Temporal Context Framework for Endoscopy Artefact Segmentation and Detection

Haili Ye; Hanpei Miao; Jiang Liu; Dahan Wang; Heng Li

Temporal Context Framework for Endoscopy Artefact Segmentation and Detection

Haili Ye, Hanpei Miao, Jiang Liu, Dahan Wang, Heng Li

Research output: Journal Publication › Conference article › peer-review

Abstract

Endoscopic video processing could facilitate pre-operative planning, intra-operative image guidance and generation of post-operative analysis of the surgical procedure. However, most of the current methods are still based on a single frame of image analysis, which makes the results of the previous frame images independent of each other and causes vibration. In this paper, we propose an temporal context framework for endoscopy artefact segmentation and detection. The framework extends the general segmentation and detection model to the form based on temporal input, and we add a Temporal Context Transformer(TCT) after the encoder of the model to improve the model’s ability to construct temporal context features. the experiments of the EndoCV 2022 challenge dataset that this framework can improve the robustness of the model.

Original language	English
Pages (from-to)	35-39
Number of pages	5
Journal	CEUR Workshop Proceedings
Volume	3148
Publication status	Published - 2022
Externally published	Yes
Event	4th International Workshop and Challenge on Computer Vision in Endoscopy, EndoCV 2022 - Kolkata, India Duration: 28 Mar 2022 → …

Keywords

Colonoscopic Image
Medical Image Analysis
Object Detection
Semantic Segmentation

ASJC Scopus subject areas

General Computer Science

Cite this

@article{c6fdb8b1c5d7494fbfaff3b8db5b5de7,

title = "Temporal Context Framework for Endoscopy Artefact Segmentation and Detection",

abstract = "Endoscopic video processing could facilitate pre-operative planning, intra-operative image guidance and generation of post-operative analysis of the surgical procedure. However, most of the current methods are still based on a single frame of image analysis, which makes the results of the previous frame images independent of each other and causes vibration. In this paper, we propose an temporal context framework for endoscopy artefact segmentation and detection. The framework extends the general segmentation and detection model to the form based on temporal input, and we add a Temporal Context Transformer(TCT) after the encoder of the model to improve the model{\textquoteright}s ability to construct temporal context features. the experiments of the EndoCV 2022 challenge dataset that this framework can improve the robustness of the model.",

keywords = "Colonoscopic Image, Medical Image Analysis, Object Detection, Semantic Segmentation",

author = "Haili Ye and Hanpei Miao and Jiang Liu and Dahan Wang and Heng Li",

note = "Publisher Copyright: {\textcopyright} 2022 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).; 4th International Workshop and Challenge on Computer Vision in Endoscopy, EndoCV 2022 ; Conference date: 28-03-2022",

year = "2022",

language = "English",

volume = "3148",

pages = "35--39",

journal = "CEUR Workshop Proceedings",

issn = "1613-0073",

publisher = "CEUR-WS",

}

TY - JOUR

T1 - Temporal Context Framework for Endoscopy Artefact Segmentation and Detection

AU - Ye, Haili

AU - Miao, Hanpei

AU - Liu, Jiang

AU - Wang, Dahan

AU - Li, Heng

PY - 2022

Y1 - 2022

N2 - Endoscopic video processing could facilitate pre-operative planning, intra-operative image guidance and generation of post-operative analysis of the surgical procedure. However, most of the current methods are still based on a single frame of image analysis, which makes the results of the previous frame images independent of each other and causes vibration. In this paper, we propose an temporal context framework for endoscopy artefact segmentation and detection. The framework extends the general segmentation and detection model to the form based on temporal input, and we add a Temporal Context Transformer(TCT) after the encoder of the model to improve the model’s ability to construct temporal context features. the experiments of the EndoCV 2022 challenge dataset that this framework can improve the robustness of the model.

AB - Endoscopic video processing could facilitate pre-operative planning, intra-operative image guidance and generation of post-operative analysis of the surgical procedure. However, most of the current methods are still based on a single frame of image analysis, which makes the results of the previous frame images independent of each other and causes vibration. In this paper, we propose an temporal context framework for endoscopy artefact segmentation and detection. The framework extends the general segmentation and detection model to the form based on temporal input, and we add a Temporal Context Transformer(TCT) after the encoder of the model to improve the model’s ability to construct temporal context features. the experiments of the EndoCV 2022 challenge dataset that this framework can improve the robustness of the model.

KW - Colonoscopic Image

KW - Medical Image Analysis

KW - Object Detection

KW - Semantic Segmentation

UR - http://www.scopus.com/inward/record.url?scp=85132139430&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:85132139430

SN - 1613-0073

VL - 3148

SP - 35

EP - 39

JO - CEUR Workshop Proceedings

JF - CEUR Workshop Proceedings

T2 - 4th International Workshop and Challenge on Computer Vision in Endoscopy, EndoCV 2022

Y2 - 28 March 2022

ER -

Temporal Context Framework for Endoscopy Artefact Segmentation and Detection

Abstract

Keywords

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this