Fusion of Selected Deep CNN and Handcrafted Features for Gastritis Detection from Wireless Capsule Endoscopy Images

Zhao, Bailiang; Sun, Wendell Q.; Wang, Liangchao; Hu, Menghan

doi:10.1109/cisp-bmei53629.2021.9624380

Cited by 1 publication

(2 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The approaches using these set of features with the classification of support vector machine (SVM) ( Hearst et al, 1998 ), random forest, and the k-nearest neighbors (KNN) classifier resulted in a good detection scores ( Tajbakhsh et al, 2014 ; Pogorelov et al, 2017a ; Iakovidis, Maroulis & Karkanis, 2006 ; Alexandre, Nobre & Casteleiro, 2008 ; Karkanis et al, 2001 ; Zhao et al, 2021 ; Bernal, Sánchez & Vilarino, 2013 ; Esgiar et al, 1998 ; Deeba et al, 2018 ; Hwang et al, 2007 ). The F1-scores of the approaches of Pogorelov et al (2017a) , Zhao et al (2021) , Hwang et al (2007) and Deeba et al (2018) are above 0.90 for the detection of various abnormalities in the GI tract. Some approaches focus on the precision, recall, or the Matthews correlation coefficient (MCC) score, resulting in some of these evaluation parameters ( Tajbakhsh et al, 2014 ; Naqvi et al, 2017 ).…”

Section: Related Workmentioning

confidence: 99%

“…The approach of FAST-NUDS based on only texture features resulted in the F1-score of 0.75 with a real-time detection ( Khan & Tahir, 2018 ). The best evaluations on detecting the various abnormalities in the GI tract are achieved by Zhao et al (2021) using a fusion of the features extracted from the deep neural network with the handcrafted features.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Real time anatomical landmarks and abnormalities detection in gastrointestinal tract

Khan,

Tahir

2023

PeerJ Computer Science

View full text Add to dashboard Cite

Gastrointestinal (GI) endoscopy is an active research field due to the lethal cancer diseases in the GI tract. Cancer treatments result better if diagnosed early and it increases the survival chances. There is a high miss rate in the detection of the abnormalities in the GI tract during endoscopy or colonoscopy due to the lack of attentiveness, tiring procedures, or the lack of required training. The procedure of the detection can be automated to the reduction of the risks by identifying and flagging the suspicious frames. A suspicious frame may have some of the abnormality or the information about anatomical landmark in the frame. The frame then can be analysed for the anatomical landmarks and the abnormalities for the detection of disease. In this research, a real-time endoscopic abnormalities detection system is presented that detects the abnormalities and the landmarks. The proposed system is based on a combination of handcrafted and deep features. Deep features are extracted from lightweight MobileNet convolutional neural network (CNN) architecture. There are some of the classes with a small inter-class difference and a higher intra-class differences, for such classes the same detection threshold is unable to distinguish. The threshold of such classes is learned from the training data using genetic algorithm. The system is evaluated on various benchmark datasets and resulted in an accuracy of 0.99 with the F1-score of 0.91 and Matthews correlation coefficient (MCC) of 0.91 on Kvasir datasets and F1-score of 0.93 on the dataset of DowPK. The system detects abnormalities in real-time with the detection speed of 41 frames per second.

show abstract