Multi-expert annotation of Crohn’s disease images of the small bowel for automatic detection using a convolutional recurrent attention neural network

Maissin, Astrid de; Vallée, Remi; Flamant, Mathurin; Fondain-Bossiere, Marie; Berre, Catherine Le; Coutrot, Antoine; Normand, Nicolas; Mouchère, Harold; Coudol, Sandrine; Trang, Caroline; Bourreille, Arnaud

doi:10.1055/a-1468-3964

Cited by 23 publications

(10 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A variable number of training images (469–483,444) were used to develop the various AI technology followed by validation. The sensitivity and specificity of the AI models were 80 to 97.1% and 89 to 98.1%, respectively 184 185 186 187 188 189 190 191 192 ( Supplementary Table S4 , available in the online version). Hence, AI can significantly reduce the examination time with excellent sensitivity and specificity.…”

Section: Resultsmentioning

confidence: 99%

“…This score also named as the Elaikim score was shown to have an excellent correlation with LS and had excellent interobserver agreement (k ¼ 0.9). [184][185][186][187][188][189][190][191][192] (►Supplementary Table S4, available in the online version). Hence, AI can significantly reduce the examination time with excellent sensitivity and specificity.…”

Section: Panenteric Capsule Endoscopy Scoresmentioning

confidence: 99%

See 1 more Smart Citation

Capsule Endoscopy in Inflammatory Bowel Disease: A Systematic Review

Pal,

Banerjee,

Gupta

et al. 2023

Journal of Digestive Endoscopy

View full text Add to dashboard Cite

The role of video capsule endoscopy (VCE) in inflammatory bowel disease (IBD) has evolved from small bowel to a panenteric evaluation tool over the past two decades. We systematically reviewed the techniques, applications, outcomes, and complications of VCE in IBD. A systematic literature search was performed using PubMed, Embase, and Medline. All relevant original articles involving VCE in IBD were included from 2003 to July 2022. After screening 3,089 citations, finally 201 references were included. The diagnostic yield of VCE in suspected Crohn's disease (CD) was highly variable (6–80%) with excellent sensitivity (77–93%) and specificity (80–89%). The diagnostic yield in known CD was 52 to 88.3% leading to a change in management (26–75%) and disease reclassification with variable retention rates. VCE was superior to small bowel series, computed tomography (CT) and could be better than magnetic resonance enterography (MRE), especially for proximal and superficial lesions. Colon or panenteric VCE has strong correlation to ileo-colonoscopy (IC) and combined magnetic resonance imaging and IC, respectively. The VCE retention rate in CD is higher in known CD which significantly decreases after the negative patency capsule test or CT/MRE. VCE can identify lesions beyond the reach of IC in postoperative CD. Colon Capsule Endoscopy is a noninvasive monitoring tool in ulcerative colitis (UC) having a strong correlation with IC and may uncover small bowel involvement. VCE is specifically useful in IBD-unclassified (IBD-U) which can lead to the diagnosis of CD in 16.7 to 61.5%. Various scoring systems have been established and validated for small bowel CD (Lewis score and capsule endoscopy CD activity index—CECDAI), UC (capsule scoring of UC: Capsule Scoring of Ulcerative Colitis), panenteric evaluation (Capsule Endoscopy Crohn's Disease Activity Index, Elaikim score), and flare prediction (APEX score). Technological advances include double head, three-dimensional reconstruction, sampling system, panoramic view (344 and 360 degree lateral), and panenteric capsule. Artificial intelligence and software like TOP100 and Quickview can help reduce capsule reading time with excellent sensitivity and specificity. VCE in IBD has widespread application in suspected and known small bowel CD, monitoring of UC, postoperative CD, IBD-U, and for panenteric evaluation. Patency capsule testing helps to reduce retention rates significantly. Artificial intelligence and technical advances can help evolve this novel technology.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Panenteric Capsule Endoscopy Scoresmentioning

confidence: 99%

Capsule Endoscopy in Inflammatory Bowel Disease: A Systematic Review

Pal,

Banerjee,

Gupta

et al. 2023

Journal of Digestive Endoscopy

View full text Add to dashboard Cite

show abstract

“… 33 – 35 CE is perhaps an even more attractive target for AI research, as there is no need for real-time diagnosis, and the variety of pathology types is somewhat limited: thus, identification of most SB pathologies by AI was very accurate, with AUC above 90%. 22 , 30 , 36 – 38 In future models of CE, strong incorporation of AI modules with automated lesion markup can be expected. However, the incorporation of AI into clinical practice and clinical trials requires a huge leap in terms of standardization, quality assessment, reproducibility, and workflow integration.…”

