eprintid: 14585
rev_number: 2
eprint_status: archive
userid: 1
dir: disk0/00/01/45/85
datestamp: 2023-11-10 03:29:10
lastmod: 2023-11-10 03:29:10
status_changed: 2023-11-10 01:57:17
type: article
metadata_visibility: show
creators_name: Eu, C.Y.
creators_name: Tang, T.B.
creators_name: Lin, C.-H.
creators_name: Lee, L.H.
creators_name: Lu, C.-K.
title: Automatic polyp segmentation in colonoscopy images using a modified deep convolutional encoder-decoder architecture
ispublished: pub
keywords: Computer aided diagnosis; Convolution; Diseases; Endoscopy; Medical robotics; Signal encoding, Automatic segmentations; Colorectal cancer; Colorectal cancer diagnosis; Computer Aided Diagnosis(CAD); Convolutional encoders; Hospital settings; Optical colonoscopy; Polyp segmentation, Image segmentation, colonoscopy; computer assisted diagnosis; factual database; image processing; methodology, Colonoscopy; Databases, Factual; Diagnosis, Computer-Assisted; Image Processing, Computer-Assisted; Neural Networks, Computer; Research Design
note: cited By 3
abstract: Colorectal cancer has become the third most commonly diagnosed form of cancer, and has the second highest fatality rate of cancers worldwide. Currently, optical colonoscopy is the pre-ferred tool of choice for the diagnosis of polyps and to avert colorectal cancer. Colon screening is time-consuming and highly operator dependent. In view of this, a computer-aided diagnosis (CAD) method needs to be developed for the automatic segmentation of polyps in colonoscopy images. This paper proposes a modified SegNet Visual Geometry Group-19 (VGG-19), a form of convolu-tional neural network, as a CAD method for polyp segmentation. The modifications include skip connections, 5 � 5 convolutional filters, and the concatenation of four dilated convolutions applied in parallel form. The CVC-ClinicDB, CVC-ColonDB, and ETIS-LaribPolypDB databases were used to evaluate the model, and it was found that our proposed polyp segmentation model achieved an accuracy, sensitivity, specificity, precision, mean intersection over union, and dice coefficient of 96.06, 94.55, 97.56, 97.48, 92.3, and 95.99, respectively. These results indicate that our model performs as well as or better than previous schemes in the literature. We believe that this study will offer benefits in terms of the future development of CAD tools for polyp segmentation for colorectal cancer diagnosis and management. In the future, we intend to embed our proposed network into a medical capsule robot for practical usage and try it in a hospital setting with clini-cians. © 2021 by the authors. Licensee MDPI, Basel, Switzerland.
date: 2021
publisher: MDPI AG
official_url: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85113262692&doi=10.3390%2fs21165630&partnerID=40&md5=381137f2a3555d55bd31c054d3653e9f
id_number: 10.3390/s21165630
full_text_status: none
publication: Sensors
volume: 21
number: 16
refereed: TRUE
issn: 14248220
citation:   Eu, C.Y. and Tang, T.B. and Lin, C.-H. and Lee, L.H. and Lu, C.-K.  (2021) Automatic polyp segmentation in colonoscopy images using a modified deep convolutional encoder-decoder architecture.  Sensors, 21 (16).   ISSN 14248220