Abstract：In order to detect an important kind of abnormal and discontinuous respiratory sounds ——crackles in lung sounds, a new detection algorithm has been proposed based on an emerging theory of fractional Hilbert transform. By applying fractional Hilbert transform to lung sound signals, a twodimension image with texture feature can be generated. The features of combination of dark and bright interlaced strips in texture images were corresponding to features of crackle signals in lung sounds, which can be employed to detect crackles. Two groups of detection experiments were carried out. First, real normal lung sounds signals with simulated crackles embedded were tested, and 100% crackles were detected. Then, lung sounds of patients with pulmonary disease were tested, and the sensitivity of detection accuracy reached to 97%. Thus, the effectiveness of crackle detection based on fractional Hilbert transform has been validated.
李真真*吴效明. 基于分数阶Hilbert 变换二维纹理特征的罗音检测算法[J]. 中国生物医学工程学报, 2013, 32(3): 299-304.
LI Zhen Zhen WU Xiao Ming*. Algorithm of Crackle Detection Based on TwoDimensional Texture Features by Fractional Hilbert Transform. journal1, 2013, 32(3): 299-304.
［1］Gavriely N, Breath sounds methodogy ［M］. // Breath sounds in the time domain: crackles. Florida: CRC Press, 1995:18-29.
［2］Reichert S, Gass R, Brandt C, et al. Analysis of respiratory sounds: state of the art ［J］. Clinical Medicine: Circulatory, Respiratory and Pulmonary Medicine, 2008, (5): 45-58.
［3］Murphy RLH, Holford SK, Knowler WC. Visual lungsound characterization by timeexpanded waveform analysis ［J］.New Eng J Med, 1977, 296(17): 968-971.
［4］Kaisla T, Sovijarvi A, Piirila P, et al. Validated method for automatic detection of lung sound crackles ［J］. Medical Biological Engineering & Computing, 1991, 29(5): 517-521.
［5］Dorantes G, Charleston S, Gonzalez R, et al. Crackles detection using a timevariant autoregressive model ［C］. // Dumont G, Galiana H,eds. 2008 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Vancouver: IEEE, 2008: 1894-1897.
［6］Yeginer M, Kahya YP. Feature extraction for pulmonary crackle representation via wavelet networks ［J］.Computers in Biology and Medicine, 2009, 39(8): 713-721.
［7］Lu Xiaoguang, Bahoura M. An integrated automated system for crackles extraction and classification ［J］. Biomedical Signal Processing and Control, 2008, 3(3): 244-254.
［8］Du Minghui, Chan FHY, Lam FK, et al. Crackle detection and classification based on matched wavelet analysis［C］. // Jaeger R,eds. 1997 19th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Chicago: IEEE, 1997:1638-1641.
［9］Sonia C, Ramon G, Tomas A. Crackle sounds analysis by empirical mode decomposition ［J］.IEEE Engineering in Medicine and Biology Magazine, 2007, 26(1): 40-47.
［10］Lohmann AW , Soffer EH. Fractional Hilbert Transform ［J］. Opt Lett, 1996, 21: 281-283.
［11］李真真，杜明辉，吴效明. 基于分数阶希尔伯特变换的罗音特征提取 ［J］. 华南理工大学学报（自然科学版）, 2011,39(12): 38-42.
［12］Ran Tao, Li Xuemei, Wang Yue. Generalization of the fractional Hilbert transform ［J］.IEEE Signal Processing Letters, 2008, 15: 365-368.
［13］Pei Soochang, Yeh Minhung. Discrete Fractional Hilbert Transform ［J］. IEEE Transaction on Circuits and Systems, Analog and Digital Signal Processing, 2000, 47 (11): 1307-1311.
［14］Pei Soochang, Hsue Wenliang, Ding Jianjiun. Discrete fractional fourier transform based on new nearly tridiagonal commuting matrices ［J］. IEEE Transaction on Signal Processing, 2006, 54(10): 3815-3828.
