Title :
Near infrared spectroscopy combined with high dimensional data analysis applied to diagnosis of endometrial carcinoma
Author :
Yang, Fan ; Zhang, Zhuoyong ; Xiang, Yuhong ; de B Harrington, Peter
Author_Institution :
Dept. of Chem., Capital Normal Univ., Beijing, China
Abstract :
The feasibility of early diagnosis of endometrial carcinoma was studied by high dimensional data analysis (HDDA) that classified near infrared (NIR) spectra of tissues. NIR spectra of 77 specimens of endometrium were collected. The spectra were pretreated by principal component orthogonal signal correction (PC-OSC) and emphatic orthogonal signal correction (EOSC) methods to improve the signal-to-noise ratio (SNR) and remove the influences of background and baseline. The effects of modeling parameters were investigated using bootstrapped Latin-partition methods. The optimal HDDA model of the PC-OSC pretreatment method successfully classified the samples with prediction accuracies of 95.2 ± 1.9%. The proposed procedure proved to be rapid and convenient, which is suitable to be developed as a non-invasive diagnosis method for cancer tissue.
Keywords :
biological tissues; cancer; data analysis; gynaecology; infrared spectroscopy; medical signal processing; patient diagnosis; patient treatment; EOSC method; HDDA model; NIR spectra; PC-OSC pretreatment method; bootstrapped Latin partition method; cancer tissue; data analysis; emphatic orthogonal signal correction; endometrial carcinoma diagnosis; near infrared spectroscopy; noninvasive diagnosis method; principal component analysis; signal-to-noise ratio; Accuracy; Cancer; Data models; Polynomials; Predictive models; Spectroscopy; Training; cancer diagnosis; endometrial carcinoma; high dimensional discriminant analysis; near infrared spectroscopy;
Conference_Titel :
Fuzzy Systems and Knowledge Discovery (FSKD), 2012 9th International Conference on
Conference_Location :
Sichuan
Print_ISBN :
978-1-4673-0025-4
DOI :
10.1109/FSKD.2012.6233734
