Voice disorders (hoarseness) are a common condition manifesting in almost 9% of the population. The causes of hoarseness can be related to common respiratory diseases and/or to vocal fatigue caused by vocal overload, however it can also be related to functional, behavioral, neurologic factors and both benign and malignant laryngeal tumors. Hoarseness can also be an early symptom of laryngeal cancer. Timely, qualified, and complex otolaryngology examination may increase diagnostic and treatment effectiveness and reduce health care expenditures in the evaluation and management of patients with diverse laryngeal/voice disorders including laryngeal carcinoma. Ensuring an early visit to an otolaryngologist, screening healthy or at-risk populations for early signs of a voice abnormality, seems to be useful and promising, because many laryngeal/voice disorders may be most successfully treated if detected early.
The planned research result would be embedded in an automated acoustic analysis-based voice-screening tool based on machine learning algorithms (in addition combining voice quality and review questionnaires) that could serve as the potential approach to help subjects (self-screening), primary care physicians and other public health care services identify the patients who require otolaryngological referral, thereby improving diagnostics and management of laryngeal/voice disorder patients. The main goal of automated pathological voice detection systems is to categorize any input voice as either normal or pathological. The research results would be original and important in practical application and scientific worldwide context.
Project funding:
KTU Research Fund
Project results:
1. Two publications are planned to be submitted together with co-authors (at least one publication when submitted to a Q1-Q3 Clarivate Analytics WOS-level journal).
2. The artificial intelligence algorithms created and adapted during the project will be illustrated by the created Android app, which is planned to be placed in the Google Play store, thus promoting the name of both universities. The prepared solution will meet the level of technological readiness TRL8 (Manufactured trial batch of products).
3. A stationary version of the device for evaluating voice disorders is planned. The completed solution will meet the level of technological readiness TRL6 (Manufactured Prototype).
Period of project implementation: 2024-04-11 - 2024-12-27
Project partners: Lithuanian University of Health Sciences
