Machine Learning Assessment of Next-Day Migraine Likelihood Using Data From 53,000 App Users Living With Migraine
The ability to anticipate an impending migraine attack has the potential to improve patients' well-being, yet previous models have shown limited performance. The current study aimed to develop and evaluate a machine learning (ML) model to identify risk of next-day migraine attacks based on electronic diaries, leveraging the longest preattack time frame (30 days) and the largest prospective dataset of patients with migraine (over 53,000) to date.
Authors
- Alon Ironi (ORCID: https://orcid.org/0000-0002-3992-4978)
- Peter J. Goadsby (ORCID: https://orcid.org/0000-0003-3260-5904)
- Robert P. Cowan (ORCID: https://orcid.org/0000-0002-6868-0108)
- Alit Stark‐Inbar (ORCID: https://orcid.org/0000-0001-6434-6447)
- Christopher Louis Robinson (ORCID: https://orcid.org/0000-0002-6276-9056)
- Liron Rabany (ORCID: https://orcid.org/0000-0003-3394-5361)
- Shira Shalev-Aviad (ORCID: https://orcid.org/0009-0001-3879-1472)
- Ronen Jashek (ORCID: https://orcid.org/0009-0001-1143-1469)
- Anastasia Markina (ORCID: https://orcid.org/0009-0000-0224-8729)
Institutions
- Johns Hopkins University (US)
- King's College London (GB)
- Johns Hopkins Medicine (US)
- H2Pro (Israel) (IL)
- King Abdullah University of Science and Technology (SA)
- Stanford University (US)
Publication Details
- Journal
- Neurology Open Access
- Published
- 2026-08-21
- DOI
- https://doi.org/10.1212/wn9.0000000000000144
- Primary Topic
- Migraine and Headache Studies
- Type
- article
- Field-Weighted Citation Impact
- 0.00