EAPSI: Validation of Three-Dimensional Endoscopy System with Synthetic Vocal Fold Models

在美国，声音障碍影响着许多人，但有效的临床治疗受到声音产生物理学知识、正确诊断和诊断挑战的限制。喉镜检查，使用专门设计的内窥镜检查声带，是临床和研究环境中声带可视化的重要技术。然而，该技术仅限于提供有关运动的信息在两个维度时，声带的运动已被证明是三维的。日本立命馆大学的Isao Tokuda博士开发了一种使用带有两个摄像头的喉镜重建三维声带运动的方法，但该技术尚未得到广泛验证。与Tokuda博士合作，该项目将使用最近用于语音生物力学研究的合成声带模型来验证立体内窥镜系统。合成声带模型最近已被用于语音生物力学研究，并将用于拟议的项目，因为它们提供更多的控制，更强大，更容易获得比切除动物模型。将使用多个摄像机和常见的计算机视觉技术来验证立体内窥镜系统，方法是用立体内窥镜系统和高速系统记录和重建合成的自振荡声带模型的振动表面，并比较结果。立体内窥镜系统的验证将允许其在临床环境中使用，以及声带运动的定量分析，从而提高对发声中所涉及的动态的理解。这个NSF EAPSI奖是与日本科学促进协会合作资助的。

Voice disorders affect many people in the United States but effective clinical treatment is limited by knowledge of the physics of voice production proper diagnosis and by challenges in diagnosis. Laryngoscopy, use of a specially designed endoscope to examine the vocal folds, is an important technique for vocal fold visualization in both clinical and research settings. However, the technique is limited to providing information regarding movement in two dimensions when the motion of vocal folds has been shown to be three-dimensional. Dr. Isao Tokuda at the Ritsumeikan University in Japan has developed a method for reconstructing three-dimensional vocal fold motion using a laryngoscope with two cameras, but the technique has not yet been extensively validated. In collaboration with Dr. Tokuda, this project will use synthetic vocal fold models, recently used in voice biomechanics research, to validate the stereo-endoscopy system. Synthetic vocal fold models have recently been used in voice biomechanics research and will be used in the proposed project because they offer more control, are more robust, and are easier to acquire than excised animal models. Multiple cameras and common computer vision techniques will be used to validate the stereo-endoscopy system by recording and reconstructing the vibrating surface of synthetic, self-oscillating vocal fold models with both the stereo-endoscopy and high speed systems and comparing the results. The validation of the stereo-endoscopy system will allow its use in clinical settings as well as for quantitative analysis of vocal fold motion, leading to improved understanding of the dynamics involved in phonation. This NSF EAPSI award is funded in collaboration with the Japan Society for the Promotion of Science.