三维声带振动的半喉设置研究

In order to understand how human and mammalian voice is created, it is important to investigate the vibrations of the vocal folds. Since they are difficult to study inside the body, excised larynges have been used to obtain more detailed information on the behavior and properties of the vocal folds. In this video, we demonstrate how to create a so-called hemi-larynx setup.

One half of the larynx is removed in order to visualize the vocal fold vibration not only from above but also from the side. This is beneficial for studying vibratory features which remain hidden in a conventional setup using a whole larynx. In this demonstration, the larynx of a red deer will be used.

Since the utilized larynx specimens are usually quick-frozen in liquid nitrogen and stored in the freezer, it is necessary to defrost the chosen specimen properly in order not to alter the biomechanical properties of the tissue. First, insert the frozen larynx into a twofold autoclave or any other plastic bag with waterproof sealing. Also add a weight, which helps to keep the larynx under the water surface.

Seal the bag and put it into a water bath with a temperature of about 30 degrees Celsius until the larynx is completely defrosted. After the larynx is defrosted completely, take it out of the bag and clean it thoroughly with saline solution. If present, remove superfluous structures such as external neck muscles or the hyoid bone.

For some bioacoustical research questions, the supralaryngeal structures might be kept intact, which however limits the visibility of the vocal folds from the top. Shorten the trachea to a length of about 45 centimeters so that it can be used later for mounting the larynx onto an air supply tube. Before proceeding with the preparation, always check the larynx for potential tissue anomalies and occasionally appearing cracks introduced during the flash freezing process.

The next step involves exposing one half of the thyroid cartilage. First, remove the external laryngeal muscles using a scalpel. Start the cutting procedure with an initial vertical cut through the anterior part of the thyroid cartilage.

Carefully lead the cut more on to the side that should be removed to avoid damage of the vocal fold that needs to be preserved and studied. If the cartilage is to stiff due to ossification, use a small saw. Then, cut the cricoid cartilage midsagittaly.

Create the incision vertically from in between the arytenoid cartilages and stop approximately and the horizontal level of the inferior thyroid notch. Connect both the incisions by making a horizontal cut starting at the inferior end of the vertical incision in the cricoid cartilage and lead the cut towards the inferior thyroid notch. Fold the side of the larynx which is going to be removed towards the anterior end and make a vertical cut through the soft tissue on the inner side of the thyroid cartilage.

This is the most crucial moment of the whole procedure. Therefore, be extremely careful when creating the incision from the anterior attachment of the vocal folds to the thyroid cartilage. If the thyroid cartilage is sectioned properly, the resulting glottis will always contain a slight anterior gap.

In order to expose the vocal folds to obtain a good superior view, it is sometimes useful to remove also the structures above the vocal folds. In some cases, the inner soft laryngeal tissues lose their connection to the thyroid cartilage after the cutting procedure and interfere with the vocal fold. In that case, removal of this superfluous tissue is inevitable.

Finally, remove a small part of the posterior thyroid cartilage in order to facilitate vocal fold adduction through the arytenoids. The larynx now has an L-shaped mortis matching the L-shaped class plate arrangement utilized in the setup. The left side of the hemi-larynx provides a medial view where the incisions in the thyroid and the cricoid cartilage are visible together with the remaining vocal fold and the arytenoid cartilage.

The experimental setup for hemi-larynges consists of three basic parts. An air supply tube which delivers warmed and humidified air into the larynx. Then two transparent plates as a substitution for the removed laryngeal parts and the prongs for adduction of the studied vocal fold and larynx stabilization.

Now, we will mount the larynx on the air supply tube. Cover the tube with denture fixative cream and mount the larynx using its trachea. The denture fixative cream works as an adhesive and seals potential gaps.

Fasten the trachea with a plastic tightened strap or some other useful clamp. Cover also the edges of the cut with the denture fixative cream and attach the transparent plates. Stabilize the thyroid cartilage and adduct the vocal fold using the prongs.

At this stage, the hemi-laryngeal setup is ready to use. Apply the air flow and check for any leaks. If the air is escaping to the side, seal the gaps by adding some more fixative cream.

To see the vocal fold vibrations in real time, stroboscopic light can be applied. The main application of the hemi-larynx setup is to observe the vibration of the vocal folds not only from the top but also from the side. For this purpose, two synchronized high-speed video cameras are used.

One for the top view and one for the side view. Here we see the vocal fold vibration of the hemi-larynx documented by the two synchronized cameras running at 6, 000 frames per second. The left panel shows the top view and the right panel shows the side view.

The vocal fold vibrates at the rate of about 70 cycles per second and the captured videos are shown here in slow motion. The space between the vocal fold and the glass plate is called the glottis. It can be quantified by measuring its area as it changes over time.

Here, this glottal area is expressed in pixels. It is zero when the vocal fold is in contact with the glass plate and non-zero when air is flowing through the glottis. Analysis of the acoustic signal resulting from the process of vocal fold vibration in the hemi-larynx shows that the main acoustic excitation events occur at two instances during a vibratory cycle.

At the moment of glottal opening, when the vocal fold loses contact with the glass plate and at the moment of glottic closure when the vocal fold has again reached full contact with the glass plate. In the next animation, the time varying contact area of the vocal fold with the glass plate is shown in orange for one vibratory cycle. Here, the hemi-larynx setup is used to validate a method called electroglottography or EGG.

The simultaneously acquired EGG signal is shown here in blue. There is a reasonably good correlation between the vocal fold contact area and the EGG signal. This is documented in further detail in a recent study from our group.

In the last section of this video, we introduce a novel method to assess vocal fold vibration in three dimensions called chemographic glottal motion analysis. In a digital chemogram, the image information found along one line within a high-speed video is assessed for each video frame and plotted from left to right. The resulting chemogram shows the motion of the vocal fold structure along the scan line as it develops in time.

Chemograms are generated for both the top and the side view with the chemographic scan lines placed exactly in the middle of the vocal fold. The temporal offset of the vocal fold edges is traced. The light blue graph shows the lateral deflection of the vocal fold and the red and green graphs show the vertical deflection of the inferior and superior edge of the vocal fold respectively.

The resulting information can be combined to synthesize an animation showing the motion of the glottis in two dimensions as seen at the glottal midline. In this video, we have provided a tutorial for creating a hemi-larynx setup. We have shown some potential applications for assessing vocal fold vibration not only from the top but also from the side, thus adding crucial empirical data for better understanding a key aspect of voice generation.