The Case for tinyML in Healthcare: CNNs for Real-Time On-Edge Blood Pressure Estimation
Bibliographic record
Abstract
More than half of the deaths from cardiovascular diseases (CVDs) can be eliminated if high Blood Pressure (BP) is brought under control. Towards this target, continuous BP monitoring is a must. Unfortunately, such continuous monitoring is hindered by the inconvenience of the traditional cuff-based method. Convolutional Neural Networks (CNNs) offer a promising alternative to address such problems. Nonetheless, existing CNN-based solutions rely on server-like infrastructure with huge computation and memory capabilities. This entails these solutions impractical with several security, privacy, reliability, and latency concerns. The contribution of this paper is twofold as follows. First, the paper contributes to the general field of tinyML by proposing novel techniques that enable the fitting of popular CNNs into extremely-constrained edge devices with limited computation, memory, and power budget. Namely, the paper successfully manages to fit the following five popular CNNs into tiny edge devices: AlexNet, LeNet, SqueezeNet, ResNet, and MobileNet. This enables us to run inference completely on the edge without dependency on connectivity or cloud infrastructure. The proposed techniques use a combination of novel architecture modifications, pruning, and quantization methods. Second, utilizing this stepping stone, the paper proposes a tinyML-based solution to enable accurate and continuous BP estimation using only photo-plethysmogram (PPG) signals. We conduct an extensive evaluation using thousands of real Intensive Care Unit (ICU) patient data and several tiny edge devices and all the five aforementioned CNNs. Results show that the proposed solutions offer comparable accuracy to server-based solutions, and also meet the Association for the Advancement of Medical Instrumentation (AAMI) and the British Hypertension Society (BHS) standards.
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 | 0.001 | 0.006 |
| 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 | 0.001 |
| Scholarly communication | 0.001 | 0.003 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| 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".