RESUMO
OBJECTIVE: Develop a pioneer surgical anonymization algorithm for reliable and accurate real-time removal of out-of-body images validated across various robotic platforms. BACKGROUND: The use of surgical video data has become a common practice in enhancing research and training. Video sharing requires complete anonymization, which, in the case of endoscopic surgery, entails the removal of all nonsurgical video frames where the endoscope can record the patient or operating room staff. To date, no openly available algorithmic solution for surgical anonymization offers reliable real-time anonymization for video streaming, which is also robotic-platform and procedure-independent. METHODS: A data set of 63 surgical videos of 6 procedures performed on four robotic systems was annotated for out-of-body sequences. The resulting 496.828 images were used to develop a deep learning algorithm that automatically detected out-of-body frames. Our solution was subsequently benchmarked against existing anonymization methods. In addition, we offer a postprocessing step to enhance the performance and test a low-cost setup for real-time anonymization during live surgery streaming. RESULTS: Framewise anonymization yielded a receiver operating characteristic area under the curve score of 99.46% on unseen procedures, increasing to 99.89% after postprocessing. Our Robotic Anonymization Network outperforms previous state-of-the-art algorithms, even on unseen procedural types, despite the fact that alternative solutions are explicitly trained using these procedures. CONCLUSIONS: Our deep learning model, Robotic Anonymization Network, offers reliable, accurate, and safe real-time anonymization during complex and lengthy surgical procedures regardless of the robotic platform. The model can be used in real time for surgical live streaming and is openly available.
Assuntos
Algoritmos , Procedimentos Cirúrgicos Robóticos , Humanos , Anonimização de Dados , Gravação em Vídeo , Aprendizado ProfundoRESUMO
BACKGROUND: Minimally invasive total gastrectomy (MITG) is a mainstay for curative treatment of patients with gastric cancer. To define and standardize optimal surgical techniques and further improve clinical outcomes through the enhanced MITG surgical quality, there must be consensus on the key technical steps of lymphadenectomy and anastomosis creation, which is currently lacking. This study aimed to determine an expert consensus from an international panel regarding the technical aspects of the performance of MITG for oncological indications using the Delphi method. METHODS: A 100-point scoping survey was created based on the deconstruction of MITG into its key technical steps through local and international expert opinion and literature evidence. An international expert panel comprising upper gastrointestinal and general surgeons participated in multiple rounds of a Delphi consensus. The panelists voted on the issues concerning importance, difficulty, or agreement using an online questionnaire. A priori consensus standard was set at > 80% for agreement to a statement. Internal consistency and reliability were evaluated using Cronbach's α. RESULTS: Thirty expert upper gastrointestinal and general surgeons participated in three online Delphi rounds, generating a final consensus of 41 statements regarding MITG for gastric cancer. The consensus was gained from 22, 12, and 7 questions from Delphi rounds 1, 2, and 3, which were rephrased into the 41 statetments respectively. For lymphadenectomy and aspects of anastomosis creation, Cronbach's α for round 1 was 0.896 and 0.886, and for round 2 was 0.848 and 0.779, regarding difficulty or importance. CONCLUSIONS: The Delphi consensus defined 41 steps as crucial for performing a high-quality MITG for oncological indications based on the standards of an international panel. The results of this consensus provide a platform for creating and validating surgical quality assessment tools designed to improve clinical outcomes and standardize surgical quality in MITG.