Automated Support of Robotic Surgical Training, Operations, and Outcomes

Background: Surgical error reduction requires addressing rapid adoption of new surgical technology. We propose development of a networked surgical assessment platform able to collect performance and outcomes data from multiple surgical platforms and sites for dynamic metrics analysis. Goal: Provide surgical signature ("Surgome") of each surgeon by delivering automated performance feedback to accelerate learning curves and deriving summative and statistically valid trends in data for comparative effectiveness of new technologies. Task 1: Algorithm Development Deliverables: 1) Documentation and implementation to mathematically define metrics for rating evaluation algorithms. 2) Validation of machine learning algorithm (Autoregressive Dynamic Bayesian Network) within framework. 3) Prototype software applying this algorithm to existing surgical performance datasets. Task 2:Server Design and Implementation Deliverables: 1) On-line server running open-source operating system and database engine for storing performance data. 2) Importing and generation of full metadata for three datasets. 3) Preliminary dynamic metrics analysis software able to compute performance scores from new surgical performance data sets.