Is it implemented by functions such as BET2? And I see the following functions in FSL.

[xingyunjohn1]

[vanandrew]

I'm not entirely sure what you're asking...

This project is a re-implementation of FSL's bet algorithm in Python. I wrote it for a class project and made it publicly available for reference.

It's not affiliated with the FSL project in any way.

[xingyunjohn1]

I am sorry for my bad expression.

1. I wonder whether there would be a bet2 reimplementation in your project.
2. Accroding to the intro of BET page, it seems bet should be used for deleting non-brain tissue from an image of the whole head while bet2 should be used for skull and scalp surface extraction. But accroding to the screenshot of the FSL software I showed above, brain extaction also uses bet2. I haven't research deep inside this. I would appriciate it if you kown something about this and tell me more about this.
3. Could you please tell me how many iterations that the FSL software runs while you set it for 1000 if you know?

At last, thank you for your great work of this reimplementation.

[vanandrew]

1. This project implements just bet (per the original paper: Smith SM. Fast robust automated brain extraction. Hum Brain Mapp. 2002 Nov;17(3):143-55. doi: 10.1002/hbm.10062. PMID: 12391568; PMCID: PMC6871816.). Features in bet2 are not implemented and won't ever be implemented at this point.
2. Both bet and bet2 will delete non-brain tissue from the image. My understanding is that bet2 is mainly an extension on bet that includes skull and scalp surface extraction, it still uses the same underlying algorithm as bet for brain extraction.
3. I'm not entirely sure how many iterations is run in the FSL software's bet. When writing this, I only used the original paper as a guide and I did not look into any of FSL's bet source code. I chose 1000 iterations because I didn't see much change in the brain extraction results beyond that point (and 1000 seemed like a nice round number).

Hope this helps, Andrew