English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Physicians use Capsule Endoscopy (CE) as a non-invasive and non-surgicalprocedure to examine the entire gastrointestinal (GI) tract for diseases andabnormalities. A single CE examination could last between 8 to 11 hoursgenerating up to 80,000 frames which is compiled as a video. Physicians have toreview and analyze the entire video to identify abnormalities or diseasesbefore making diagnosis. This review task can be very tedious, time consumingand prone to error. While only as little as a single frame may capture usefulcontent that is relevant to the physicians' final diagnosis, frames coveringthe small bowel region alone could be as much as 50,000. To minimizephysicians' review time and effort, this paper proposes a novel unsupervisedand computationally efficient temporal segmentation method to automaticallypartition long CE videos into a homogeneous and identifiable video segments.However, the search for temporal boundaries in a long video using highdimensional frame-feature matrix is computationally prohibitive andimpracticable for real clinical application. Therefore, leveraging both spatialand temporal information in the video, we first extracted high level framefeatures using a pretrained CNN model and then projected the high-dimensionalframe-feature matrix to lower 1-dimensional embedding. Using this 1-dimensionalsequence embedding, we applied the Pruned Exact Linear Time (PELT) algorithm tosearched for temporal boundaries that indicates the transition points fromnormal to abnormal frames and vice-versa. We experimented with multiple realpatients' CE videos and our model achieved an AUC of 66\% on multiple testvideos against expert provided labels.

Unsupervised Shot Boundary Detection for Temporal Segmentation of Long Capsule Endoscopy Videos