I am working with resting state data obtained from children with neurodevelopmental disorders like ADHD and I use FIX to remove the noisy components. 

When I check my motion parameters before and after FIX, there is a huge difference. A good thing is that I don't need to exclude any of my participants based on their FD after denoising. 

However, I was wondering if it's better to have stringent exclusion criteria based on FD before denoising. 

Should I exclude participants with high motion first and denoise the rest only.

Thanks for your help.

Sara