Artificial Intelligence in Upper Gastrointestinal Endoscopy

Tokat, Meltem; Tilburg, Laurèlle van; Koch, Arjun D.; Spaander, Manon C.W.

doi:10.1159/000518232

Cited by 17 publications

(14 citation statements)

References 70 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Te past fve years witness a series of studies assessing the performance of DL in the diagnosis of esophageal diseases [16][17][18][19][20][21][22]. Te main application is the computer vision task, consisting of the detection and segmentation lesions in esophageal endoscopic images or video [23,24]. Te CAD system is designed to detect and diferentiate lesions based on the mucosal/ vascular pattern, to stratify the progression of the diseases or to assist the decision-making of therapy [20,25,26].…”

Section: Discussionmentioning

confidence: 99%

Development and Validation of Deep Learning Models for the Multiclassification of Reflux Esophagitis Based on the Los Angeles Classification

Zhou

Wang

et al. 2023

Journal of Healthcare Engineering

View full text Add to dashboard Cite

This study is to evaluate the feasibility of deep learning (DL) models in the multiclassification of reflux esophagitis (RE) endoscopic images, according to the Los Angeles (LA) classification for the first time. The images were divided into three groups, namely, normal, LA classification A + B, and LA C + D. The images from the HyperKvasir dataset and Suzhou hospital were divided into the training and validation datasets as a ratio of 4 : 1, while the images from Jintan hospital were the independent test set. The CNNs- or Transformer-architectures models (MobileNet, ResNet, Xception, EfficientNet, ViT, and ConvMixer) were transfer learning via Keras. The visualization of the models was proposed using Gradient-weighted Class Activation Mapping (Grad-CAM). Both in the validation set and the test set, the EfficientNet model showed the best performance as follows: accuracy (0.962 and 0.957), recall for LA A + B (0.970 and 0.925) and LA C + D (0.922 and 0.930), Marco-recall (0.946 and 0.928), Matthew’s correlation coefficient (0.936 and 0.884), and Cohen’s kappa (0.910 and 0.850), which was better than the other models and the endoscopists. According to the EfficientNet model, the Grad-CAM was plotted and highlighted the target lesions on the original images. This study developed a series of DL-based computer vision models with the interpretable Grad-CAM to evaluate the feasibility in the multiclassification of RE endoscopic images. It firstly suggests that DL-based classifiers show promise in the endoscopic diagnosis of esophagitis.

show abstract

Section: Discussionmentioning

confidence: 99%

Development and Validation of Deep Learning Models for the Multiclassification of Reflux Esophagitis Based on the Los Angeles Classification

Zhou

Wang

et al. 2023

Journal of Healthcare Engineering

View full text Add to dashboard Cite

show abstract

“…Compared to OGD, more research is focused on colonoscopy. Some recent reviews highlighted a few works from selected groups on upper and lower GI 25 , 26 , 30 ; however, the distinction between train-test dataset or type-of learning-based method used in these studies or both was not presented. A more generic CADe and CADx systems with deep learning (DL) terms were used in the presentation of most found review papers.…”

Section: Methodsmentioning

confidence: 99%

“…However, there is a clear opportunity and need for similar developments of these computer-assistive technologies in all endoscopic procedures to improve patient care. For example, tackling gastrointestinal pathology using AI has an overwhelming number of papers 25 , 31 (see section ‘Computer-aided gastrointestinal endoscopy’). In contrast, despite cystoscopy and ureteroscopy procedures being equally challenging, literature shows minimal work reported so far 66 , 71 .…”

Section: Conclusion and Directionsmentioning

confidence: 99%

“…Most previous review works on artificial intelligence (AI)-driven applications for endoscopic image analysis are published in clinical journals (22 versus only 11 published in the technical journal from 2020 to 2022). The clinical review papers are mostly focused on CADe and CADx systems for lesions in GI endoscopy 25 , 26 , while the technical review articles are mostly concentrated around the laparoscopic surgery 27 , 28 . Reviews concerning upper GI (Barrett’s oesophagus, squamous cell carcinoma (SCC) and stomach 25 , 29 ) and lower GI disease (colorectal polyps 26 , 30 , 31 , and ulcerative colitis 32 , 33 ) are dominant compared to other organs.…”

Section: Introductionmentioning

confidence: 99%

“…The clinical review papers are mostly focused on CADe and CADx systems for lesions in GI endoscopy 25 , 26 , while the technical review articles are mostly concentrated around the laparoscopic surgery 27 , 28 . Reviews concerning upper GI (Barrett’s oesophagus, squamous cell carcinoma (SCC) and stomach 25 , 29 ) and lower GI disease (colorectal polyps 26 , 30 , 31 , and ulcerative colitis 32 , 33 ) are dominant compared to other organs. In addition, even though some reviews’ titles were generic 26 , 34 , they only presented studies from GI endoscopic surveillance and did not cover other areas of endoscopic image analysis.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Where do we stand in AI for endoscopic image analysis? Deciphering gaps and future directions

Ali

2022

npj Digit. Med.

View full text Add to dashboard Cite

Recent developments in deep learning have enabled data-driven algorithms that can reach human-level performance and beyond. The development and deployment of medical image analysis methods have several challenges, including data heterogeneity due to population diversity and different device manufacturers. In addition, more input from experts is required for a reliable method development process. While the exponential growth in clinical imaging data has enabled deep learning to flourish, data heterogeneity, multi-modality, and rare or inconspicuous disease cases still need to be explored. Endoscopy being highly operator-dependent with grim clinical outcomes in some disease cases, reliable and accurate automated system guidance can improve patient care. Most designed methods must be more generalisable to the unseen target data, patient population variability, and variable disease appearances. The paper reviews recent works on endoscopic image analysis with artificial intelligence (AI) and emphasises the current unmatched needs in this field. Finally, it outlines the future directions for clinically relevant complex AI solutions to improve patient outcomes.

show abstract

Ascending the “Hill” of Artificial Intelligence in Upper Gastrointestinal Endoscopy

Simsek,

Parsa,

Kunovsky

2024

UEG Journal

View full text Add to dashboard Cite

Artificial Intelligence in Upper Gastrointestinal Endoscopy

Cited by 17 publications

References 70 publications

Development and Validation of Deep Learning Models for the Multiclassification of Reflux Esophagitis Based on the Los Angeles Classification

Development and Validation of Deep Learning Models for the Multiclassification of Reflux Esophagitis Based on the Los Angeles Classification

Where do we stand in AI for endoscopic image analysis? Deciphering gaps and future directions

Ascending the “Hill” of Artificial Intelligence in Upper Gastrointestinal Endoscopy

Contact Info

Product

Resources

About