Section: Discussionmentioning

confidence: 99%

Key research questions for implementation of artificial intelligence in capsule endoscopy

Leenhardt

Koulaouzidis

Histace

et al. 2022

Therap Adv Gastroenterol

Self Cite

View full text Add to dashboard Cite

Background: Artificial intelligence (AI) is rapidly infiltrating multiple areas in medicine, with gastrointestinal endoscopy paving the way in both research and clinical applications. Multiple challenges associated with the incorporation of AI in endoscopy are being addressed in recent consensus documents. Objectives: In the current paper, we aimed to map future challenges and areas of research for the incorporation of AI in capsule endoscopy (CE) practice. Design: Modified three-round Delphi consensus online survey. Methods: The study design was based on a modified three-round Delphi consensus online survey distributed to a group of CE and AI experts. Round one aimed to map out key research statements and challenges for the implementation of AI in CE. All queries addressing the same questions were merged into a single issue. The second round aimed to rank all generated questions during round one and to identify the top-ranked statements with the highest total score. Finally, the third round aimed to redistribute and rescore the top-ranked statements. Results: Twenty-one (16 gastroenterologists and 5 data scientists) experts participated in the survey. In the first round, 48 statements divided into seven themes were generated. After scoring all statements and rescoring the top 12, the question of AI use for identification and grading of small bowel pathologies was scored the highest (mean score 9.15), correlation of AI and human expert reading-second (9.05), and real-life feasibility-third (9.0). Conclusion: In summary, our current study points out a roadmap for future challenges and research areas on our way to fully incorporating AI in CE reading.

show abstract

“…A consensus annotation for each image was provided, obtained from the independent review of several experts. The multi-reader annotation process was described in [31]. This article highlighted the importance of generating a consensus diagnosis.…”

Section: State Of the Artmentioning

confidence: 99%

Automatic detection of crohn disease in wireless capsule endoscopic images using a deep convolutional neural network

et al. 2022

View full text Add to dashboard Cite

The diagnosis of Crohn’s disease (CD) in the small bowel is generally performed by observing a very large number of images captured by capsule endoscopy (CE). This diagnostic technique entails a heavy workload for the specialists in terms of time spent reviewing the images. This paper presents a convolutional neural network capable of classifying the CE images to identify those ones affected by lesions indicative of the disease. The architecture of the proposed network was custom designed to solve this image classification problem. This allowed different design decisions to be made with the aim of improving its performance in terms of accuracy and processing speed compared to other state-of-the-art deep-learning-based reference architectures. The experimentation was carried out on a set of 15,972 images extracted from 31 CE videos of patients affected by CD, 7,986 of which showed lesions associated with the disease. The training, validation/selection and evaluation of the network was performed on 70%, 10% and 20% of the total images, respectively. The ROC curve obtained on the test image set has an area greater than 0.997, with points in a 95-99% sensitivity range associated with specificities of 99-96%. These figures are higher than those achieved by EfficientNet-B5, VGG-16, Xception or ResNet networks which also require an average processing time per image significantly higher than the one needed in the proposed architecture. Therefore, the network outlined in this paper is proving to be sufficiently promising to be considered for integration into tools used by specialists in their diagnosis of CD. In the sample of images analysed, the network was able to detect 99% of the images with lesions, filtering out for specialist review 96% of those with no signs of disease.

show abstract

Multi-expert annotation of Crohn’s disease images of the small bowel for automatic detection using a convolutional recurrent attention neural network

Cited by 23 publications

References 26 publications

Capsule Endoscopy in Inflammatory Bowel Disease: A Systematic Review

Capsule Endoscopy in Inflammatory Bowel Disease: A Systematic Review

Key research questions for implementation of artificial intelligence in capsule endoscopy

Automatic detection of crohn disease in wireless capsule endoscopic images using a deep convolutional neural network

Contact Info

Product

Resources

About