Machine Learning-Enabled Hypertension Screening Through Acoustical Speech Analysis: Model Development and Validation
Bibliographic record
Abstract
Hypertension, referred to as the “silent killer” by the World Health Organization, affects over 35% of the global population. Early diagnosis and behavioural interventions have been shown to mitigate morbidity and mortality associated with this condition. However, conventional methods of measuring blood pressure and accordingly identifying hypertension, such as sphygmomanometry, require technical expertise and may not be readily accessible, particularly in remote or underserved areas. Automatic blood pressure measurement devices offer an alternative but are often inaccessible in certain populations. In this study, we propose a novel framework for detecting hypertension through acoustic analysis of speech. By recording speech across multiple sessions and analyzing its temporal and spectral characteristics, we aim to identify indicators of hypertension. We explore two thresholds for labeling individuals with hypertension: 1) systolic blood pressure (SBP)$\geq 135$mmHg or diastolic blood pressure (DBP)$\geq 85$mmHg and 2) SBP$\geq 140$mmHg or DBP$\geq 90$mmHg. Our study involved 245 participants, including 91 females. We developed predictive models for each gender and assessed their performance using leave-one-subject-out validation. For the first threshold, the balanced accuracy achieved was 84% for females and 77% for males. For the second threshold, the corresponding balanced accuracies were 63% for females and 86% for males. These results demonstrate the potential of utilizing speech-based representations for non-invasive screening of hypertension.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.007 | 0.010 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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 source (direct Gemma or distilled Codex), 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".