Development and validation of the ENDOLAP artificial intelligence framework for inflammation severity classification in laparoscopic cholecystectomy: a cross-sectional study

Norman A Rendón Mejía ¹, Said De la Cruz Rey ², Carlos R Cervantes-Sánchez ^3,4

⁰General Surgery Department, Chihuahua City General Hospital "Dr. Salvador Zubirán Anchondo", 510 Cristobal Columbus Avenue, 31000, Chihuahua, Mexico.

Surgical Endoscopy +

|

August 20, 2025

View abstract on PubMed

Summary

This study introduces ENDOLAP-IA, an AI tool that objectively classifies gallbladder inflammation severity during laparoscopic cholecystectomy. The AI framework improves procedural assessment and reduces variability in surgical decision-making.

Area Of Science

Artificial Intelligence in Surgery
Medical Imaging Analysis
Surgical Workflow Optimization

Background

Intraoperative assessment of gallbladder inflammation severity in laparoscopic cholecystectomy is subjective and lacks standardization.
Objective, real-time classification is needed to predict procedural difficulty and potential complications.

Purpose Of The Study

To develop and validate an AI framework (ENDOLAP-IA) for objective, real-time classification of gallbladder inflammation severity.
To assess the AI's ability to predict procedural difficulty and complications during laparoscopic cholecystectomy.

Main Methods

A prospective cohort of 53 patients undergoing elective cholecystectomy.
Development and validation of a 9-item checklist (ENDOLAP-IA tool) for image quality.
Fine-tuning a YOLOv8 AI model on over 2000 intraoperative images annotated per Parkland severity grades.
Validation using fivefold cross-validation and external testing with blinded surgeon assessment.

Main Results

The ENDOLAP tool demonstrated excellent content validity (CVI > 0.85) and inter-rater reliability (ICC = 0.82).
The AI model achieved 87.3% accuracy, with an AUC of 0.923, and eliminated interobserver variability.
External validation showed 82.6% agreement with surgeons and 91.3% sensitivity for severe inflammation.

Conclusions

ENDOLAP-IA is the first AI framework for standardizing intraoperative inflammatory severity classification in laparoscopic cholecystectomy.
The AI framework provides clinically reliable, objective, real-time decision support.
Integration into surgical workflows can enhance safety, training, and comparative outcome analyses.

Keywords:

Artificial intelligence Deep learning Inflammation Laparoscopic cholecystectomy Observer agreement