Open AccessClassification for breast cancer diagnosis with Raman spectroscopy
Author(s) -
Qingbo Li,
Qishuo Gao,
Guangjun Zhang
Publication year - 2014
Publication title -
biomedical optics express
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.362
H-Index - 86
ISSN - 2156-7085
DOI - 10.1364/boe.5.002435
Subject(s) - hyperplane , raman spectroscopy , k nearest neighbors algorithm , breast cancer , computer science , pattern recognition (psychology) , artificial intelligence , support vector machine , optics , cancer , materials science , mathematics , physics , medicine , geometry
In order to promote the development of the portable, low-cost and in vivo cancer diagnosis instrument, a miniature laser Raman spectrometer was employed to acquire the conventional Raman spectra for breast cancer detection in this paper. But it is difficult to achieve high discrimination accuracy. Then a novel method of adaptive weight k-local hyperplane (AWKH) is proposed to increase the classification accuracy. AWKH is an extension and improvement of K-local hyperplane distance nearest-neighbor (HKNN). It considers the features weights of the training data in the nearest neighbor selection and local hyperplane construction stage, which resolve the basic shortcoming of HKNN works well only for small values of the nearest-neighbor. Experimental results on Raman spectra of breast tissues in vitro show the proposed method can realize high classification accuracy.