search

BIAPT / awareness-perturbation-complexity-index

Development of an index for assessing the level of consciousness of healthy and disorder of consciousness individuals.
1 stars 0 forks source link

Improve the dpli-dri to be useable in an article #7

Closed yacineMahdid closed 4 years ago

yacineMahdid commented 4 years ago

I had some feedback and the index is actually really good. The only problem is that the range is very narrow 0.5 to 0.53 and this is because we are dividing by 2*matrix.
This is the definition we were using prior: dri_dpli = sum(BvR) / (w1*(sum(BvA)) + w2*(sum(RvA))) where w1 and w2 = 1 To fix the range problem we could go with something like this: dri_dpli = 2*sum(BvR) - ( w1*sum(BvA) + w2*sum(RvA) ) This we are equilibrating the two side of the contrast, we still will have a high number if the similarity is higher in the BvR and a lower number if the similarity is 'similar' in each matrix irrespective of the anesthesia.

yacineMahdid commented 4 years ago

This formula won't be able to bring the result in between 0 and 1, but we don't need to if we do a in-between subject comparison.

yacineMahdid commented 4 years ago

I'm still waiting on feedback for this, but will improve it iteratively.