Page 114 - Read Online
P. 114

Shen et al. Art Int Surg. 2025;5:150-9  https://dx.doi.org/10.20517/ais.2024.71                                                             Page 156

               informed decisions.


               Moreover, patients undergoing breast reconstruction, especially those from racial minority groups, report
               significant dissatisfaction with the information provided to them regarding preoperative decision making
                                      [47]
               and postoperative recovery . They often perceive significant gaps in their knowledge and understanding of
               expected outcomes, leading to increased anxiety and fear during the surgical process. Patients are
               increasingly turning to online resources for advice and insights prior to a potential surgical journey, often
               because of the stress and frustration of navigating the complex or inaccessible materials provided by their
               surgical teams. Yet, even as the available number of existing resources targeted toward breast reconstruction
               continues to grow, most are not written at a broadly accessible level . This calls for the need to improve
                                                                          [48]
               this process, which will not only elevate immediate patient experience, but may also lead to overall increased
               patient satisfaction with their breast reconstruction. AI chatbots in their current state have shown the ability
               to perform such tasks effectively and could be easily integrated to bridge this gap with almost immediate
                    [49]
               effect . Notably, a concern in validating AI software is that it is limited by the biases of the datasets on
               which it is built, including those disproportionately affecting populations that are poorly represented even
               in large clinical trials. With even high-level evidence succumbing to this weakness [50,51] , it is important that
               clinically facing tools are continuously refined to ensure high-quality care.


               CONCLUSION
               The integration of AI in breast reconstruction holds great promise for improving surgical outcomes and
               patient satisfaction. AI-driven image analysis, predictive modeling, and personalized planning tools enhance
               the precision and efficiency of surgical planning. Additionally, VR and AR technologies provide innovative
               ways to educate and engage patients, helping them make informed decisions about their care. As AI
               technologies continue to evolve, their applications in breast reconstruction are likely to expand, offering
               even greater benefits to patients and surgeons alike. Overcoming current limitations, including costs,
               accuracy, and reliability, and further evaluating these applications through prospective, randomized, and
               large-scale clinical trials will be necessary to leverage AI’s full potential to improve surgical care and patient
               outcomes in breast reconstruction.


               DECLARATIONS
               Authors’ contributions
               Conceptualization, drafting and editing of the manuscript, technical support, administrative support: Shen
               Y, Malek AJ
               Drafting and editing of the manuscript, technical support: Gao R
               Conceptualization, editing of the manuscript, administrative support: Broyles JM

               Availability of data and materials
               Not applicable.

               Financial support and sponsorship
               None.

               Conflicts of interest
               Broyles JM serves on the scientific advisory board for Healshape LLC, receives consulting support from The
               Agency for Healthcare Research and Quality, and serves on the Medical Device Advisory Board for the
               Food and Drug Administration. The other authors declared that there are no conflicts of interest.
   109   110   111   112   113   114   115   116   117   118   119