Computer-aided estimation for the risk of development of gastric cancer by image processing

Abstract : The aim of this study was to establish a computer-aided estimating system for determining the risk of development of gastric cancer, achieved by image processing on an ordinary endoscopic picture. Digital endoscopic pictures of the background gastric mucosa in 26 Helicobacter pylori (H. pylori) positive patients with early intestinal type gastric cancer and age-gender-matched H. pylori positive subjects without cancer were used. The pictures were processed for 15 pictorial parameters. Out of the 15 pictorial parameters, 3 parameters were found to characterize the background gastric mucosa with gastric cancer against that without. Based on the Bayes decision theory, the computer-aided estimating system has been established. Sensitivity, specificity, positive predictive value and negative predictive value of the Bayes classifier were found to be 0.64, 0.64, 0.65 and 0.63, respectively. This method may permit an effective selection of the high risk population of gastric cancer needing follow-up endoscopy.
Document type :
Conference papers
Complete list of metadatas

Cited literature [11 references]  Display  Hide  Download

https://hal.inria.fr/hal-01054593
Contributor : Hal Ifip <>
Submitted on : Thursday, August 7, 2014 - 3:18:17 PM
Last modification on : Thursday, November 23, 2017 - 2:56:13 PM
Long-term archiving on : Wednesday, November 26, 2014 - 1:47:56 AM

File

The_second_submission_YSasaki....
Files produced by the author(s)

Licence


Distributed under a Creative Commons Attribution 4.0 International License

Identifiers

Citation

Yoshihiro Sasaki, Ryukichi Hada, Tetsuro Yoshimura, Norihiro Hanabata, Tatsuya Mikami, et al.. Computer-aided estimation for the risk of development of gastric cancer by image processing. Third IFIP TC12 International Conference on Artificial Intelligence (AI) / Held as Part of World Computer Congress (WCC), Sep 2010, Brisbane, Australia. pp.197-204, ⟨10.1007/978-3-642-15286-3_19⟩. ⟨hal-01054593⟩

Share

Metrics

Record views

265

Files downloads

162