Takayoshi Nakai

A visual stethoscope to detect the position of the tracheal tube

BACKGROUND: Advancing a tracheal tube into the bronchus produces unilateral breath sounds. We created a Visual Stethoscope that allows real-time fast Fourier transformation of the sound signal and 3-dimensional (frequency-amplitude-time) color rendering of the results on a personal computer with simultaneous processing of 2 individual sound signals. The aim of this study was to evaluate whether the Visual Stethoscope can detect bronchial intubation in comparison with auscultation. METHODS: After induction of general anesthesia, the trachea was intubated with a tracheal tube. The distance from the incisors to the carina was measured using a fiberoptic bronchoscope. While the anesthesiologist advanced the tracheal tube from the trachea to the bronchus, another anesthesiologist auscultated breath sounds to detect changes of the breath sounds and/or disappearance of bilateral breath sounds for every 1 cm that the tracheal tube was advanced. Two precordial stethoscopes placed at the left and right sides of the chest were used to record breath sounds simultaneously. Subsequently, at a later date, we randomly entered the recorded breath sounds into the Visual Stethoscope. The same anesthesiologist observed the visualized breath sounds on the personal computer screen processed by the Visual Stethoscope to examine changes of breath sounds and/or disappearance of bilateral breath sound. We compared the decision made based on auscultation with that made based on the results of the visualized breath sounds using the Visual Stethoscope. RESULTS: Thirty patients were enrolled in the study. When irregular breath sounds were auscultated, the tip of the tracheal tube was located at 0.6 ± 1.2 cm on the bronchial side of the carina. Using the Visual Stethoscope, when there were any changes of the shape of the visualized breath sound, the tube was located at 0.4 ± 0.8 cm on the tracheal side of the carina (P < 0.01). When unilateral breath sounds were auscultated, the tube was located at 2.6 ± 1.2 cm on the bronchial side of the carina. The tube was also located at 2.3 ± 1.0 cm on the bronchial side of the carina when a unilateral shape of visualized breath sounds was obtained using the Visual Stethoscope (not significant). CONCLUSIONS: During advancement of the tracheal tube, alterations of the shape of the visualized breath sounds using the Visual Stethoscope appeared before the changes of the breath sounds were detected by auscultation. Bilateral breath sounds disappeared when the tip of the tracheal tube was advanced beyond the carina in both groups.

Single Channel Speech Enhancement Using Adaptive Soft-Thresholding with Bivariate EMD

This paper presents a novel data adaptive thresholding approach to single channel speech enhancement. The noisy speech signal and fractional Gaussian noise (fGn) are combined to produce the complex signal. The fGn is generated using the noise variance roughly estimated from the noisy speech signal. Bivariate empirical mode decomposition (bEMD) is employed to decompose the complex signal into a finite number of complex-valued intrinsic mode functions (IMFs). The real and imaginary parts of the IMFs represent the IMFs of observed speech and fGn, respectively. Each IMF is divided into short time frames for local processing. The variance of IMF of fGn calculated within a frame is used as the reference term to classify corresponding noisy speech frame into noise and signal dominant frames. Only the noise dominant frames are soft-thresholded to reduce the noise effects. Then, all the frames as well as IMFs of speech are combined, yielding the enhanced speech signal. The experimental results show the improved performance of the proposed algorithm compared to the recently reported methods.

Noise reduction of speech signal based on phase spectrum estimation

In the process of short-time Fourier transform (STFT), signal is departed into two components, which are magnitude spectrum and phase spectrum. It was believed that the magnitude spectrum includes most of the information of the speech, and phase spectrum was kept unmodified due to the human auditory system is phase deaf. In this paper, we focused on phase spectrum estimation of noisy speech for a speech-enhancement system. A novel phase spectrum estimation method for speech enhancement system using high resolution STFT is proposed. An approximate value of the phase spectrum is estimated from the high resolution STFT. Then proposed phase estimation method is compared with analysis modification synthesis (AMS) and phase spectrum compensation (PSC) methods. The experiment results show the effectiveness of the phase spectrum estimation and significant contribution to the speech quality.

