Optimizing Voice Sample Quantity and Recording Settings for the Prediction of Type 2 Diabetes Mellitus: Retrospective Study
Abstract BackgroundThe use of acoustic biomarkers derived from speech signals is a promising non-invasive technique for diagnosing type 2 diabetes mellitus (T2DM). Despite its potential, there remains a critical gap in knowledge regarding the optimal number of voice recordings...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
JMIR Publications
2025-06-01
|
| Series: | JMIR Biomedical Engineering |
| Online Access: | https://biomedeng.jmir.org/2025/1/e64357 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849421265586094080 |
|---|---|
| author | Atousa Assadi Jessica Oreskovic Jaycee Kaufman Yan Fossat |
| author_facet | Atousa Assadi Jessica Oreskovic Jaycee Kaufman Yan Fossat |
| author_sort | Atousa Assadi |
| collection | DOAJ |
| description |
Abstract
BackgroundThe use of acoustic biomarkers derived from speech signals is a promising non-invasive technique for diagnosing type 2 diabetes mellitus (T2DM). Despite its potential, there remains a critical gap in knowledge regarding the optimal number of voice recordings and recording schedule necessary to achieve effective diagnostic accuracy.
ObjectiveThis study aimed to determine the optimal number of voice samples and the ideal recording schedule (frequency and timing), required to maintain the T2DM diagnostic efficacy while reducing patient burden.
MethodsWe analyzed voice recordings from 78 adults (22 women), including 39 individuals diagnosed with T2DM. Participants had a mean (SD) age of 45.26 (10.63) years and mean (SD) BMI of 28.07 (4.59) kg/m². In total, 5035 voice recordings were collected, with a mean (SD) of 4.91 (1.45) recordings per day; higher adherence was observed among women (5.13 [1.38] vs 4.82 [1.46] in men). We evaluated the diagnostic accuracy of a previously developed voice-based model under different recording conditions. Segmented linear regression analysis was used to assess model accuracy across varying numbers of voice recordings, and the Kendall tau correlation was used to measure the relationship between recording settings and accuracy. A significance threshold of P
ResultsOur results showed that including up to 6 voice recordings notably improved the model accuracy for T2DM compared to using only one recording, with accuracy increasing from 59.61 to 65.02 for men and from 65.55 to 69.43 for women. Additionally, the day on which voice recordings were collected did not significantly affect model accuracy (P
ConclusionsThis study underscores the optimal voice recording settings to reduce patient burden while maintaining diagnostic efficacy. |
| format | Article |
| id | doaj-art-e2c6b6c9fd924338a79605398f90e342 |
| institution | Kabale University |
| issn | 2561-3278 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | JMIR Publications |
| record_format | Article |
| series | JMIR Biomedical Engineering |
| spelling | doaj-art-e2c6b6c9fd924338a79605398f90e3422025-08-20T03:31:30ZengJMIR PublicationsJMIR Biomedical Engineering2561-32782025-06-0110e64357e6435710.2196/64357Optimizing Voice Sample Quantity and Recording Settings for the Prediction of Type 2 Diabetes Mellitus: Retrospective StudyAtousa Assadihttp://orcid.org/0000-0002-1634-7821Jessica Oreskovichttp://orcid.org/0009-0007-9778-8728Jaycee Kaufmanhttp://orcid.org/0000-0001-8183-0206Yan Fossathttp://orcid.org/0000-0002-1271-2633 Abstract BackgroundThe use of acoustic biomarkers derived from speech signals is a promising non-invasive technique for diagnosing type 2 diabetes mellitus (T2DM). Despite its potential, there remains a critical gap in knowledge regarding the optimal number of voice recordings and recording schedule necessary to achieve effective diagnostic accuracy. ObjectiveThis study aimed to determine the optimal number of voice samples and the ideal recording schedule (frequency and timing), required to maintain the T2DM diagnostic efficacy while reducing patient burden. MethodsWe analyzed voice recordings from 78 adults (22 women), including 39 individuals diagnosed with T2DM. Participants had a mean (SD) age of 45.26 (10.63) years and mean (SD) BMI of 28.07 (4.59) kg/m². In total, 5035 voice recordings were collected, with a mean (SD) of 4.91 (1.45) recordings per day; higher adherence was observed among women (5.13 [1.38] vs 4.82 [1.46] in men). We evaluated the diagnostic accuracy of a previously developed voice-based model under different recording conditions. Segmented linear regression analysis was used to assess model accuracy across varying numbers of voice recordings, and the Kendall tau correlation was used to measure the relationship between recording settings and accuracy. A significance threshold of P ResultsOur results showed that including up to 6 voice recordings notably improved the model accuracy for T2DM compared to using only one recording, with accuracy increasing from 59.61 to 65.02 for men and from 65.55 to 69.43 for women. Additionally, the day on which voice recordings were collected did not significantly affect model accuracy (P ConclusionsThis study underscores the optimal voice recording settings to reduce patient burden while maintaining diagnostic efficacy.https://biomedeng.jmir.org/2025/1/e64357 |
| spellingShingle | Atousa Assadi Jessica Oreskovic Jaycee Kaufman Yan Fossat Optimizing Voice Sample Quantity and Recording Settings for the Prediction of Type 2 Diabetes Mellitus: Retrospective Study JMIR Biomedical Engineering |
| title | Optimizing Voice Sample Quantity and Recording Settings for the Prediction of Type 2 Diabetes Mellitus: Retrospective Study |
| title_full | Optimizing Voice Sample Quantity and Recording Settings for the Prediction of Type 2 Diabetes Mellitus: Retrospective Study |
| title_fullStr | Optimizing Voice Sample Quantity and Recording Settings for the Prediction of Type 2 Diabetes Mellitus: Retrospective Study |
| title_full_unstemmed | Optimizing Voice Sample Quantity and Recording Settings for the Prediction of Type 2 Diabetes Mellitus: Retrospective Study |
| title_short | Optimizing Voice Sample Quantity and Recording Settings for the Prediction of Type 2 Diabetes Mellitus: Retrospective Study |
| title_sort | optimizing voice sample quantity and recording settings for the prediction of type 2 diabetes mellitus retrospective study |
| url | https://biomedeng.jmir.org/2025/1/e64357 |
| work_keys_str_mv | AT atousaassadi optimizingvoicesamplequantityandrecordingsettingsforthepredictionoftype2diabetesmellitusretrospectivestudy AT jessicaoreskovic optimizingvoicesamplequantityandrecordingsettingsforthepredictionoftype2diabetesmellitusretrospectivestudy AT jayceekaufman optimizingvoicesamplequantityandrecordingsettingsforthepredictionoftype2diabetesmellitusretrospectivestudy AT yanfossat optimizingvoicesamplequantityandrecordingsettingsforthepredictionoftype2diabetesmellitusretrospectivestudy |