Least Square Saliency Transformation of Capsule Endoscopy Images for PDF Model Based Multiple Gastrointestinal Disease Classification

Amit Kumar Kundu,Khan A Wahid,Shaikh Anowarul Fattah

doi:10.1109/access.2020.2982870

Abstract

Wireless capsule endoscopy (WCE) is an effective video technology to diagnose gastrointestinal (GI) diseases, such as bleeding, ulcer, and tumor. In order to avoid a tedious manual review process of long duration WCE video recordings, automatic disease detection schemes have received significant attention from the researchers. In particular, instead of the conventional approach of dealing with a single disease, developing a unified scheme, which is capable of detecting multiple GI diseases, is getting more importance while being challenging. In this paper, a unified computer-aided scheme is developed for detecting multiple GI diseases from WCE videos based on a proposed least-square saliency transformation (LSST) followed by a probabilistic model-fitting approach. Commonly in the training phase, image-level labeling of images is used, as pixel-level annotations are available only for a small number of images. In view of utilizing the knowledge from pixel-level annotated diseased images, an LSST scheme is proposed to extract a set of optimum prior coefficient-vectors which is later used to capture the salient pixels of interest (POI) in a larger WCE image dataset that do not have pixel-annotations. The intensity distributions of salient POI are modeled by a suitable probability density function (PDF) and the fitted PDF parameters are utilized as features in the proposed supervised hierarchical classification scheme. A large number of WCE images obtained from publicly available WCE videos are used for performance evaluation and it is found that the result obtained by the proposed method outperforms the results obtained in case of some state-of-the-art methods.

Highlights

Detecting multiple gastrointestinal (GI) diseases, like bleeding, ulcer and tumor via manual inspection of wireless capsule endoscopy (WCE) videos require a significant amount of time and effort [1]
The objective of this paper is to develop a unified automatic scheme for detecting multiple GI diseases from WCE videos based on proposed least square saliency transformation (LSST) and probabilistic model fitting scheme using a minimum number of pixel-level annotated images of different diseases
In this paper, a unified two-stage computer aided multiple GI disease detection scheme is proposed where, first, the LSST based salient pixels of interest (POI) of a given image are extracted and a suitable probability density function (PDF) model is fitted on the extracted POI to utilize the model parameters in the supervised classification scheme

Summary

Introduction

Detecting multiple gastrointestinal (GI) diseases, like bleeding, ulcer and tumor via manual inspection of wireless capsule endoscopy (WCE) videos require a significant amount of time and effort [1]. Automatic disease detection schemes from WCE videos have received great attention from the researchers [2]. Most of the research efforts given so far deal only with one type of disease detection from WCE videos. The reported automatic bleeding detection schemes are mainly based on suspected blood indicator [3], histogram-based features [4]–[7], statistical features [8], block-based approaches [7], [9], features from salient points [11], [15] and deep learning framework [12]. Automatic ulcer detection schemes are proposed based on convolutional neural network (CNN) based architecture [16], completed local binary patterns (LBP), and laplacian

Objectives

Results

Conclusion