RUS  ENG
Full version
JOURNALS // Computer Optics // Archive

Computer Optics, 2017 Volume 41, Issue 5, Pages 726–731 (Mi co443)

This article is cited in 12 papers

IMAGE PROCESSING, PATTERN RECOGNITION

Pulmonary emphysema recognition by CT scan

N. A. Smelkinaa, A. V. Kolsanovb, S. S. Chaplyginb, P. M. Zelterb, A. G. Khramova

a Samara National Research University, Samara, Russia
b Samara State Medical University, Samara, Russia

Abstract: We discuss a simple method for automatic recognition of pulmonary emphysema in three-dimensional computer tomography (CT) images. This technique allows one to quantify the disease progress, calculating some numerical characteristics, such as the percentage of the lung tissue affected, as well as visualizing its location and intensity histogram in the region of interest. An experiment on the test data shows that the recognition error is not higher than 7.5%.

Keywords: CT scan, pulmonary emphysema, diagnostic images, pathology segmentation, data mining.

Received: 28.01.2017
Accepted: 25.08.2017

DOI: 10.18287/2412-6179-2017-41-5-726-731



© Steklov Math. Inst. of RAS, 2026