Inter-and Intraobserver Variability in Bowel Preparation Scoring for Colon Capsule Endoscopy: Impact of AI-Assisted Assessment Feasibility Study

This study assessed the reliability of AI-assisted bowel cleansing scoring in colon capsule endoscopy using the CC-CLEAR scale. While interobserver agreement was excellent with manual scoring among experienced readers, AI-assisted reads did not improve agreement but showed reduced consistency, parti...

Descripción completa

Detalles Bibliográficos
Autores: Io Lei, Ian, Gaya, Daniel R., Robertson, Alexander, Schelde-Olesen, Benedicte, Mapiye, Alice, Bhandare, Anirudh, Lui, Bei Bei, Shekhar, Chander, Valentiner, Ursula, Gilabert Roca, Pere, Laiz Treceño, Pablo, Seguí Mesquida, Santi, Parsons, Nicholas, Huhulea, Cristiana, Wenzek, Hagen, White, Elizabeth, Koulaouzidis, Anastasios, Arasaradnam, Ramesh P.
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2025
País:España
Institución:Universidad de Barcelona
Repositorio:Dipòsit Digital de la UB
OAI Identifier:oai:diposit.ub.edu:2445/226894
Acceso en línea:https://hdl.handle.net/2445/226894
Access Level:acceso abierto
Palabra clave:Càpsula endoscòpica
Aprenentatge automàtic
Intel·ligència artificial
Capsule endoscopy
Machine learning
Artificial intelligence
Descripción
Sumario:This study assessed the reliability of AI-assisted bowel cleansing scoring in colon capsule endoscopy using the CC-CLEAR scale. While interobserver agreement was excellent with manual scoring among experienced readers, AI-assisted reads did not improve agreement but showed reduced consistency, particularly among less experienced users. The mean AI-assisted scores were significantly lower than manual scores, highlighting potential interpretive challenges. These findings suggest that AI’s effectiveness currently depends on user expertise, reinforcing the importance of further development and refinement required for a robust AI implementation in CCE.

Similares en LA Referencia