A visual stethoscope for pediatric patient

SIR––In pediatric patients, a tracheal tube can be advanced until breath sound is apparent in one lung and then the tube is pulled back by 2–4 cm to find an appropriate position (1). We evaluated a system for analysis and visualization of breath and heart sounds (Kou Planning Corporation, Hamamatsu, Japan) to assist with this procedure. The system allows real-time fast Fourier transform of the sound signal and three dimensional (frequency– amplitude–time) rendering of the results on a personal computer with processing of two individual sound signals simultaneously. The analyzing and visualizing software was developed by Shizuoka University Graduate School of Engineering (Hamamatsu, Japan). Two precordial stethoscopes for the left and right sides were used to obtain breath and heart sounds. A stethoscope connected to a microphone (AT805F; audio-technica, Tokyo, Japan) was placed at the midpoint between the nipple and the midclavicle. The normal breath and heart sounds of a pediatric patient after induction of anesthesia with tracheal intubation are shown in Figure 1a. A frequency band of the heart sound was located between 60 and 120 Hz. A frequency band of the breath sound was located between 120 and 480 Hz. Left breath sound disappeared after advancement of the tracheal tube into the right bronchus (Figure 1b). The tube was then pulled back and left heart sound resumed (not shown in figure). Further clinical evaluation of this system is in progress under various conditions. This study was supported by Japan Society for the Promotion of Science KAKENHI (18591699) and Shizuoka prefecture research and development grant for industries, universities and government in Japan 2003. Akira Suzuki* Hiroshi Makino* Yoshimitsu Sanjo† Takayoshi Nakai‡ Keita Mochizuki‡ Yoshito Shiraishi* Takasumi Katoh* Shigehito Sato* *Department of Anesthesiology and Intensive Care, School of Medicine, Hamamatsu University, Hamamatsu, Japan †Department of Anesthesiology, National Defense Medical College, Tokorozawa, Japan Graduate School of Engineering, Shizuoka University, Hamamatsu, Japan (email: akiras@hama-med.ac.jp)

Analysis of acoustic properties of the nasal tract using 3-D FEM

In order to examine the acoustic effects of the complicated morphological construction of the nasal tract, we have analyzed acoustic models constructed according to measurements by magnetic resonance imaging (MRI). A prototype model was made to be as similar as possible with the actual nasal tract, which is asymmetrical in the left and right passages and has complicated cross-sectional shapes. Sound pressure, particle velocity and sound intensity in the model were calculated by the finite element method (FEM). Several modifications were applied to the shape of the prototype model in order to learn what acoustical effects are produced by the following modifications: (1) models having an elliptic shape, (2) models with and without a pair of maxillary sinuses, (3) a left-right symmetry model in which one passage is modified to be identical with the other passage, and (4) models having narrowed or blocked passages. The results show that the poles and zeros are produced and shifted by a mutual branching effect caused by the left-right asymmetry of the nasal passages and an additional side-branch effect of the sinus cavities. Those effects are also produced by complicated cross-sectional shapes of the nasal tract in the 3 kHz region. The reduced cross-section in the narrowed-passage models causes a shift of the pole and zero frequency and weakens the mutual side-branch effect of the left and right passages when either of them are excessively narrowed.

Noise reduction using modified phase spectra and Wiener Filter

In this paper our aim is to investigate a modified Wiener filter for the enhancement of speech that has been corrupted by fluctuating noise. In real noise environment, where the SNR of speech maybe fluctuating about 5 to 10dB, because typical speech enhancement methods always estimate the noise power spectrum directly by the average noise power spectrum of the pauses in speech, but there still some error and it is impossible to track the noise in speech interval. In order to settle this problem, our proposed method estimates the noise spectrum in the form of Posteriori SNR according to its Gaussian statistical model and track the noise spectrum form degraded speech directly and changes at every frame in the noise reduction processing of speech. Then use an iterative structure contains noise estimation, modified wiener filter and phase spectrum. Finally, the performance of the proposed method was tested in 4 kinds of fluctuating noises, and resulting in improved results over classical MMSE algorithms at the low fluctuating SNR.

