BreastGAN: Artificial Intelligence-Enabled Breast Augmentation Simulation

Abstract

Background Managing patient expectations is important to ensuring patient satisfaction in aesthetic medicine. To this end, computer technology developed to photograph, digitize, and manipulate three-dimensional (3D) objects has been applied to the female breast. However, the systems remain complex, physically cumbersome, and extremely expensive. Objectives The authors of the current study wish to introduce the plastic surgery community to BreastGAN, a portable, artificial intelligence-equipped tool trained on real clinical images to simulate breast augmentation outcomes. Methods Charts of all patients who underwent bilateral breast augmentation performed by the senior author were retrieved and analyzed. Frontal before and after images were collected from each patient’s chart, cropped in a standardized fashion, and used to train a neural network designed to manipulate before images to simulate a surgical result. AI-generated frontal after images were then compared to the real surgical results. Results Standardizing the evaluation of surgical results is a timeless challenge which persists in the context of AI-synthesized after images. In this study, AI-generated images were comparable to real surgical results. Conclusions This study features a portable, cost-effective neural network trained on real clinical images and designed to simulate surgical results following bilateral breast augmentation. Tools trained on a larger dataset of standardized surgical image pairs will be the subject of future studies.

Assessment of Patient Satisfaction Using a New Augmented Reality Simulation Software for Breast Augmentation: A Prospective Study

“…Worldwide, breast augmentation is the most common surgical procedure performed by plastic surgeons, representing 17.6% of all plastic surgical procedures [ 6 ]. Of the 1,862,506 breast augmentations performed worldwide in 2018, 17.3% were performed in the United States, followed by Brazil (14.8%), Mexico (3.8%), Germany (3.5%), Italy (3.5%), Argentina (2.7%), and Colombia (2.3%) [ 11 ].…”

“…Pre-operative planning involves testing several breast implants in order to choose the one that best suits the patient. This kind of simulation has been demonstrated to improve post-operative patient satisfaction overall and with respect to breast size [ 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 ]. Nevertheless, this kind of simulation may not be realistic, as sizers are applied in the bra over the breast, patients may not be wearing the right garments, and a large variety of implants may not be available (round or anatomical: XP, HP, MP) in the surgeon’s office.…”

“…Furthermore, if a mastopexy is indicated, it is not possible to obtain a realistic post-operative simulation with breast sizers. These disadvantages have led to the development of 3D simulation software [ 6 ]. Arbrea Breast Software (ABS) is a 3D, live augmented reality simulation software for pre-visualization of breast surgery results that allows a more accurate prediction of the appropriate breast implant size and volume.…”

Artificial intelligence in plastic surgery: current developments and future perspectives

“…During the stage of online patient consultation, artificial intelligent virtual assistants (AIVA) using NLP technology understands the intent of human speech and provide answers in the form of dialogue. A study used AIVA to chat with patients on common plastic surgery topics [17] . The overall answer accuracy of the plastic surgery AIVA was 92.3%, and 83.3% of the participants thought the answers were correct [17] .…”

A Synopsis of Artificial Intelligence and its Applications in Surgery

“…This constitutes an impetus for a consensus media quality standard to be respected across all institutions and research groups, such that datasets are no longer siloed by heterogeneity. 30 Lastly, establishing a "ground truth" for algorithm training purposes has proven difficult in surgical education because unanimity of dissection technique or instrument use is rarely achieved between multiple surgeons. 29 In a recent study, Madani et al 31 proposed a framework to achieve consensus among minimally invasive surgeons.…”