Classification of FTIR cancer data using wavelets and BPNN

Cungui Cheng; Yumei Tian; Changjiang Zhang

doi:10.1117/12.758497

8 January 2008 Classification of FTIR cancer data using wavelets and BPNN

Cungui Cheng, Yumei Tian, Changjiang Zhang

Proceedings Volume 6826, Optics in Health Care and Biomedical Optics III; 682633 (2008) https://doi.org/10.1117/12.758497
Event: Photonics Asia 2007, 2007, Beijing, China

Abstract

In this paper, a feature extracting method based on wavelets for horizontal attenuated total reflectance Fourier transform infrared spectroscopy (HATR-FTIR) cancer data analysis and classification using artificial neural network trained with back-propagation algorithm is presented. 168 Spectra were collected from 84 pairs of fresh normal and abnormal lung tissue's samples. After preprocessing, 12 features were extracted with continuous wavelet analysis. Based on BPNN classification, all spectra were classified into two categories : normal or abnormal. The accuracy of identifying normal, early carcinoma, and advanced carcinoma were 100%, 90% and 100% respectively. This result indicated that FTIR with continuous wavelet transform (CWT) and the back-propagation neural network (BPNN) could effectively and easily diagnose lung cancer in its early stages.

Citation Download Citation

Cungui Cheng, Yumei Tian, and Changjiang Zhang "Classification of FTIR cancer data using wavelets and BPNN", Proc. SPIE 6826, Optics in Health Care and Biomedical Optics III, 682633 (8 January 2008); https://doi.org/10.1117/12.758497

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available