Speech production by a vocal cords‐vocal tract‐vocal tract wall vibration model

This research was aimed at finding the effects of finitehess in the mechanical impedance of the vocal organs on speech sounds. A speech synthesis program has been developed considering the vocal tract wall vibration that causes sound leakage from the vocal cavity to the nasal cavity even when the velum is closed. The radiation of sound from the outer surface of the vocal tract is not always negligibly small, and its effect is enhanced by the nasal cavity. The synthesis program has the parameters of the physical properties and pressure of ambient gases, as well as the mechanical impedance of a yielding vocal tract wall. The wall vibration is treated as a small perturbation in the area function of the vocal tract. The following items will be discussed in the paper: comparisons between vowels with and without vocal tract wall vibration, sounds from the mouth opening, nostrils, and vocal tract wall, and the distribution of the strength of the wall vibration along the vocal tract. The effect of the pressure of...

Simulation of Normal Incidence Sound Absorption Coefficients of Perforated Panels With/Without Glass Wool by Transmission Line Parameters in a Two-Port Network

This paper describes simulation of normal incidence sound absorption coefficients of perforated panels by transmission line parameters in a two-port network. Maa and Sakagami have investigated micro perforated panels, MPP. But their theories can treat only near 1 % perforation rates of perforated panels with back cavities. If sound propagates as a plane wave, sound propagation can be represented as transmission line parameters in a two-port network. Perforated panels, back cavities, and glass wool absorption materials are represented as matrix of transmission line parameters, respectively. Transmission line parameters of a perforated panel with a back cavity are calculated as multiplication of their matrices. An input impedance can be calculated from the transmission line parameters. A normal incident absorption coefficient is calculated from the input impedance. Holes of the perforated panels have losses of viscous friction and thermal conduction at their walls. Simulations are done in the condition of 0.25 mm to 5 mm diameters of holes, 0.25 % to 25 % perforation rates, 0.5 mm to 5 mm thickness of the perforated panels with back cavities in which there are or are not glass wool absorption materials. The results of these simulations are good agreements with the results of our measurements by transfer function method except in the condition of more than 1 mm diameter of holes.

Simulation of frequency characteristics of bone conduction by own speech

When a speaker himself/herself hear his/her own speech, it is different from recorded speech. It is known that it is due to bone conduction. But we measured sound at the entrance of the ear, and it is shown that speech heard by speaker himself/herself is almost agreement with sound at the entrance of the ear at more than 1 kHz. This time, we simulate frequency characteristics of bone conduction. The vocal tract have loss factor of air viscosity and heat conduction of the wall, and it is assumed that the ear drum is vibrated by vibration of the vocal tract. These results are reported.

A study on transvelar coupling for non-nasalized sounds

Previous studies have found that the velum in speech production may not only serve as a binary switch with on-off states for nasal and non-nasal sounds, but also partially alter the acoustic characteristics of non-nasalized sounds. The present study investigated the unique functions of the velum in the production of non-nasalized sounds by using morphological, mechanical, and acoustical measurements. Magnetic resonance imaging movies obtained from three Japanese speakers were used to measure the behaviors of the velum and dynamic changes in the pseudo-volume of the pharyngeal cavity during utterances of voiced stops and vowels. The measurements revealed no significant enlargements in the supraglottal cavity as subjects uttered voiced stops. It is found that the velum thickness varied across utterances in a way that depended on vowels, but not on consonants. The mechanical and acoustical observations in the study suggested that the velum is actively controlled to augment the voice bars of voiced stops, and nostril-radiated sound is one of the most important sources for voice bars, just as is laryngeal wall vibration. This study also proposed a two-layer diaphragm model that simulates transvelar coupling during the production of non-nasalized speech sounds. The simulation demonstrated that the model accurately represented the basic velar functions involved in speech production.