Hypertension Screening Using Acoustic Analysis and Machine Learning of Random Speech Samples: A Feasibility Study
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Introduction: Hypertension is the leading risk factor for cardiovascular disorders. Early detection and initiation of treatment have been identified as the most effective ways to reduce the burden of hypertension. The most common method for detecting hypertension is blood pressure measurement, typically performed with cuff-based devices, where systolic pressure (SBP) and diastolic pressure (DBP) are measured through Korotkoff sounds. Although this method is accurate and non-invasive, it requires technical expertise and is often inaccessible in rural and remote areas. In this study, we investigated the feasibility of using overt speech (random speech corpora) through multiple short recordings for hypertension screening based on two hypertension guidelines: (1) SBP ≥135 mm Hg OR DBP ≥85 mm Hg, and (2) SBP ≥140 mm Hg OR DBP ≥90 mm Hg. Methods: We incorporated speech recordings from 573 participants (197 women) with diverse ages and body mass index and extracted temporal, spectral, and nonlinear acoustic features through three different frameworks, all of which are based on classical and boosted machine learning models. The models were evaluated using a leave-one-subject-out cross-validation scheme. Results: Our proposed pipeline achieved a balanced accuracy (BACC) of 61% for males and 70% for females under the relaxed criterion (SBP ≥135 OR DBP ≥85), and a BACC of 71% for males and 78% for females under the stricter European Society of Hypertension (ESH) guidelines (SBP ≥140 OR DBP ≥90). Conclusion: These results demonstrate the potential of employing overt speech alongside acoustic analysis for hypertension screening.
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.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.002 |
| 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 it