Screening of Voice Pathologies: Identifying the Predictive Value of Voice Acoustic Parameters for Common Voice Pathologies.
Lady Catherine Cantor-Cutiva, Sai Aishwarya Ramani, Patrick R Walden and 1 others
PMID 38143203WHAT IT FOUND
A combination of five acoustic measures correctly identified 8 out of 10 people with voice disorders, but only 4 out of 10 healthy people.
This high sensitivity makes it useful for screening, but the low specificity means many normal voices would be falsely flagged.
Key findings
01The combined acoustic measure (PC1) had high sensitivity but low specificity, correctly classifying 8 out of 10 participants with voice disorders and only 4 out of 10 speakers with healthy voices.
02Individual parameters like CPPS, shimmer, and HNR showed moderate predictive capabilities with areas under the curve of 0.74, 0.72, and 0.69, respectively.
03The first principal component, which included shimmer, jitter, HNR, CPPS, and PPE, explained 34.7% of the variance and was the best predictor among the components.
STILL TO COME
How it was doneWhat they foundWhat it means for SLPs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
The study used a secondary database, meaning the researchers did not control how the data was collected or the specific diagnoses of the participants. Some participants in the original database lacked clear diagnoses and were excluded, potentially introducing selection bias. The sample size for specific pathologies like Parkinson's Disease and Adductor Spasmodic Dysphonia was small, limiting the generalizability of findings for those groups. The low specificity of the best predictive model (PC1) means it produces a high rate of false positives in healthy individuals.
Declared interests
No conflict of interest exists.
The easy way to misread this
Do not interpret the high sensitivity of the five-parameter model as evidence that it is a definitive diagnostic tool. Because its specificity was low (only 4 out of 10 healthy voices were correctly classified), relying on it alone would result in many healthy patients being incorrectly identified as having a voice disorder.
Summarised by AI from the full paper, without a clinician reviewing it. Check it against the source before it changes what you do. Read it on PubMed →