New Study Aims to Develop AI-Based "Precision Medicine" for Migraine Relief

Healint, the developer and owner of Migraine Buddy - the world's largest migraine tracking and research platform used by more than 3 million people - is announcing a collaboration with the University of Geneva.

Prof. Katarzyna Wac, Director of the Quality-of-life technologies lab and mQoL Living lab, at the University of Geneva:      

"This collaboration with Healint is a great opportunity for our lab to get data that would not be available anywhere else," said Prof. Wac, "and we expect to help Prediction Research improve".    

"Working with the Geneva University Quality of life technologies is a wonderful opportunity! For the past 10 years, we, at Healint, have been gathering new types of data for the patient community, to improve communication on the efficacy of treatment, search for relief, and for the improvement of evaluation of Quality Of Life. This exploratory work with the university of Geneva will allow us to bring a new understanding on the dramatic domino effect we face before and during migraine attacks!" François Cadiou, CEO at Healint.

For this world first, the Migraine Buddy, 45.000 Migraine Buddy volunteers shared nearly 1 million events associated with the attacks, and 69 dimensions are associated per event, which constitutes:

 (i) Sensors, sleep, weather, to use as much as possible automation 

(ii) Self-report, we explain to our doctor, such as the level of pain, the drug we took, etc.

(iii) Some information allowing to group users in a relevant way and ensuring non-discrimination in the results (year of birth, sex, and gender, etc) 

The objective of the study will support "precision medicine", with a new method of evaluation of prediction of relief provided molecules (drugs) combined with the actions taken during an attack, such as sleep, staying in a dark room, cold showers, etc. The study also aims to isolate what triggers and contexts lead most often to a migraine attack on the next day.

Two approaches are being explored the auto persistence of Migraines and their covariation with the amount of sleep, and the Bayesian Hierarchical Modelling and Clustering of migraine attacks

Every patient and every attack is different. A specific treatment may work for some but not for all. Hence the need to take a "precision medicine" approach. The first part of the study will focus on making a prediction when a certain molecule will be the most helpful based on individual characteristics, with the ultimate goal to recommend medication based on the users' history and symptoms.

The second part of the study aims to model the migraine disease progression through the use of Markov chains. Markov chains are memoryless stochastic processes describing a sequence of possible events. Hidden Markov models will complement the approach and take into account the possibility that hidden variables and hidden states influence further states in the model when predicting migraine days. It is well accepted that the full mechanism of migraine attacks is far from being understood, hence the relevance of an approach allowing one to cope with the unknown.          

Healint's objective is to help research and promote a better understanding of migraine to improve disease management. For this project, the data and support from Healint the time allocated, and the resources provided by Healint are sponsored by the migraine-buddy users' subscriptions to its migraine app.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this news story?

Leave your feedback
Your comment type

While we only use edited and approved content for Azthena answers, it may on occasions provide incorrect responses. Please confirm any data provided with the related suppliers or authors. We do not provide medical advice, if you search for medical information you must always consult a medical professional before acting on any information provided.

Your questions, but not your email details will be shared with OpenAI and retained for 30 days in accordance with their privacy principles.

Please do not ask questions that use sensitive or confidential information.

Read the full Terms & Conditions.