Evaluation of an AI-Based Voice Biomarker Tool to Detect Signals Consistent With Moderate to Severe Depression
Bibliographic record
Abstract
PURPOSE: Mental health screening is recommended by the US Preventive Services Task Force for all patients in areas where treatment options are available. Still, it is estimated that only 4% of primary care patients are screened for depression. The goal of this study was to evaluate the efficacy of machine learning technology (Kintsugi Voice, v1, Kintsugi Mindful Wellness, Inc) to detect and analyze voice biomarkers consistent with moderate to severe depression, potentially allowing for greater compliance with this critical primary care public health need. METHODS: We performed a cross-sectional study from February 1, 2021 to July 31, 2022 to examine ≥25 seconds of free-form speech content from English-speaking samples captured from 14,898 unique adults in the United States and Canada. Participants were recruited via social media, provided informed consent, and their voice biomarker results were compared with a self-reported Patient Health Questionnaire-9 (PHQ-9) at a cut-off score of 10 (moderate to severe depression). RESULTS: From as few as 25 seconds of free-form speech, machine learning technology was able to detect vocal characteristics consistent with an increased PHQ-9 ≥10, with a sensitivity of 71.3 (95% CI, 69.0-73.5) and a specificity of 73.5 (95% CI, 71.5-75.5). CONCLUSIONS: Machine learning has potential utility in helping clinicians screen patients for moderate to severe depression. Further research is needed to measure the effectiveness of machine learning vocal detection and analysis technology in clinical deployment.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.005 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".