AutoFFS: Adversarial Deformations for Facial Feminization Surgery Planning

2026-03-02

AI Summary

This study addresses the lack of quantitative and reproducible anatomical guidance in facial feminization surgery, which often relies on subjective assessments. The authors propose a counterfactual generation method based on adversarial free-form deformation, wherein individual cranial structures are morphologically optimized toward a target gender by strategically perturbing a pre-trained skull-based gender classifier through targeted adversarial attacks. This approach uniquely integrates adversarial deformation with a binary gender classifier to enable data-driven, precise morphological transformation. Experimental results demonstrate that the generated cranial models consistently exhibit pronounced target-gender characteristics, as validated by both classifier outputs and human perceptual evaluations, thereby confirming the method’s efficacy and clinical potential for preoperative planning in gender-affirming care.

Technology Category

• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Machine Learning: Adversarial Learning & Robustness
• Multiagent Systems: Adversarial Agents

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Facial feminization surgery (FFS) is a key component of gender affirmation for transgender and gender diverse patients, aiming to reshape craniofacial structures toward a female morphology. Current surgical planning procedures largely rely on subjective clinical assessment, lacking quantitative and reproducible anatomical guidance. We therefore propose AutoFFS, a novel data-driven framework that generates counterfactual skull morphologies through adversarial free-form deformations. Our method performs a deformation-based targeted adversarial attack on an ensemble of pre-trained binary sex classifiers that learned sexual dimorphism, effectively transforming individual skull shapes toward the target sex. The generated counterfactual skull morphologies provide a quantitative foundation for preoperative planning in FFS, driving advances in this largely overlooked patient group. We validate our approach through classifier-based evaluation and a human perceptual study, confirming that the generated morphologies exhibit target sex characteristics.

Problem

• Facial Feminization Surgery
• Sexual Dimorphism
• Surgical Planning
• Craniofacial Morphology
• Gender Affirmation

Innovation

• adversarial deformation
• facial feminization surgery
• counterfactual morphology
• sexual dimorphism
• surgical planning