Brianne Hamilton and Dr. Scott L. Thomson, Mechanical Engineering
The study of vocal fold vibration allows for greater understanding of how speech production occurs and of the prevention, diagnosis, and treatment of associated physiological disorders. Until now an environment to test vocal fold models which are anatomically similar in both form and function has not been available. It was necessary to have a larynx model in which artificial vocal folds can be placed and subjected to conditions found in the human body in order to attain a more comprehensive understanding of vocal folds and their function. This project was part of a team working with Dr. Scott L. Thomson to generate a working model of the human larynx.
This project’s aim was to construct a computerized model of the laryngeal cartilage framework and to use this model to fabricate synthetic cartilage models. The understanding gained from these models will be of benefit to the clinical, surgical, and pedagogical voice communities. In the closer future these replicas of larynx cartilage will continue to be used to study the flow of air through the vocal folds.
Two dimensional anonymous MRI images of the human neck larynx were acquired from magnetic resonance images. These images had to be combined into a three dimensional model to accurately reproduce the human larynx. An obstacle which had to be overcome was to deal with converting the images into data which the computer could read into a three-dimensional computerized model. Originally I was working with the computer science department to convert the images into a format that could be read by a CAD system using several different file formats and creating loops of data points around individual selected regions. Each of the formats tried was unable to be read properly into the CAD system and had several errors.
Eventually it was decided to purchase an image reading software package which read the different shades of grey which occur in MRI images. By recognizing and selecting different regions of images it was possible to separate the different grays into individual cartilage pieces. An understanding of the location and shape of the cartilage was essential in constructing these models. The thyroid, cricoid, and arytenoid cartilages were selected from the images and assembled into four different three-dimensional models, which were then converted into .stl files. This file was then read by a three-dimensional printer which deposits plastic in layers to form models.
After the individual cartilage pieces were printed. Rubber molds of the pieces were made so a white plastic could be used to make multiple sets of cartilage. Another challenge was trying to find an appropriate way to assemble the cartilage to construct the airway. Dual layer vocal folds were made by another part of the research group from MRI images and silicone models were made to fit inside the cartilage. Silicone of different consistencies was used when assembling the larynx to allow movement when force was applied. Also a membrane was added between the cricoid and the thyroid cartilage to prevent air from escaping through the front. One of the finished models is shown in Figure 1.
The completed models were presented at the Acoustical Society of America Conference , and the response of the vocal folds when subjected to pressure in the artificial environment. Figure 2 shows the setup used for applying pressure to the vocal folds.
This experience has been valuable and has helped me to understand better what goes into engineering research. In the future, if I decide to continue my education further, this has helped me to understand better what would be involved. The experience collaborating with a team for a finished product has also helped me as I have started my career. I graduated in August in Mechanical Engineering and currently have a job in the medical industry doing mechanical engineering for cancer treatment devices.