Vowel Segmentation for Classification of Chronic Obstructive Pulmonary Disease Using Machine Learning

Blekinge Institute of Technology

Status: Active, not recruiting

Conditions: Chronic Obstructive Pulmonary Disease

Treatments: Other: COPD

Study type: Observational

Funder types: Other

Identifiers

NCT06160674

BTH-6.1.1-0169-2023

Details and patient eligibility

About

This work aims to evaluate whether the segmentation of vowel recordings collected from patients diagnosed with COPD and healthy control groups can increase the classification precision of machine learning techniques.

Full description

Voice data and sociodemographic data on gender and age will be collected through the "VoiceDiganostic" application from the company Voice Diagnostic. Collected vowel recordings will be segmented and tested to determine whether some segments contain more information for the discrimination of COPD from healthy control groups.

Each segment will be transformed into mathematical vocal measures called voice features. A dataset consisting of voice features in conjunction with demographics and health data will be constructed for each segment which in turn will be evaluated for classification performance using several machine learning algorithms.

Descriptive statistical analysis will be held on attributes containing information on input data and gained outcomes from ML algorithms. The achieved results will be presented in the form of summary tables and graphs.

Enrollment: 68 patients

Sex: All

Ages: 18+ years old

Volunteers: Accepts Healthy Volunteers

Inclusion criteria

• being 18 years old and older.

Exclusion criteria

• being under 18 years old and older.

Trial design

68 participants in 2 patient groups

COPD

Description:

30 COPD participants, 16 Female and 14 Male.

Treatment:

Other: COPD

HC

Description:

38 HC participants, 20 Female and 18 Male.

Treatment:

Other: COPD