Study of the Methods of Wavelet Feature Extraction and Neural Network Classification of Fluorescence Spectra to Improve the Diagnostic Rate of Colonic Earlier Stage Cancer
XIA Dai-lin1, MENG Hong-xia2, ZHANG Yang-de3, HE Ji-shan4
1. The Key Laboratory for Biomedical Photonics of Ministry of Education, Huazhong University of Science and Technology, Wuhan 430074, China 2. Department of Automatization of Collage of Power & Mechanical Engineering of Wuhan University, Wuhan 430072, China 3. National Hepatobiliary and Enteric Surgery Research Center, Central South University, Changsha 410083, China 4. School of Info-physics and Geomatrics Engineering, Central South University, Changsha 410083，China
Abstract：In order to improve the diagnostic rate of earlier stage colonic cancer with laser-induced 5-ALA-PpⅨ fluorescence spectra, a novel method of extraction of fluorescence spectral feature using wavelet analysis and classification using artificial neural network trained with resilient back-propagation algorithm (R-BPNN) was developed. 504 spectra were collected from 8 normal SD rats, and 20 1,2-DMH-induced SD colon cancer models and 12 second generation rats of induced rats. 150 min later trail intravenous injections of 5-ALA dose of 25 mg·kg-1 body weight (BW), and fluorescence spectra excited with 370 nm Ti-laser were collected in vivo. After preprocessing, 12 feature variants were extracted with wavelet analysis. With R-BPNN, all spectra were classified into two categories: normal or abnormal, which included dysplasia, early carcinoma (EC) and advanced carcinoma (AC). The sensitivity and specificity were 98.91% and 97.2% respectively. The accuracy of discriminating dysplasia, early carcinoma, and advanced carcinoma from normal tissue were 91.3%, 98.9% and 98.8% respectively. The result indicated that this method could effectively and easily diagnoses earlier stage colonic carcinomas.
Key words：Fluorescence spectroscopy;Wavelet feature extraction;BP neural network;Colonic cancer
夏代林1,孟红霞2,张阳德3,何继善4 . 提高结肠早癌诊断率的荧光光谱小波特征提取与神经网络分类方法研究[J]. 光谱学与光谱分析, 2006, 26(11): 2076-2079.
XIA Dai-lin1, MENG Hong-xia2, ZHANG Yang-de3, HE Ji-shan4 . Study of the Methods of Wavelet Feature Extraction and Neural Network Classification of Fluorescence Spectra to Improve the Diagnostic Rate of Colonic Earlier Stage Cancer . SPECTROSCOPY AND SPECTRAL ANALYSIS, 2006, 26(11): 2076-2079.
[1] ZHANG Yang-de, LIU Wei-dong(张阳德, 刘蔚东). China Journal of Endoscopy(中国内镜杂志), 2000, 6(5): 1. [2] LUO Qing-ming, GONG Hui, LIU Xian-de, et al(骆清铭, 龚 辉, 刘贤德, 等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 1997, 17(3): 105. [3] Messmann H. Gastrointest Endoscopy Clin. N. Am., 2000, 10(3): 497. [4] Vo-Dinh T, Panjehpour M, Overholt B F, et al. Lasers Surg. Med., 1995, 16: 41. [5] Cothern R M, Sivak M V, Dam J V, et al. Gastrointest Endosc., 1996, 44(2): 168. [6] Jianan Y, Wing P, Zhijian H, et al. Lasers Surg. Med., 2000, 26: 432. [7] Kapadia C R, Cuteuzzila F W, O’Brien K M, et al. Gastroenter., 1990, 99: 150. [8] Schomacker K T, Frisoli J K, Compton C C, et al. Lasers Surg. Med., 1992, 12: 63. [9] Ramanujam N, Mitchell M F, Mahadevan A, et al. Lasers Surg. Med., 1996, 19: 46. [10] Eker C, Rydel R, Svanberg K, et al. Lasers Surg. Med., 2001, 28: 259. [11] XIA Dai-lin, HE Ji-shan, ZHANG Yang-de, et al(夏代林, 何继善, 张阳德, 等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2005, 25(12): 2029. [12] XIA Dai-lin, HE Ji-shan, ZHANG Yang-de(夏代林, 何继善, 张阳德). Journal of Huazhong University of Science and Technology(Nature Science Edition)(华中科技大学学报·自然科学版)，2005, 33(9): 120. [13] Riedmiller M, Braun H. In Ruspini, editor, Proceedings of the IEEE International Conference on Neural Networks (ICNN), San Franciso: IEEE Press, 1993, 586.
