraine: A Review on Its History, Global Epidemiology, Risk Factors, and Comorbidities”, Frontiers in neurology, volume 12-2021, February 2022.
4. Gulati S. et al. , ”Classification of Migraine Disease using Supervised Machine Learning”, 10th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO)., 2022.
5. Jalannavar A. et al., ”Migraine Prediction Using Deep Learning Model”, Fourth International Conference on Emerging Research in Electronics, Computer Science and Technology (ICERECT),2022.
6. Srinivasan S. et al. , ”Predictive Analysis of Chronic Migraine Symptoms Detection and Management Using IoT and ML”, 2025 International Conference on Advances in Modern Age Technologies for Health and Engineering Science (AMATHE), 2025.
7. Velgina R. et al., "Deciphering Migraine Types: A Machine Learning Odyssey for Precision Prediction", 2024 2nd International Conference on Intelligent Data Communication Technologies and Internet of Things (IDCIoT), 2024.
8. Chokri Baccouch. et al., "Advanced Machine Learning Approaches for Accurate Migraine Prediction and Classification" (IJACSA) International Journal of Advanced Computer Science and Applications, 2025.
9. Khan, Lal, et al. "Migraine headache (MH) classification using machine learning methods with data augmentation." Scientific Reports, volume 14, no. 1, 2 Mar. 2024, p. 5180. Nature.