An automated three-dimensional visualization and classification of emphysema using neural network

Tan Kok Liang, Toshiyuki Tanaka, Hidetoshi Nakamura, Toru Shirahata, Hiroaki Sugiura

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Chronic Obstructive Pulmonary Disease (COPD) is a disease in which the airways and tiny air sacs (alveoli) inside the lungs are partially obstructed or destroyed. Emphysema is what occurs as more and more of the walls between air sacs get destroyed. Computed tomography (CT) image has been a useful modality for assessing diffuse lung diseases, particularly, emphysema. At present, diagnosis of emphysema is done by using spirometry, X-rays, spiral chest computed tomography (CT)-scan, bronchoscopy, blood tests and pulse oximetry. In this study, we extracted the two-dimensional emphysematous lung tissues in the lung CT automatically using digital image processing techniques, then we visualized the extracted emphysematous lung tissues by implementing a three-dimensional (3D) lung model which was computed using 55 pre-processed CT images, and finally we divided the lung model into eight sub-volumes and classified each sub-volume into five classes of emphysema related severity using an artificial neural network. The performance of the classifier was assessed using the leave-one-out method on 120 sub-volumes of the lungs generated from 15 COPD-verified patients' CT data sets.

Original languageEnglish
Title of host publication2008 42nd Asilomar Conference on Signals, Systems and Computers, ASILOMAR 2008
Pages1936-1940
Number of pages5
DOIs
Publication statusPublished - 2008 Dec 1
Event2008 42nd Asilomar Conference on Signals, Systems and Computers, ASILOMAR 2008 - Pacific Grove, CA, United States
Duration: 2008 Oct 262008 Oct 29

Publication series

NameConference Record - Asilomar Conference on Signals, Systems and Computers
ISSN (Print)1058-6393

Other

Other2008 42nd Asilomar Conference on Signals, Systems and Computers, ASILOMAR 2008
CountryUnited States
CityPacific Grove, CA
Period08/10/2608/10/29

ASJC Scopus subject areas

  • Signal Processing
  • Computer Networks and Communications

Fingerprint Dive into the research topics of 'An automated three-dimensional visualization and classification of emphysema using neural network'. Together they form a unique fingerprint.

Cite this