tinyCare: A tinyML-based Low-Cost Continuous Blood Pressure Estimation on the Extreme Edge
Bibliographic record
Abstract
We propose a solution that deploys Machine Learning (ML) techniques on resource-constrained edge devices (tinyMl)for the healthcare domain. In particular, we construct a complete end-to-end prototyped system that conducts ML inference with various ML techniques on microcontroller unit (MCU)-powered edge devices to predict blood-pressure-related vital metrics such as systolic (SBP), diastolic (DBP), and mean arterial (MAP) blood pressures using electrocardiogram (ECG) and photoplethysmogram (PPG) sensors. The proposed solution is trained and tested using over 500 hours of 12, 000 real intensive care unit data instances. Despite running on an extremely limited computation, power and memory budget, the proposed solution achieves comparable results to server-based state-of-the-art solutions. Furthermore, it meets the British Hypertension Society (BHS) standard for grade B (C in extremely-constrained devices). This is achieved by careful investigation of the correlation between a wide-set of ECG and PPG features and BP. Afterwards, we compress the ML inference models by only incorporating the minimal features that meet i) the edge constraints from one side, and ii) the standard's acceptable accuracy from the other side. Unlike existing solutions, the inference is entirely conducted on MCU-based edge devices without depending on any cloud-based infrastructure. Hence, the proposed solution improves robustness, accessibility, reliability, security, as well as data privacy.
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 | 0.000 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 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".