Digital Voice Analysis as a Biomarker of Acromegaly
Authors' conclusion
Digital voice analysis can identify patients with acromegaly from short voice recordings with high accuracy. Patients with acromegaly experience more voice disorders than matched controls.
Study context
- Population
- Swedish patients with acromegaly (13% biochemically active, 10% newly diagnosed) and matched controls, Swedish university hospitals151 patients · 139 controls
- Comparison
- Machine learning digital voice analysis vs experienced endocrinologists for identifying acromegaly
- Outcome
- Classification of speaker as acromegaly or control (ROC AUC)
Methodological detailsmatched controls, machine learning, multicentre
- Matched controls
- Machine learning
- Multicentre
Performance
As reported in the abstract
Identification of acromegaly (ROC AUC)
| Test or model | AUC |
|---|---|
| Machine learning model | 0.84 |
| Endocrinologists | 0.69 |
AUC
Machine learning model0.84
Endocrinologists0.69
Further findings
Clinically significant voice handicap (VHI ≥20)
Median
| Outcome | Acromegaly | Matched controls | P value |
|---|---|---|---|
| Self-reported voice problems (VHI)More pronounced in acromegaly | 6 | 2 | P < 0.01 |
Original publication
Digital Voice Analysis as a Biomarker of Acromegaly. J Clin Endocrinol Metab. Published 4 October 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →