Publications by authors named "K N Fockens"
Gastrointestinal endoscopic image analysis presents significant challenges, such as considerable variations in quality due to the challenging in-body imaging environment, the often-subtle nature of abnormalities with low interobserver agreement, and the need for real-time processing. These challenges pose strong requirements on the performance, generalization, robustness and complexity of deep learning-based techniques in such safety-critical applications. While Convolutional Neural Networks (CNNs) have been the go-to architecture for endoscopic image analysis, recent successes of the Transformer architecture in computer vision raise the possibility to update this conclusion.
View Article and Find Full Text PDFArticle Synopsis
- Pre-training deep learning models for endoscopic image analysis using large datasets like ImageNet is common, but it's unclear if natural image features are the best for medical tasks due to the lack of quality medical imagery.
- This study introduces a new dataset called GastroNet-5M, with over 5 million gastrointestinal endoscopic images, and tests the effectiveness of in-domain pre-training using methods like DINO against pre-training on natural images.
- Results show that self-supervised pre-training with DINO significantly improves model performance on various gastrointestinal tasks, achieving an average boost of 1.63% for ResNet50 and 4.62% for Vision-Transformer-small compared to models trained on natural images.
Background And Aims: Characterization of visible abnormalities in patients with Barrett's esophagus (BE) can be challenging, especially for inexperienced endoscopists. This results in suboptimal diagnostic accuracy and poor interobserver agreement. Computer-aided diagnosis (CADx) systems may assist endoscopists.
View Article and Find Full Text PDFBackground And Aims: This pilot study evaluated the performance of a recently developed computer-aided detection (CADe) system for Barrett's neoplasia during live endoscopic procedures.
Methods: Fifteen patients with a visible lesion and 15 without were included in this study. A CAD-assisted workflow was used that included a slow pullback video recording of the entire Barrett's segment with live CADe assistance, followed by CADe-assisted level-based video recordings every 2 cm of the Barrett's segment.
Background: Computer-aided detection (CADe) systems could assist endoscopists in detecting early neoplasia in Barrett's oesophagus, which could be difficult to detect in endoscopic images. The aim of this study was to develop, test, and benchmark a CADe system for early neoplasia in Barrett's oesophagus.
Methods: The CADe system was first pretrained with ImageNet followed by domain-specific pretraining with GastroNet.