CC BY-NC-ND 4.0 · Endosc Int Open 2018; 06(08): E1044-E1050
DOI: 10.1055/a-0627-7136
Original article
Owner and Copyright © Georg Thieme Verlag KG 2018

Assessment of bowel cleansing quality in colon capsule endoscopy using machine learning: a pilot study

Maria Magdalena Buijs
1  Department of Surgery, Odense University Hospital, Svendborg, Denmark
2  Institute of Clinical Research, University of Southern Denmark, Odense, Denmark
,
Mohammed Hossain Ramezani
3  Mads Clausen Institute, University of Southern Denmark, Sønderborg, Denmark
,
Jürgen Herp
4  Applied Statistical Signal Processing Group, Embodied Systems for Robotics and Learning, Faculty of Engineering, University of Southern Denmark, Denmark
,
Rasmus Kroijer
1  Department of Surgery, Odense University Hospital, Svendborg, Denmark
2  Institute of Clinical Research, University of Southern Denmark, Odense, Denmark
,
Morten Kobaek-Larsen
2  Institute of Clinical Research, University of Southern Denmark, Odense, Denmark
,
Gunnar Baatrup
1  Department of Surgery, Odense University Hospital, Svendborg, Denmark
2  Institute of Clinical Research, University of Southern Denmark, Odense, Denmark
,
Esmaeil S. Nadimi
4  Applied Statistical Signal Processing Group, Embodied Systems for Robotics and Learning, Faculty of Engineering, University of Southern Denmark, Denmark
› Author Affiliations
Further Information

Publication History

Publication Date:
10 August 2018 (online)

Abstract

Background and study aims The aim of this study was to develop a machine learning-based model to classify bowel cleansing quality and to test this model in comparison to a pixel analysis model and assessments by four colon capsule endoscopy (CCE) readers.

Methods A pixel analysis and a machine learning-based model with four cleanliness classes (unacceptable, poor, fair and good) were developed to classify CCE videos. Cleansing assessments by four CCE readers in 41 videos from a previous study were compared to the results both models yielded in this pilot study.

Results The machine learning-based model classified 47 % of the videos in agreement with the averaged classification by CCE readers, as compared to 32 % by the pixel analysis model. A difference of more than one class was detected in 12 % of the videos by the machine learning-based model and in 32 % by the pixel analysis model, as the latter tended to overestimate cleansing quality. A specific analysis of unacceptable videos found that the pixel analysis model classified almost all of them as fair or good, whereas the machine learning-based model identified five out of 11 videos in agreement with at least one CCE reader as unacceptable.

Conclusions The machine learning-based model was superior to the pixel analysis in classifying bowel cleansing quality, due to a higher sensitivity to unacceptable and poor cleansing quality. The machine learning-based model can be further improved by coming to a consensus on how to classify cleanliness of a complete CCE video, by means of an expert panel.