Object-Centric Process Mining of Laparoscopic Cholecystectomy Workflows: Identifying Bottlenecks and Optimization...

Area of Science:

• Medical Informatics
• Process Mining
• Surgical Workflow Analysis

Background:

• Traditional process mining struggles with complex workflows involving multiple entities.
• Object-centric process mining (OCPM) offers a novel approach for analyzing multi-entity processes.
• Laparoscopic cholecystectomy (LC) procedures present a complex, multi-entity surgical workflow.

Purpose of the Study:

• To construct an Object-Centric Event Log (OCEL) for LC procedures.
• To discover multi-entity process patterns and identify workflow bottlenecks.
• To analyze the impact of patient complexity on perioperative dynamics.

Main Methods:

• Analysis of an OCEL from 1,186 LC cases using PM4Py.
• Discovery of object-centric directly-follows graphs (OC-DFG) and variant exploration.
• Statistical analysis of ICU admission and ASA classification effects on perioperative duration.

Main Results:

• 92% adherence to the reference clinical pathway was observed.
• Intensive care unit (ICU) admitted patients had significantly longer perioperative durations (13.83h vs. 8.85h).
• Major bottlenecks identified in booking-to-OR transfer, pre-anesthesia, and PACU discharge.

Conclusions:

• OCPM provides unique multi-perspective insights into surgical workflows.
• Preoperative scheduling and postoperative discharge are key areas for improvement.
• Targeted interventions can optimize surgical throughput and patient flow.