Basic Study
Copyright ©The Author(s) 2016. Published by Baishideng Publishing Group Inc. All rights reserved.
World J Gastroenterol. Aug 21, 2016; 22(31): 7124-7134
Published online Aug 21, 2016. doi: 10.3748/wjg.v22.i31.7124
Computer-aided texture analysis combined with experts' knowledge: Improving endoscopic celiac disease diagnosis
Michael Gadermayr, Hubert Kogler, Maximilian Karla, Dorit Merhof, Andreas Uhl, Andreas Vécsei
Michael Gadermayr, Dorit Merhof, Institute of Imaging and Computer Vision, RWTH Aachen University, D-52074 Aachen, Germany
Hubert Kogler, Maximilian Karla, Andreas Vécsei, Department of Pediatrics, Pediatric Gastroenterology, St, Anna Children’s Hospital, Medical University Vienna, A-1090 Vienna, Austria
Andreas Uhl, Department of Computer Sciences, University of Salzburg, A-5020 Salzburg, Austria
Author contributions: Gadermayr M and Kogler H contributed equally to this work; Gadermayr M, Kogler H, Karla M and Vécsei A jointly wrote the first draft of the manuscript; Furthermore, Gadermayr M, Kogler H, Uhl A and Vécsei A developed the study design and the concept; Gadermayr M and Uhl A developed the statistical analysis plan, interpreted the data, did the statistical analysis; Kogler H, Uhl A and Vécsei A participated in data collection; Uhl A and Vécsei A obtained the funding; Vécsei A supervised the study; all authors revised the manuscript for important intellectual content, read and approved the final manuscript.
Supported by the Austrian Science Fund (FWF), No. KLI 429-B13 to Vécsei A.
Institutional review board statement: The study was reviewed and approved by the Institutional Review Board of the St. Anna Children’s Hospital.
Conflict-of-interest statement: Gadermayr M and Karla M have received research funding of the Austrian Science Fund (FWF). Kogler H, Merhof D, Uhl A and Vécsei A have no financial or other conflict of interest relevant to the subject of this article.
Data sharing statement: Statistical code is available from the corresponding author at Participants gave informed consent for data sharing.
Open-Access: This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See:
Correspondence to: Michael Gadermayr, PhD, Institute of Imaging and Computer Vision, RWTH Aachen University, Kopernikusstraße 16, D-52074 Aachen, Germany.
Telephone: +49-241-8022906 Fax: +49-241-8022200
Received: March 15, 2016
Peer-review started: March 18, 2016
First decision: March 31, 2016
Revised: April 28, 2016
Accepted: May 21, 2016
Article in press: May 23, 2016
Published online: August 21, 2016

AIM: To further improve the endoscopic detection of intestinal mucosa alterations due to celiac disease (CD).

METHODS: We assessed a hybrid approach based on the integration of expert knowledge into the computer-based classification pipeline. A total of 2835 endoscopic images from the duodenum were recorded in 290 children using the modified immersion technique (MIT). These children underwent routine upper endoscopy for suspected CD or non-celiac upper abdominal symptoms between August 2008 and December 2014. Blinded to the clinical data and biopsy results, three medical experts visually classified each image as normal mucosa (Marsh-0) or villous atrophy (Marsh-3). The experts’ decisions were further integrated into state-of-the-art texture recognition systems. Using the biopsy results as the reference standard, the classification accuracies of this hybrid approach were compared to the experts’ diagnoses in 27 different settings.

RESULTS: Compared to the experts’ diagnoses, in 24 of 27 classification settings (consisting of three imaging modalities, three endoscopists and three classification approaches), the best overall classification accuracies were obtained with the new hybrid approach. In 17 of 24 classification settings, the improvements achieved with the hybrid approach were statistically significant (P < 0.05). Using the hybrid approach classification accuracies between 94% and 100% were obtained. Whereas the improvements are only moderate in the case of the most experienced expert, the results of the less experienced expert could be improved significantly in 17 out of 18 classification settings. Furthermore, the lowest classification accuracy, based on the combination of one database and one specific expert, could be improved from 80% to 95% (P < 0.001).

CONCLUSION: The overall classification performance of medical experts, especially less experienced experts, can be boosted significantly by integrating expert knowledge into computer-aided diagnosis systems.

Keywords: Celiac disease, Diagnosis, Endoscopy, Computer-aided texture analysis, Biopsy, Pattern recognition

Core tip: A hybrid system for the detection of villous atrophy integrating human texture recognition into computer-aided diagnosis systems outperforms human judgement alone in the diagnosis of pediatric celiac disease. In the classification of 2835 endoscopic images from the duodenum into one of two categories (“normal mucosa or villous atrophy”) using 27 different classification settings the hybrid system was superior to human experts in 24 settings. This superiority was significant in 17 of these 24 settings. Less experienced endoscopists in particular can benefit from this new method because their diagnostic accuracy can be improved the most.