A comprehensive analysis of classification methods in gastrointestinal endoscopy imaging

Jha, Debesh; Ali, Sharib; Hicks, Steven A.; Thambawita, Vajira; Borgli, Hanna; Smedsrud, Pia H.; Lange, Thomas de; Pogorelov, Konstantin; Wang, Xiaowei; Harzig, Philipp; Tran, Minh–Triet; Meng, Wenhua; Hoàng, Trung-Hiếu; Dias, Danielle; Ko, Tobey H.; Agrawal, Taruna; Ostroukhova, Olga; Khan, Zeshan Aslam; Tahir, Muhammad Atif; Liu, Yang; Chang, Yuan; Kirkerød, Mathias; Johansen, Dag; Lux, Mathias; Riegler, Michael; Halvorsen, Pål

doi:10.1016/j.media.2021.102007

Cited by 21 publications

(8 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similar studies have been proposed for anatomical site segmentation in colonoscopy. A longitudinal analysis of the performances of reference CNN approaches 65 on three reference colonoscopy image classification challenges [66][67][68] has recently been performed. The most recent challenges [68][69][70] aim at evaluating image classification methods for distinguishing between anatomical location (z-line, pylorus, cecum, retroflex rectum, retroflex stomach), abnormalities (polyps, ulcerative colitis), polyp removal cases (dyed and lifted polyps, dyed resection margins), and surgical context (normal colon mucosa, moderate stool inclusion, significant stool inclusion, useless blurry image, surgical instrument detected, out of patient).…”

Section: C2 Anatomical Structure Recognitionmentioning

confidence: 99%

Artificial intelligence and automation in endoscopy and surgery

Chadebecq

Lovat

Stoyanov

2022

Nat Rev Gastroenterol Hepatol

View full text Add to dashboard Cite

Modern endoscopic procedures rely on digital technology ranging from high resolution imaging sensors and displays through to electronics connecting configurable illumination and actuation systems for robotic articulation. In addition to enabling more effective diagnostic and therapeutic interventions, the digitization of the procedural toolset also enables video data capture of the internal human anatomy at unprecedented levels. Interventional video data encapsulates functional and structural information about the patients' anatomy, events, activity as well as action log about surgical process. This detailed but difficult to interpret record from endoscopic procedures can potentially be linked to pre-and post-operative records or patient imaging information.Rapid advances in Artificial Intelligence (AI), especially in supervised deep learning, can utilize data from endoscopic procedures to develop systems for assisting procedures leading to computer assisted interventions (CAI). CAI systems can provide a wide range of enhanced capabilities, for example better navigation during procedures, automation of image interpretation or even of robotically assisted tool manipulation. In this review, we summarize the state-of-theart AI for CAI in gastroenterology and surgery.

show abstract

Section: C2 Anatomical Structure Recognitionmentioning

confidence: 99%

Artificial intelligence and automation in endoscopy and surgery

Chadebecq

Lovat

Stoyanov

2022

Nat Rev Gastroenterol Hepatol

View full text Add to dashboard Cite

show abstract

“…These studies aim to achieve a general classi˝cation of the di˙erent GI ˝ndings that can appear during endoscopy instead of concentrating on a particular su˙ering or symptom. For a detailed review of papers using AI to classify images of the gastrointestinal tract, refer to Jha et.al [10].…”

Section: Classi˝cation Of Endoscopy Imagesmentioning

confidence: 99%

“…Numerous studies have stated their concerns about the relevance of the existing DL models in real-world applications. The focal limitations found for these systems are their interpretability scarcity [2,9,16,20] (often referred to as the "black box" condition [11,21,25]) and questionable generalizability [2,20,26,28].…”

Section: Introductionmentioning

confidence: 99%

Towards an Interpretable Model for Automatic Classification of Endoscopy Images

García-Aguirre¹,

Torres-Treviño²,

Navarro-López³

et al. 2022

Advances in Computational Intelligence

View full text Add to dashboard Cite

Deep learning strategies have become the mainstream for computer-assisted diagnosis tools development since they outperform other machine learning techniques. However, these systems can not reach their full potential since the lack of understanding of their operation and questionable generalizability provokes mistrust from the users, limiting their application. In this paper, we generate a Convolutional Neural Network (CNN) using a genetic algorithm for hyperparameter optimization. Our CNN has state-of-the-art classi˝cation performance, delivering higher evaluation metrics than other recent papers that use AI models to classify images from the same dataset. We provide visual explanations of the classi˝cations made by our model implementing Grad-CAM and analyze the behavior of our model on misclassi˝cations using this technique.

show abstract

“…They further extended their study to classify and evaluate upper, middle, and lower stomachs using a validation set of 13,048 items, achieving a four-class classification accuracy of 97% and a three-position study accuracy of 99%. As indicated above, while studies exist on the identification of upper gastrointestinal landmarks through wired conventional endoscopy, research on upper gastrointestinal landmarks with WCE images has only recently emerged [ 17 , 18 ]. Further investigations are needed for landmark identification in this specific context.…”

Section: Introductionmentioning

confidence: 99%

Color-Transfer-Enhanced Data Construction and Validation for Deep Learning-Based Upper Gastrointestinal Landmark Classification in Wireless Capsule Endoscopy

Kim,

Cho,

Park

et al. 2024

Diagnostics

View full text Add to dashboard Cite

While the adoption of wireless capsule endoscopy (WCE) has been steadily increasing, its primary application remains limited to observing the small intestine, with relatively less application in the upper gastrointestinal tract. However, there is a growing anticipation that advancements in capsule endoscopy technology will lead to a significant increase in its application in upper gastrointestinal examinations. This study addresses the underexplored domain of landmark identification within the upper gastrointestinal tract using WCE, acknowledging the limited research and public datasets available in this emerging field. To contribute to the future development of WCE for gastroscopy, a novel approach is proposed. Utilizing color transfer techniques, a simulated WCE dataset tailored for the upper gastrointestinal tract is created. Using Euclidean distance measurements, the similarity between this color-transferred dataset and authentic WCE images is verified. Pioneering the exploration of anatomical landmark classification with WCE data, this study integrates similarity evaluation with image preprocessing and deep learning techniques, specifically employing the DenseNet169 model. As a result, utilizing the color-transferred dataset achieves an anatomical landmark classification accuracy exceeding 90% in the upper gastrointestinal tract. Furthermore, the application of sharpen and detail filters demonstrates an increase in classification accuracy from 91.32% to 94.06%.

show abstract

A comprehensive analysis of classification methods in gastrointestinal endoscopy imaging

Cited by 21 publications

References 22 publications

Artificial intelligence and automation in endoscopy and surgery

Artificial intelligence and automation in endoscopy and surgery

Towards an Interpretable Model for Automatic Classification of Endoscopy Images

Color-Transfer-Enhanced Data Construction and Validation for Deep Learning-Based Upper Gastrointestinal Landmark Classification in Wireless Capsule Endoscopy

Contact Info

Product

Resources

About