Spaciotmeporal Machine Learning Analysis of Complete Small Bowel Endoscopy Videos for Prediction of Outcomes in Crohn's Disease

Kellerman, Raizy; Bleiweiss, Amit; Samuel, Shimrit; Yehuda, Reuma Margalit-; Barzilay, Oranit; Ben‐Horin, Shomron; Eliakim, Rami; Klang, Eyal; Kopylov, Uri

doi:10.1016/j.gie.2022.04.1179

Cited by 1 publication

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The diagnostic yield of the SBCE examination could be improved utilizing artificial intelligent systems 46,47 . The presence of water bubbles, chime, foam, debris, or food residues in the small bowel may reduce mucosal visibility, increase the risks of missing lesions, and lead to misdiagnosis.…”

Section: Discussionmentioning

confidence: 99%

“…The diagnostic yield of the SBCE examination could be improved utilizing artificial intelligent systems. 46,47 The presence of water bubbles, chime, foam, debris, or food residues in the small bowel may reduce mucosal visibility, increase the risks of missing lesions, and lead to misdiagnosis. To decrease the false positive rate (FPR) in abnormality detection, it is beneficial to localize and reject contaminated regions of a frame, which do not convey clinical information.…”

Section: Number Of Test Imagesmentioning

confidence: 99%

See 1 more Smart Citation

Real‐time small bowel visualization quality assessment in wireless capsule endoscopy images using different lightweight embeddable models

Sadeghi,

Mehridehnavi,

Sanahmadi

et al. 2024

Int J Imaging Syst Tech

View full text Add to dashboard Cite

Wireless capsule endoscopy (WCE) captures huge number of images, but only a fraction are medically relevant. We propose automated real‐time small bowel visualization quality (SBVQ) assessment to eliminate transmission of irrelevant frames. Our aim is to design lightweight color‐based models for segmenting clean and contaminated regions with minimal parameters, short training, and fast inference, suitable for WCE hardware integration. Using the Kvasir Capsule endoscopy dataset, we constructed models based on distinctive color patterns of clean and contaminated regions. While different classifiers have been trained and evaluated, the k‐nearest neighbors (KNNs), multilayer perceptron (MLP), and gradient‐boosted machine (GBM) obtained superior performance (accuracy: 0.87±0.12, Dice similarity score (DSC): 0.87±0.15, intersection over union (IOU): 0.80±0.19). Logistic regression (LR) had the shortest training and inference times. Our models offer simplicity, compactness, and robustness, delivering satisfactory real‐time performance. Evaluation on the SEE‐AI project dataset confirms good generalization capabilities, demonstrating practical solutions for WCE image analysis.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Number Of Test Imagesmentioning

confidence: 99%