Reliability of sensor-based real-time workflow recognition in laparoscopic cholecystectomy
Michael Kranzfelder1,2, Armin Schneider3, Adam Fiolka3
1Department of Surgery, Technische Universität München, 81675, Munich, Germany. michael.kranzfelder@tum.de.
International Journal of Computer Assisted Radiology and Surgery
|February 22, 2014
Summary
Sensor data reliably identified procedural tasks during laparoscopic cholecystectomy. Expert observers could accurately determine each step, paving the way for future automated surgical analysis.
Area of Science:
- Minimally Invasive Surgery
- Surgical Workflow Analysis
- Medical Device Technology
Background:
- Laparoscopic cholecystectomy is a common procedure.
- Assistance systems can potentially improve surgical outcomes.
- Model-based surgery offers a framework for analyzing surgical processes.
Purpose of the Study:
- To implement and test a model-based surgery approach for laparoscopic cholecystectomy.
- To define and analyze distinct procedural tasks (PTs) within the operation.
- To evaluate the reliability of real-time workflow recognition using sensor data.
Main Methods:
- Defined ten procedural tasks (PTs) for laparoscopic cholecystectomy.
- Acquired continuous sensor data, including instrument detection, room/table light status, intra-abdominal pressure, table tilt, irrigation/aspiration volume, and coagulation/cutting current.
- Utilized two independent observers to record PT start/endpoints from sensor data, cross-checked with laparoscopic video recordings.
Main Results:
- Bland-Altman analysis showed agreement within limits for 95% of cases.
- Sensor data and laparoscopic video generally matched across different PTs.
- Observer results from sensor data often exceeded those from video analysis, though empirical knowledge was needed for phase transitions.
Conclusions:
- A sensor set was sufficient for expert observers to reliably identify procedural tasks in laparoscopic cholecystectomy.
- Future computer systems could automate task identification with improved data inflow.
- This study supports the development of advanced surgical assistance systems.
