Abstract
641 ARTIFICIAL INTELLIGENCE DYSPLASIA DETECTION (AIDD) ALGORITHM FOR BARRETT’S ESOPHAGUS
Gastrointestinal endoscopy, Vol.89(6 Supplement), pp.AB99-AB100
06/2019
DOI: 10.1016/j.gie.2019.04.095
Abstract
Background
The gold standard and most widely used approach for screening and surveillance of Barrett’s esophagus (BE) is esophagogastroduodenoscopy. However, the visual detection of early esophageal neoplasia (high grade dysplasia and T1 stage adenocarcinoma) in BE with white light and electronic virtual chromoendoscopy is still often difficult. The aim of this study is to assess if a convolutional neural artificial intelligence network can aid in the recognition of early esophageal neoplasia in BE.
Methods
Over 800 images from 65 patients were retrospectively collected of histology-proven early esophageal neoplasia in BE containing high grade dysplasia or T1 stage adenocarcinoma (Dysplasia Group). Within each image, the area of neoplasia was masked using image annotation software by two experts in Barrett’s esophagus imaging (Fig 1). Over 800 control images were collected of either histology-proven or confocal endomicroscopy-proven BE without high grade dysplasia (Non -Dysplastic Group).
A training set with ∼1200 images split 50/50 Dysplasia/Non-Dysplasia was used to train the algorithm. We used a convolutional neural network (CNN) and hybrid algorithm design including Inception blocks to deepen the neural net and maximize efficiency and accuracy. The algorithm was pre-trained on ImageNet and then fine-tuned with the goal to provide the correct binary classification: “Dysplastic” (1) or “Non-dysplastic” (0). Adam optimizer performed stochastic optimization of a binary cross-entropy loss function to produce a probability value between 0 and 1. A set 458 images unique of the training set was used for algorithm validation.
Results
The CNN analyzed 458 test images (225 dysplasia/233 non-dysplasia) and correctly detected early neoplasia in BE cases with sensitivity of 95.6% and specificity of 91.8%. The accuracy was 93.7% and the AUC was 0.94 (Fig.2).
Conclusion
This early Artificial Intelligence algorithm using convolutional neural networks was able to detect early esophageal neoplasia in Barrett’s Esophagus images with 93.7% accuracy and an area under the curve of 0.94. This algorithm appears promising and may increase an endoscopist's sensitivity to detect early neoplasia during surveillance endoscopy of Barrett’s esophagus. Further real-time live video validation of the algorithm is needed and is currently underway.
Details
- Title: Subtitle
- 641 ARTIFICIAL INTELLIGENCE DYSPLASIA DETECTION (AIDD) ALGORITHM FOR BARRETT’S ESOPHAGUS
- Creators
- Rintaro Hashimoto - University of California, IrvineMichael Lugo - University of California, IrvineDaniel Mai - University of California, IrvineNabil E. Chehade - University of California, IrvineElise Tran - University of California, IrvineTyler Dao - University of California, IrvineJohn Lee - University of California, IrvineKenneth J. Chang - University of California, IrvineAndrew Ninh - University of California, IrvineJames Requa - University of California, IrvineWilliam E. Karnes - University of California, IrvineJason B. Samarasena - University of California, Irvine
- Resource Type
- Abstract
- Publication Details
- Gastrointestinal endoscopy, Vol.89(6 Supplement), pp.AB99-AB100
- Publisher
- MOSBY-ELSEVIER
- DOI
- 10.1016/j.gie.2019.04.095
- ISSN
- 0016-5107
- eISSN
- 1097-6779
- Language
- English
- Date published
- 06/2019
- Academic Unit
- Internal Medicine
- Record Identifier
- 9984697050202771
Metrics
1 Record Views