Facial Feminization Research New York City
Dr. Bradley’s aim is to provide the best, most up-to-date care for each patient. He has many years of craniofacial experience. He believes that the use of the latest technology helps providing optimal results with shorter and less painful recovery. Part of staying on the cutting edge of surgical procedures is conducting scientific research and teaching others surgical technique.
- Dr. Bradley and his team were the first to show that use virtual surgical planning for FFS allows for more efficient, more accurate, and safer care. (see below)
- Dr. Bradley and his team were the first to show that artificial intelligence gender identification demonstrated the success of FFS. (see below)
- Dr. Bradley and his team were the first to show that public perception in the form of crowd sourcing verified that less mis-gendering occurs following FFS. (see below)
Osseous Transformation with FFS: Improved Anatomical accuracy with virtual planning
In this study 3D CT scans were used:
- to access morphologic type to various regions of the face;
- to create patient-specific cutting guides;
- to manufacture custom fixation plates.
Data collection showed that FFS performed with virtual planning was:
- more accurate (97% vs 79%);
- more efficient (less OR time);
- safer (minimized injury and complications)
Facial recognition neural networks confirm success of FFS
In this study facial recognition neural networks used artificial intelligence:
- to gender-type cis-male and cis-female images;
- to gender-type preoperative FFS and postoperative FFS images;
- access confidence in femininity in the four comparative groups: cis-male, cis-female preop FFS post FFS.
Data collection showed that artificial intelligence:
- correctly identified cis-males and cis-females (100% vs 98%);
- preoperative FFS were mis-gendered 47% of the time;
- BUT postoperative FFS were correctly gendered 98% of the time (similar to cis-female).
FFS changes perception of patient gender
In this study public perception was looked at using crowd sourcing (large sample size):
- to identify gender-type in cis-male and cis-female images;
- to identify gender-type in preoperative FFS and postoperative FFS images;
- to assign a confidence score to their decision.
Data collection showed that the general public were able to:
- correctly gender-type cis-male and cis-female (99% vs 99%);
- correctly gender-type preop FFS only 57% BUT was correct with postop FFS 95% of the time;
- improve their confidence in accessing femininity greatly after FFS.