Swallowing dysfunction (dysphagia) is a common disorder in patients with neurological impairments, head/neck injuries or brain-stem stroke. The main objectives of this thesis were to use acoustical analysis of swallowing and breath sounds for 1) understanding the swallowing mechanism and the main cause of dysphagia, and 2) developing a noninvasive diagnostic technology to detect swallowing aspiration (the entry of bolus into airway); thus, identifying individuals at high risk of severe dysphagia.
As the first objective of the study, swallowing mechanism modeling in two groups of healthy individuals and dysphagic patients (due to cerebral palsy or stroke) was approached by using two different assumptions to relate the swallowing sounds either to the pharyngeal response or to the neural activities that initiate the swallow. The results showed that the model with the assumption of neural activities as the cause of dysphagia was a better fit to the available data.
As the second main objective of the study, we analyzed breathing and swallowing sounds of 50 dysphagic individuals during the fiberoptic endoscopic evaluation of swallowing (FEES) or the videofluoroscopic swallowing study (VFS). The results showed 91% sensitivity and 85% specificity in identifying patients with severe aspirations. Also, the algorithm was able to detect the silent aspiration among the swallows of each patient.
The proposed methods led to development of a non-invasive and reliable diagnostic/screening tool as an aid to the clinical examination of swallowing. The proposed acoustic method can be performed at the patients’ bedside to determine the appropriate further assessment or a dietetic treatment; thus, reducing the health care cost by prioritizing the patients’ referrals to the VFS/FEES tests.
| Identifer | oai:union.ndltd.org:MANITOBA/oai:mspace.lib.umanitoba.ca:1993/23974 |
| Date | 01 1900 |
| Creators | Sarraf Shirazi, Samaneh |
| Contributors | Moussavi, Zahra (Electrical and Computer Engineering), Sherif, Sherif (Electrical and Computer Engineering) Pawlak, Miroslaw (Electrical and Computer Engineering) Popplewell, Neil (Mechanical Engineering) Ellis, Christopher G. (Medical Biophysics, University of Western Ontario) |
| Publisher | Springer, Springer, Springer, IEEE |
| Source Sets | University of Manitoba Canada |
| Detected Language | English |
Page generated in 0.0025 seconds