Breath Sound Characteristics

Respiratory sounds have different characteristics depending on the location of recording. However, they are mainly divided into two classes: upper airway (tracheal) sounds usually recorded over the suprasternal notch of trachea, and lung sounds that are recorded over different locations of the chest wall either in the front or back. Tracheal sounds do not have much of diagnostic value as the upper airway may not be affected in serious lung diseases, while lung sounds have long been used for diagnosis purposes.

Lung sounds amplitude is different between persons and different locations on the chest surface and varies with flow. The peak of lung sound is in frequencies below 100 Hz. The lung sound energy drops off sharply between 100 and 200 Hz but it can still be detected at or above 800 Hz with sensitive microphones. The left top graph of Fig. 4.1 shows a typical airflow signal measured by a mouth-piece pneumotachograph. The positive values refer to inspiration and the negative values refer to expiration airflow. The left bottom graph shows the spectrogram (or sonogram) of the lung sound recorded simultaneously with that airflow signal. The spectrogram is a representation of the power spectrum for each time segment of the signal. The horizontal axis is the duration of the recording in seconds and the vertical axis is the frequency range. The magnitude of the power spectrum is therefore shown by color, where the pink color represents above 40 dB whereas the dark gray represents less than 4 dB of the power in Fig. 4.1. As it can be observed, the inspiration segments of the lung sound have much higher frequency components than expiration segments. In other words, inspiration sounds are louder than expiration sounds over the chest wall and this observation is fairly consistent among the subjects [14]. The right graph shows the average spectrum of all inspiration segments compared to that of expiration segments. Again, as it can be observed, there is about 6-10 dB difference between inspiration and expiration power spectra over a fairly large frequency range.

On the other hand, tracheal sound is strong and covers a wider frequency range than lung sound. Tracheal sound has a direct relationship with airflow and covers a frequency range up to 1500 Hz at the normal flow rate. Similar to the previous figure, the left graphs of Fig. 4.2 show a typical airflow signal on the top and the associated spectrogram of the tracheal signal on the bottom. As it can be observed, the tracheal sound signal is much louder than that of lung sound. However, the difference in inspiration and expiration power of the tracheal sound signal

FIGURE 4.1: A typical lung sound signal spectrogram (left graph) along with the average spectra of inspiration and expiration (right graph) and the corresponding flow (top graph)

FIGURE 4.2: A typical tracheal sound signal spectrogram (left graph) along with the average spectra of inspiration and expiration (right graph) and the corresponding flow (top graph)

FIGURE 4.2: A typical tracheal sound signal spectrogram (left graph) along with the average spectra of inspiration and expiration (right graph) and the corresponding flow (top graph)

varies among the subjects greatly. In some people, there is not much difference while in others such as the subject in this example the expiratory sound is louder than the inspiratory sound.

The relationship of flow with power density of tracheal and lung sounds leads to the idea that at least the breath phases, i.e., inspiration/expiration, and the onset of breaths can be determined acoustically without the actual flow measurement; this was investigated a few years ago. The actual flow estimation by acoustical means, however, requires many more signal processing techniques and investigations. We will discuss this issue in more details in the following sections.

Like all other biological signals, respiratory sounds also differ among the subjects as their chest size and body mass are different. However, using digital signal processing techniques, researchers have sought methods to extract some characteristic features ofthe respiratory sounds that can be used for diagnostic purposes between healthy individuals and patients with various respiratory diseases. This has been the main motivation for most of respiratory sound researches, which we will address in more detail in the following sections.

Was this article helpful?

0 0

Post a comment