Risk prediction in surgery using case-based reasoning and agent-based modelization

Managing the risks arising from actions and conditions of various elements that make up an operating room is a major concern during surgical procedure. One main challenges to define alert thresholds in non-deterministic context where unpredictable adverse events occur. In response this problematic, paper presents architecture couples Multi-Agent System (MAS) with Case-Based Reasoning (CBR). The possibility emulating large number situations thanks MAS, combined analytical data management CBR, original efficient way determining are not defined priori. We also compared different similarity calculation methods (Retrieve phase CBR). results presented article show our model can manage environment manages as disparate infectious agents, patient's vitals human fatigue. addition, they reveal proposed by system more than predefined ones. These tend prove simulator effective generator. Nevertheless, remains simulation mode we would like enrich real from, for example, monitoring sensors (bracelet fatigue, monitoring, etc).