Improving CAD Hemorrhage Detection in Capsule Endoscopy  

作　　者：Polydorou Alexios Sergaki Eleftheria Polydorou Andreas Barbagiannis Christos Vardiambasis Ioannis George Giakos Zervakis Michail Polydorou Alexios;Sergaki Eleftheria;Polydorou Andreas;Barbagiannis Christos;Vardiambasis Ioannis;George Giakos;Zervakis Michail(School of Electrical and Computer Engineering, Technical University of Crete, Chania, Greece;Department of Surgery, School of Medicine, National and Kapodistrian University of Athens, Athens, Greece;Department of Electronic Engineering, Faculty of Engineering, Hellenic Mediterranean University, Chania Crete, Greece;School of ECE, Manhattan College, New York, USA)

机构地区：[1]School of Electrical and Computer Engineering, Technical University of Crete, Chania, Greece [2]Department of Surgery, School of Medicine, National and Kapodistrian University of Athens, Athens, Greece [3]Department of Electronic Engineering, Faculty of Engineering, Hellenic Mediterranean University, Chania Crete, Greece [4]School of ECE, Manhattan College, New York, USA

出　　处：《Journal of Biomedical Science and Engineering》2021年第3期103-118,共16页生物医学工程（英文）

摘　　要：This study explores an automated framework to assist the recognition of hemorrhage traces and bleeding lesions in video streams of small bowel capsule endoscopy (SBCE). The proposed methodology aims to achieve fast image control (<10 minutes), save valuable time of the physicians, and enable high performance diagnosis. A specialized elimination algorithm excludes all identical consecutive frames by utilizing the difference of gray levels in pixel luminance. An image filtering algorithm is proposed based on an experimentally calculated bleeding index and blood-color chart, which inspects all remaining frames of the footage and identifies pixels that reflect active or potential hemorrhage in color. The bleeding index and blood-color chart are estimated of the chromatic thresholds in RGB and HSV color spaces, and have been extracted after experimenting with more than 3200 training images, derived from 99 videos of a pool of 138 patients. The dataset has been provided by a team of expert gastroenterologist surgeons, who have also evaluated the results. The proposed algorithms are tested on a set of more than 1000 selected frame samples from the entire 39 testing videos, to a prevalence of 50% pathologic frames (balanced dataset). The frame elimination of identical and consecutive frames achieved a reduction of 36% of total frames. The best statistical performance for diagnosis of positive pathological frames from a video stream is achieved by utilizing masks in the HSV color model, with sensitivity up to 99%, precision 94.41% to a prevalence of 50%, accuracy up to 96.1%, FNR 1%, FPR 6.8%. The estimated blood-color chart will be clinically validated and used in feature extraction schemes supporting machine learning ML algorithms to improve the localization potential.This study explores an automated framework to assist the recognition of hemorrhage traces and bleeding lesions in video streams of small bowel capsule endoscopy (SBCE). The proposed methodology aims to achieve fast image control (<10 minutes), save valuable time of the physicians, and enable high performance diagnosis. A specialized elimination algorithm excludes all identical consecutive frames by utilizing the difference of gray levels in pixel luminance. An image filtering algorithm is proposed based on an experimentally calculated bleeding index and blood-color chart, which inspects all remaining frames of the footage and identifies pixels that reflect active or potential hemorrhage in color. The bleeding index and blood-color chart are estimated of the chromatic thresholds in RGB and HSV color spaces, and have been extracted after experimenting with more than 3200 training images, derived from 99 videos of a pool of 138 patients. The dataset has been provided by a team of expert gastroenterologist surgeons, who have also evaluated the results. The proposed algorithms are tested on a set of more than 1000 selected frame samples from the entire 39 testing videos, to a prevalence of 50% pathologic frames (balanced dataset). The frame elimination of identical and consecutive frames achieved a reduction of 36% of total frames. The best statistical performance for diagnosis of positive pathological frames from a video stream is achieved by utilizing masks in the HSV color model, with sensitivity up to 99%, precision 94.41% to a prevalence of 50%, accuracy up to 96.1%, FNR 1%, FPR 6.8%. The estimated blood-color chart will be clinically validated and used in feature extraction schemes supporting machine learning ML algorithms to improve the localization potential.

关 键 词：Capsule Endoscopy Small Bowel Bleeding Detection Computer Aided Diagnosis (CAD) Color Models Color Feature 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]