I wanted to solve a persistent migraine/aura problem my daughter suffers with. I didn't have the data so I extracted all migraine events programmatically from five years of our text message history.
For each event identified I pulled 96-hours of weather data preceding its onset, including barometric pressure, temperature, and humidity.
The result is a dataset of 79 migraines with their preceding weather patterns, which proved my hypothesis that a dip in pressure followed by a spike, precedes a migraine by about ~12-14 hours.
A temperature dip, and spike in humidity accompanies that pattern, every single time. So with accurate forecasting data it would likely be possible to provide fairly confident windows of migraine risk, which can be mitigated ahead of time.
If you have any kind of chronic or persistent health issue the best thing you can do is collect data. Track symptoms daily. Soon enough someone somewhere will be able to do something with it that solves it.