1 Journal of Medical Engineering & Technology, 2011, 1-10, Early Online Copyright © 2011 Informa UK, Ltd. ISSN 0309-1902 print/ISSN 1464-522X online DOI: 10.3109/03091902.2011.595529 *Corresponding author. Tel: +91-3222-283556/1470. Fax: +91-3222-255303 Email: ashokrpe@gmail.com, gsaha.iitkgp@gmail.com During the recording time of lung sound (LS) signals from the chest wall of a subject, there is always heart sound (HS) signal interfering with it. This obscures the features of lung sound signals and creates confusion on pathological states, if any, of the lungs. A novel method based on empirical mode decomposition (EMD) technique is proposed in this paper for reducing the undesired heart sound interference from the desired lung sound signals. In this, the mixed signal is split into several components. Some of these components contain larger proportions of interfering signals like heart sound, environmental noise etc. and are filtered out. Experiments have been conducted on simulated and real-time recorded mixed signals of heart sound and lung sound. The proposed method is found to be superior in terms of time domain, frequency domain, and time–frequency domain representations and also in listening test performed by pulmonologist. Keywords: Heart sound (HS), Lung sound (LS), Empirical mode decomposition (EMD), Power spectral density (PSD), Spectrogram Introduction Lung sounds are produced by vertical and disruptive low [1] within lung airways during inspiration and expiration of air [2]. Heart sounds are created by the low of blood into and out of the heart and by the movement of structures involved in the control of this low [3] and the dominant frequency components are in the range of 20–150 Hz [4]. Heart sounds are clearly audible in lung sounds recorded on the anterior chest and may be heard to a lesser extent in lung sounds recorded over posterior lung lobes. Ater the invention of stethoscope by Lannec [5] in 1816, it is employed as a primary tool to detect lung diseases, but traditional auscultation with a stethoscope constraints a diagnostic test mainly due to the interference of the heart sound with lung sounds. he heart sound signal interference obscures the interpretation of lung sounds and leads to an inaccurate diagnostic result of the lung diseases. It is highly desirable, for accurate assessment of the lung diseases based on lung sound information, to remove the heart sound interfer- ence. he secondary drawback of the auscultation technique is the limitation of the human auditory system [6]. A less experienced physician may ind it more diicult to interpret LS in presence of HS. However, this problem may be solved by the skilled and experienced physician. Apart from the Heart sound interference, LS signals are afected by noises gener- ated from surrounding environment and friction rubs. he environment and friction rubs noise are reduced by using a sound proof room and with proper placement of microphone. he methods used for HS separation from LS signal remained an area of interest among researchers. Methods like adaptive iltering [4],[7]–[9], wavelet-based iltering [10–12], and short time Fourier transform (STFT) based approach [13] have been tried out with some success. In this paper, we propose a new method based on EMD technique [14], for separation of HS signals from LS signals. he EMD technique decomposes the signal into a number of diferent time scales or intrinsic mode functions (IMFs). his allows us to ilter signal components individually instead of iltering the original signal. It is found that some components of the signal highlight the presence of interference or noise present in the LS signal. hus it becomes easier to remove them from the signal using the EMD technique [15]. he proposed method has the ability to remove eiciently the HS interference from LS signals without any requirement of the reference signal unlike [7]. he idea introduced by S. Ari et al. [16] was employed in our experiment for detection of the HS peaks within the LS signals. his is followed by removal of HS signal and estimation of corresponding sample values. his paper is organized as follows: A brief description of the empirical mode decomposition (EMD) technique is presented inEMD technique section . Data acquisition section discusses about the database used in this study, Methodology section presents the proposed method. Experimental outcomes and its discussion is presented in Results and discussion section followed by conclusions. INNOVATION Reduction of heart sound interference from lung sound signals using empirical mode decomposition technique Ashok Mondal*, P.S. Bhattacharya and Goutam Saha Department of Electronics and Electrical Communication Engineering, Indian Institute of Technology, Kharagpur, India J Med Eng Technol Downloaded from informahealthcare.com by Indian Institute of Technology Kharagpur on 09/02/11 For personal use only.