I found a potential virtual brain digital twin which could be fun to look into. See this link and let me know if you all have any ideas Virtual Brain Twin for personalized treatment of Psychiatric Disorders - EBRAINS
They seem to be progressing quite well:
with respect to what is stated here:
at this moment I can easily predict whether a patient will respond well to a medication if there is sufficient EEG data from those for whom it worked and those for whom it did not; I can also identify, in broad terms, what the difference is (though not yet in causal terms).
But it definitely sounds good; I’m only left wondering whether you want feedback regarding the possibilities, participation, or something more specific like suggesting an experiment.
Thanks for responding! Good to see you think it is a good project. I have nothing in particular in mind, I just think it is interesting and maybe down the line could become a fun project with the OSDR data.
I wonder if we could reach out to the authors? They list a few papers related to virtual patients (PD, MS), and it could be interesting to see if their approach could be adapted to spaceflight conditions.
Maybe these kinds of virtual twins could be used to extrapolate past the time frame and type of data we have (longer duration missions, different levels/combinations/time-courses of radiation exposure, etc.) or to test interventions.
It could be a challenge to re-purpose their work, though; we may not have enough data to train new models, and their inverted models may not apply to spaceflight. For instance, they have differential equation models of dopaminergic release and re-uptake, but would their extracted parameters fit spaceflight data well (not that I’m aware of any dopaminergic tone data in OSDR, but just as an example)? But perhaps we could test the fit even if we can’t train new data, or we could re-purpose ground experiment data from more sources to refine parameters?
That’s a really good idea! You never know, it could turn into an interesting conversation.