A statistical feature selection method for lung cancer classification in CT scans

Hamada R.H. Al-Absi, Brahim Belhaouari Samir

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This paper presents a computer aided diagnosis for lung nodules in CT images. The system consists of feature extraction, feature selection and classification. A two-step feature selection process is introduced to reduce the number of coefficients produced in the feature extraction step. This helps in enhancing the classification performance as it removes unneeded and redundant information. The classification rate of the system reached 98.10 % with minimum false negatives and zero false positives.

Original languageEnglish
Title of host publication11th International Conference of Numerical Analysis and Applied Mathematics 2013, ICNAAM 2013
Pages2524-2527
Number of pages4
DOIs
Publication statusPublished - 2013
Externally publishedYes
Event11th International Conference of Numerical Analysis and Applied Mathematics 2013, ICNAAM 2013 - Rhodes, Greece
Duration: 21 Sept 201327 Sept 2013

Publication series

NameAIP Conference Proceedings
Volume1558
ISSN (Print)0094-243X
ISSN (Electronic)1551-7616

Conference

Conference11th International Conference of Numerical Analysis and Applied Mathematics 2013, ICNAAM 2013
Country/TerritoryGreece
CityRhodes
Period21/09/1327/09/13

Keywords

  • Lung nodules
  • cluster k-Nearest Neighbor
  • computer aided diagnosis
  • feature selection

Fingerprint

Dive into the research topics of 'A statistical feature selection method for lung cancer classification in CT scans'. Together they form a unique fingerprint.

Cite this