A computer-assisted algorithm for narrow-band-imaging–based tissue characterization in Barrett’s esophagus

Maarten R. Struyvenberg (Corresponding author), Albert J. De Groof (Corresponding author), Joost van der Putten, Fons van der Sommen, Francisco Baldaque-silva, Masami Omae, Roos Pouw, Raf Bisschops, Michael Vieth, Erik J. Schoon, Wouter L. Curvers, Peter H. de With, Jacques J. Bergman (Corresponding author)

Research output: Contribution to journalArticleAcademicpeer-review

43 Citations (Scopus)
18 Downloads (Pure)


Background and Aims
The endoscopic evaluation of narrow-band imaging (NBI)-zoom imagery in Barrett’s esophagus (BE) is associated with suboptimal diagnostic accuracy and poor interobserver agreement. Computer-aided diagnosis (CAD) systems may assist endoscopists in the characterization of Barrett’s mucosa. Our aim was to demonstrate feasibility of a deep-learning CAD system for tissue characterization of NBI-zoom imagery in BE.

The CAD system was first trained using 494,364 endoscopic images of general endoscopic imagery. Next, 690 neoplastic BE and 557 nondysplastic (ND)BE white-light endoscopy overview images were used for refinement training. Subsequently, a third dataset of 112 neoplastic and 71 NDBE NBI-zoom images with histological correlation was used for training and internal validation. Finally, the CAD system was further trained and validated with a fourth, histologically confirmed, dataset of 59 neoplastic and 98 NDBE NBI-zoom videos. Performance was evaluated using fourfold cross-validation. Primary outcome was the diagnostic performance of the CAD system for classification of neoplasia in NBI-zoom videos.

The CAD system demonstrated an accuracy, sensitivity and specificity for detection of BE neoplasia using NBI-zoom images of 84%, 88%, and 78%, respectively. In total 30,021 individual video frames were analyzed by the CAD system. Accuracy, sensitivity and specificity of the video-based CAD system were 83% (95% CI, 78%-89%), 85% (95% CI, 76%-94%) and 83% (95% CI, 76%-90%), respectively. Mean assessment speed was 38 frames per second.

We have demonstrated promising diagnostic accuracy of predicting the presence/absence of Barrett’s neoplasia on histologically confirmed unaltered NBI-zoom videos with fast corresponding assessment time.
Original languageEnglish
Pages (from-to)89-98
Number of pages10
JournalGastrointestinal Endoscopy
Issue number1
Early online date3 Jun 2020
Publication statusPublished - Jan 2021


Dive into the research topics of 'A computer-assisted algorithm for narrow-band-imaging–based tissue characterization in Barrett’s esophagus'. Together they form a unique fingerprint.

Cite this