Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

Noninvasive Dynamic Characterization of Swallowing Kinematics and Impairments in High Resolution Cervical Auscultation via Deep Learning

Khalifa, Yassin (2022) Noninvasive Dynamic Characterization of Swallowing Kinematics and Impairments in High Resolution Cervical Auscultation via Deep Learning. Doctoral Dissertation, University of Pittsburgh. (Unpublished)

Updated Version

Download (61MB) | Preview


Swallowing is a complex sensorimotor activity by which food and liquids are transferred from the oral cavity to the stomach. Swallowing requires the coordination between multiple subsystems which makes it subject to impairment secondary to a variety of medical or surgically related conditions. Dysphagia refers to any swallowing disorder and is common in patients with head and neck cancer and neurological conditions such as stroke. Dysphagia affects nearly 9 million adults and causes death for more than 60,000 yearly in the US. In this research, we utilize advanced signal processing techniques with sensor technology and deep learning methods to develop a noninvasive and widely available tool for the evaluation and diagnosis of swallowing problems. We investigate the use of modern spectral estimation methods in addition to convolutional recurrent neural networks to demarcate and localize the important swallowing physiological events that contribute to airway protection solely based on signals collected from non-invasive sensors attached to the anterior neck. These events include the full swallowing activity, upper esophageal sphincter opening duration and maximal opening diameter, and aspiration. We believe that combining sensor technology and state of the art deep learning architectures specialized in time series analysis, will help achieve great advances for dysphagia detection and management in terms of non-invasiveness, portability, and availability. Like never before, such advances will enable patients to get continuous feedback about their swallowing out of standard clinical care setting which will extremely facilitate their daily activities and enhance the quality of their lives.


Social Networking:
Share |


Item Type: University of Pittsburgh ETD
Status: Unpublished
CreatorsEmailPitt UsernameORCID
Khalifa, Yassinyak33@pitt.eduyak330000-0001-8017-5959
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Committee ChairSejdic, Ervinesejdic@pitt.eduesejdic0000-0003-4987-8298
Committee MemberCoyle, Jamesjcoyle@pitt.edujcoyle
Committee MemberAkcakaya, Muratakcakaya@pitt.eduakcakaya
Committee MemberZhan, Liangliz119@pitt.eduliz119
Committee MemberDallal, Ahmedahd12@pitt.eduahd12
Date: 16 January 2022
Date Type: Publication
Defense Date: 2 November 2021
Approval Date: 16 January 2022
Submission Date: 26 October 2021
Access Restriction: No restriction; Release the ETD for access worldwide immediately.
Number of Pages: 192
Institution: University of Pittsburgh
Schools and Programs: Swanson School of Engineering > Electrical and Computer Engineering
Degree: PhD - Doctor of Philosophy
Thesis Type: Doctoral Dissertation
Refereed: Yes
Uncontrolled Keywords: Swallowing, Dysphagia, Cervical Auscultation, Deep Learning, Signal Processing, Recurrent Neural Networks, Convolutional Neural Networks
Date Deposited: 16 Jan 2022 17:23
Last Modified: 16 Jan 2022 17:23


Monthly Views for the past 3 years

Plum Analytics

Actions (login required)

View Item View Item