Abstract：To solve the problems of short data and noisy recordings in pathological voice signals, this paper extracted some entropy feature parameters of pathological voice proposed in recent years, including sample entropy, multiscale entropy, fuzzy entropy and hierarchical entropy. Based on hierarchical decomposition method, we developed hierarchical multiscale entropy and hierarchical fuzzy entropy. Support vector machine (SVM) was used to distinguish the test set including 39 cases of normal and 36 cases of pathological voices. Results showed that three level hierarchical entropy，hierarchical multiscale entropy and hierarchical fuzzy entropy all achieved higher recognition rates and better stabilities using the proposed method. Pathological voice’s three level hierarchical fuzzy entropy feature got a better and more stable SVM recognition rate of 97.33% by extracting 2000 points. Compared with fuzzy entropy, the recognition rate was increased about 4.00%. The entropy method provide valuable preference for clinical analysis of short pathological voice time series contaminated by noise, which is benefit for clinical application of pathological voice analysis.
赵冰心 胡维平*. 基于熵和支持向量机的病态嗓音识别[J]. 中国生物医学工程学报, 2013, 32(5): 546-552.
ZHAO Bing Xin HU Wei Ping*. Recognition of Pathological Voice Based on Entropy and Support Vector Machine. journal1, 2013, 32(5): 546-552.
［1］GodinoLlorente JI, GomezVilda P, BlancoVelasco M. Dimensionality reduction of a pathological voice quality assessment system based on gaussian mixture models and shortterm cepstral parameters［J］. IEEE Trans Biomed Eng, 2006,53(10): 1943-1953.
［2］Palumbo A, Amato F, Calabrese B, et al. A novel portable device for pathological voice analysis［C］// International Workshop on Medical Measurements and Applications. Cetraro: IEEE,2009:51-54.
［3］Mitev P, Hadjitodorov S. Fundamental frequency estimation of voice of patients with laryngeal disorders［J］. Information Sciences, 2003,156(1-2):3-19.
［4］Umapathy K, Krishnan S,Parsa V, et al.Discrimination of pathological voices using a timefrequency approach［J］. IEEE Transactions on Biomedical Engineering, 2005,52(3):421-430.
［5］Alonso JB,Díaz F,Travieso CM, et al.Using nonlinear features for voice disorder detection［C］// 3rd Int Conf Nonlinear Speech Process.Barcelona : Speech Communication, 2005:94-106.
［6］Henríquez P,Alonso JB,Ferrer MA, et al. Characterization of healthy and pathological voice through measures based on nonlinear dynamics［J］. IEEE Transactions on Audio, Speech, and Language Processing,2009,17(6):1186-1195.
［7］高俊芬, 胡维平. 基于非线性动力学和高斯混合模型/支持向量机的病态嗓音识别与研究［J］. 生物医学工程学杂志,2012,29(4):750-753.
［8］于燕平，胡维平. 病态嗓音特征的神经网络选择［J］. 中国生物医学工程学报,2009,28(3):327-331.
［9］Pincus SM. Approximate entropy as a measure of system complexity［J］. Proc Natl Acad Sci USA,1991,88(6):2297-2301.
［10］Richman JS, Moorman JR. Physiological timeseries analysis using approximate entropy and sample entropy［J］. Am J Physio:Heart Circ Physiol, 2000,278(6):H2039-H2049.［11］Costa M, Goldberger AL, Peng CK. Multiscale entropy analysis of complex physiologic time series［J］. Phys Rev Lett,2002,89(6):068102.
［12］Chen Weiting, Wang Zhizhong, Xie Hongbo, et al. Characterization of surface EMG signal based on fuzzy entropy［J］. IEEE Transactions on Neural Systems and Rehabilitation Engineering,2007,15(2):266-272.
［13］Jiang Ying, Peng CK, Xu Yuesheng. Hierarchical entropy analysis for biological signals［J］. Journal of Computational and Applied Mathematics, 2011, 236(5): 728-742.
［14］Lake DE, Richman JS, Griffin MP, et al. Sample entropy analysis of neonatal heart rate variability［J］. Am J Physio Regul Integr Comp Physio, 2002,283(3):789-797.
［15］Chen Weiting, Zhang Jun, Yu Wangxin, et al. Measuring complexity using FuzzyEn, ApEn, and SampEn ［J］. Medical Engineering & Physics, 2009, 31(1):61-68.
［16］Xie Hongbo, Chen Weiting, He Weixing, et al. Complexity analysis of the biomedical signal using fuzzy entropy measurement ［J］. Applied Soft Computing, 2011, 11(2) 2871-2879.
［17］Saeedi NE, Almasganj F, Torabinejad F. Support vector wavelet adaptation for pathological voice assessment ［J］. Comput Biol Med, 2011, 41 (9): 822-828.