Lesion2Vec: Deep Meta Learning for Few-Shot Lesion Recognition in Capsule Endoscopy Video

Adewole, Sodiq; Fernandes, Philip; Jablonski, James; Copland, Andrew; Porter, Michael D.; Syed, Sana; Brown, Donald E.

doi:10.1007/978-3-030-89880-9_57

Cited by 6 publications

(1 citation statement)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similarly, Kim et al have used the Inception-Resnet-V2 model for the binary classi cation of frames containing a clinically signi cant lesion, such as in amed mucosa, abnormal vascularity, or bleeding [18]. Table S1 in the supplementary material provides a comprehensive review of the progress made over the years in classifying WCE videos using deep neural networks [19][20][21][22][23][24][25][26].…”

Section: Introductionmentioning

confidence: 99%

Wireless Capsule Endoscopy Multiclass Classification Using 3D Deep CNN Model

Bordbar,

Helfroush,

Danyali

et al. 2023

Preprint

View full text Add to dashboard Cite

Wireless capsule endoscopy (WCE) is a patient-friendly and non-invasive technology that scans the whole of the gastrointestinal tract, including difficult-to-access regions like the small bowel. The major drawback of this technology is that the visual inspection of a large number of video frames produced during each examination makes the physician diagnosis process tedious and prone to error. Several computer-aided diagnosis (CAD) systems, such as deep network models, have been developed for the automatic recognition of abnormalities in WCE frames. Nevertheless, most of these studies have only focused on spatial information within individual WCE frames, missing the crucial temporal data within consecutive frames. In this article an automatic multiclass classification system based on a 3D deep convolutional neural network (3D-CNN) is proposed, which utilizes the spatiotemporal information to facilitate the WCE diagnosis process. 3D-CNN model is fed with a series of sequential WCE frames in contrast to the 2D model, which exploits frames as independent ones. Moreover, the proposed 3D deep model is compared with some pre-trained networks. The proposed models are trained and evaluated with 29 subject WCE videos (14691 frames before augmentation). The performance advantages of 3D-CNN over 2D-CNN and pre-trained networks are verified in terms of sensitivity, specificity, and accuracy. 3D-CNN outperforms the 2D technique in all evaluation metrics (Sensitivity: 98.92 vs. 98.05, Specificity: 99.50 vs. 86.94, Accuracy: 99.20 vs. 92.60). In conclusion, a novel 3D-CNN model for lesion detection in WCE frames is proposed in this study. The results indicate the performance of 3D-CNN over 2D-CNN and some well-known pre-trained classifier networks. The proposed 3D-CNN model uses the rich temporal information in adjacent frames as well as spatial data to develop an accurate and efficient model.

show abstract