Modeling of Surgical Procedures Using Statecharts for Semi-Autonomous Robotic Surgery

In this paper we propose a new methodology to model surgical procedures that is specifically tailored semi-autonomous robotic surgery. We use restricted version of statecharts merge the bottom-up approach, based on data-driven techniques (e.g., machine learning), with top-down approach knowledge representation techniques. consider medical about procedure and sensing environment in two concurrent regions facilitate re-usability adaptability modules. Our allows producing well defined procedural exploiting hierarchy capability statecharts, while learning modules act as soft sensors trigger state transitions. Integrating data driven prior provides robust, modular, flexible re-configurable define which comprehensible by both humans machines. validate our three phases Robot-Assisted Radical Prostatectomy (RARP) directly involve assistant surgeon: bladder mobilization, neck transection, vesicourethral anastomosis, all performed synthetic manikins.