<< Back to the abstract archive
Artificial Intelligence Approach To Soft Tissue Deformation Modeling For Surgery: A Systematic Review
Berk B. Ozmen, Sean Doherty, Ahmet Erdemir, Graham Schwarz
Cleveland Clinic
2024-02-01
Presenter: Berk B. Ozmen
Affidavit:
Yes.
Director Name: Graham Schwarz
Author Category: Fellow Plastic Surgery
Presentation Category: Clinical
Abstract Category: General Reconstruction
BACKGROUND: Accurate modeling of soft tissue deformation holds key potential for novel surgical innovations. However, traditional methods for tissue deformation modeling are not suitable for real-time surgical use due to lack of computational efficiency. Artificial intelligence (AI) offers to overcome many data driven challenges and can expedite the innovations. The purpose of this study is to conduct a systematic review to determine the extent of currently available published research on soft tissue deformation modeling with AI for surgery.
METHODS: We conducted a systematic review following the PRISMA 2020 guidelines, reviewing literature on soft tissue deformation modeling with AI for surgery. We searched for relevant literature on databases including PubMed, Elsevier Scopus, IEEE Xplore, and the Cochrane Library from inception of databases to May 8, 2023.
RESULTS: We identified 73 potential articles for our study, and ultimately included 10 research studies in our systematic review. Included studies were published between 2003 and 2023 and primarily focused on the application of various AI approaches to model soft tissue deformation for surgical simulation. Cellular Neural Networks (30%) and Convolutional Neural Networks (CNN) (20%) were the most frequently utilized AI algorithms. The datasets used for training and evaluating AI models varied significantly and were mainly derived from Finite Element Method (FEM) (60%). Majority of the research studies (60%) incorporated real-time processing for surgical simulation use cases.
CONCLUSION: The findings emphasize the successful use of AI in soft tissue deformation modeling, highlighting its potential for innovative applications in real-time surgical use cases.
