External Email - Use Caution        

Dear Bruce Fischl,

Hi, I'm Yoonho.
I know that new features, such as SynthSeg, SynthStrip, segment_subfields (beta version), etc, based on deep learning have been added after the FreeSurfer 7.3.2.
Here I have questions (I read papers related to SynthSeg, SynthStrip, segment_subfields).
  1. In FreeSurfer 7.3.2, the recon-all function includes the mri_synthse function. Does the recon-all process include SynthSeg by default? Likewise, does SynthStrip have the same process by default?
  2. There are already processed files with FreeSurfer 7.2.0,  can we apply new functions (e.g., SynthSeg, SynthStrip, etc) to them?
  3. From SynthSeg related papers (e.g., Billot et al, 2021 and 2022), in the following quoted text, "Therefore, there is a clear need for a fast, accurate, and reproducible automated method, for segmentation of brain scans of any contrast and resolution, and that can adapt to a wide range of subject populations", Synthseg is said to be able to process MRI data of any contrast and resolution. Does this also apply to data taken from other manufacturers (Siemens, GE, Philips) and protocols (FSPGR, MPRAGE, etc)? Also, there are many studies that attempt to harmonize MRI data obtained from various sites. If there are different parts such as MRI data acquisition time, manufacturer, and protocol, the harmonization can be solved using the SynthSeg function of FreeSurfer 7.3.2. (In other words, can we simply proceed with the recon-all process?)  
Thank you for reading my long question
Please let me know if there is anything you don't understand.
Looking forward to your feedback.
Thank you very much.    

Sincerely,
Yoonho

--
Hwang, Yoonho
Research professor
Institute of Human Genomic Study,
College of Medicine, Korea University

#123, Jeokgeum-ro, Danwon-gu, Ansan-si, Gyeonggi-do, 15355, Republic of Korea.