Prospective development and validation of a volumetric laser endomicroscopy computer algorithm for detection of Barrett's neoplasia

TY - JOUR

T1 - Prospective development and validation of a volumetric laser endomicroscopy computer algorithm for detection of Barrett's neoplasia

AU - Struyvenberg, Maarten R.

AU - de Groof, Albert J.

AU - Fonollà, Roger

AU - van der Sommen, Fons

AU - de With, Peter H.N.

AU - Schoon, Erik J.

AU - Weusten, Bas L.A.M.

AU - Leggett, Cadman L.

AU - Kahn, Allon

AU - Trindade, Arvind J.

AU - Ganguly, Eric K.

AU - Konda, Vani J.A.

AU - Lightdale, Charles J.

AU - Pleskow, Douglas K.

AU - Sethi, Amrita

AU - Smith, Michael S.

AU - Wallace, Michael B.

AU - Wolfsen, Herbert C.

AU - Tearney, Gary J.

AU - Meijer, Sybren L.

AU - Vieth, Michael

AU - Pouw, Roos E.

AU - Curvers, Wouter L.

AU - Bergman, Jacques J.

N1 - Publisher Copyright: © 2021 American Society for Gastrointestinal Endoscopy

PY - 2021/4

Y1 - 2021/4

N2 - Background and Aims: Volumetric laser endomicroscopy (VLE) is an advanced imaging modality used to detect Barrett's esophagus (BE) dysplasia. However, real-time interpretation of VLE scans is complex and time-consuming. Computer-aided detection (CAD) may help in the process of VLE image interpretation. Our aim was to train and validate a CAD algorithm for VLE-based detection of BE neoplasia. Methods: The multicenter, VLE PREDICT study, prospectively enrolled 47 patients with BE. In total, 229 nondysplastic BE and 89 neoplastic (high-grade dysplasia/esophageal adenocarcinoma) targets were laser marked under VLE guidance and subsequently underwent a biopsy for histologic diagnosis. Deep convolutional neural networks were used to construct a CAD algorithm for differentiation between nondysplastic and neoplastic BE tissue. The CAD algorithm was trained on a set consisting of the first 22 patients (134 nondysplastic BE and 38 neoplastic targets) and validated on a separate test set from patients 23 to 47 (95 nondysplastic BE and 51 neoplastic targets). The performance of the algorithm was benchmarked against the performance of 10 VLE experts. Results: Using the training set to construct the algorithm resulted in an accuracy of 92%, sensitivity of 95%, and specificity of 92%. When performance was assessed on the test set, accuracy, sensitivity, and specificity were 85%, 91%, and 82%, respectively. The algorithm outperformed all 10 VLE experts, who demonstrated an overall accuracy of 77%, sensitivity of 70%, and specificity of 81%. Conclusions: We developed, validated, and benchmarked a VLE CAD algorithm for detection of BE neoplasia using prospectively collected and biopsy-correlated VLE targets. The algorithm detected neoplasia with high accuracy and outperformed 10 VLE experts. (The Netherlands National Trials Registry (NTR) number: NTR 6728.)

AB - Background and Aims: Volumetric laser endomicroscopy (VLE) is an advanced imaging modality used to detect Barrett's esophagus (BE) dysplasia. However, real-time interpretation of VLE scans is complex and time-consuming. Computer-aided detection (CAD) may help in the process of VLE image interpretation. Our aim was to train and validate a CAD algorithm for VLE-based detection of BE neoplasia. Methods: The multicenter, VLE PREDICT study, prospectively enrolled 47 patients with BE. In total, 229 nondysplastic BE and 89 neoplastic (high-grade dysplasia/esophageal adenocarcinoma) targets were laser marked under VLE guidance and subsequently underwent a biopsy for histologic diagnosis. Deep convolutional neural networks were used to construct a CAD algorithm for differentiation between nondysplastic and neoplastic BE tissue. The CAD algorithm was trained on a set consisting of the first 22 patients (134 nondysplastic BE and 38 neoplastic targets) and validated on a separate test set from patients 23 to 47 (95 nondysplastic BE and 51 neoplastic targets). The performance of the algorithm was benchmarked against the performance of 10 VLE experts. Results: Using the training set to construct the algorithm resulted in an accuracy of 92%, sensitivity of 95%, and specificity of 92%. When performance was assessed on the test set, accuracy, sensitivity, and specificity were 85%, 91%, and 82%, respectively. The algorithm outperformed all 10 VLE experts, who demonstrated an overall accuracy of 77%, sensitivity of 70%, and specificity of 81%. Conclusions: We developed, validated, and benchmarked a VLE CAD algorithm for detection of BE neoplasia using prospectively collected and biopsy-correlated VLE targets. The algorithm detected neoplasia with high accuracy and outperformed 10 VLE experts. (The Netherlands National Trials Registry (NTR) number: NTR 6728.)

UR - https://www.scopus.com/pages/publications/85095854894

UR - https://www.scopus.com/pages/publications/85095854894#tab=citedBy

U2 - 10.1016/j.gie.2020.07.052

DO - 10.1016/j.gie.2020.07.052

M3 - Article

C2 - 32735947

SN - 0016-5107

VL - 93

SP - 871

EP - 879

JO - Gastrointestinal Endoscopy

JF - Gastrointestinal Endoscopy

IS - 4

ER -