Residual food in the pharynx after swallowing is one of the causes of aspiration. For the diagnosis of dysphagia, screening tests such as oral diadochokinetic and repeated salivary swallowing test (RSST) are applied. If necessary, a detailed examination is performed by videofluorography (VF) or video endoscopic (VE). Since VF and VE are invasive tests, accurate and easy-to-apply screening tests with less physical burden on patients are needed. However, it is difficult to detect food residues in the pharynx using current screening tests. Therefore, an objective, non-invasive method based on speech signal processing technology is one possible approach for detecting pharyngeal residues [1, 2]. This study focuses on the piriform fossa, which may be likely to be aspirated if food remains. We first hypothesized that twisting the neck and rotating the head simulates the residue of food in the piriform fossa. We confirmed this hypothesis by comparing the log power spectral envelope of the simulated speech by the vocal tract model and human speech. Finally, we designed a face-direction identification system, and its identification accuracy was 0.94 for the long-vowels/e/.
CITATION STYLE
Hosoyama, T., Koto, M., Nishimura, M., Nishida, M., Horiuchi, Y., & Kuroiwa, S. (2020). Analysis of acoustic features affected by residual food in the piriform fossa toward early-detection of dysphagia. In Smart Innovation, Systems and Technologies (Vol. 192, pp. 171–177). Springer. https://doi.org/10.1007/978-981-15-5852-8_16
Mendeley helps you to discover research relevant for your work.