Few-Shot PPG Signal Generation via Guided Diffusion Models
Recent advancements in deep learning for predicting arterial blood pressure (ABP) have prominently featured photoplethysmography (PPG) signals. Notably, PPG signals exhibit significant variability due to differences in measurement environments, alongside stark disparities in the distribution of coll...
Gespeichert in:
Veröffentlicht in: | IEEE sensors journal 2024-10, Vol.24 (20), p.32792-32800 |
---|---|
Hauptverfasser: | , , , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | Recent advancements in deep learning for predicting arterial blood pressure (ABP) have prominently featured photoplethysmography (PPG) signals. Notably, PPG signals exhibit significant variability due to differences in measurement environments, alongside stark disparities in the distribution of collected signal data among different labels. To address these challenges, this study introduces a bi-guided diffusion (BG-Diff) model designed to generate PPG signals with expected features of ABP within a few-shot setting for each label group. We propose a guided diffusion model architecture that simultaneously considers both the determinant group condition and the continuous label condition for each group in a few-shot setting. To the best of our knowledge, this is the first study to use a diffusion model for generating PPG signals with a limited dataset. Initially, we categorized them into four groups based on systolic blood pressure (SBP) and diastolic blood pressure (DBP) values: Hypo, Normal, Prehyper, and Hyper2. In each group, we sample an equal number of data points according to the few-shot setting and then generate appropriate PPG signals for each group through guidance. In addition, our study proposes a postprocessing technique to address the limitations of generative models in few-shot settings, consistently boosting performance across various methods, such as training from scratch, transfer learning, and linear probing (LP). When benchmarked, our methodology demonstrated performance improvements across all datasets, including BCG, PPGBP, and Sensors. We confirmed data quality by comparing training, generated, and actual data. We analyzed error cases, morphology features, and t-SNE distribution to highlight the role of synthetic data in enhancing performance. |
---|---|
ISSN: | 1530-437X 1558-1748 |
DOI: | 10.1109/JSEN.2024.3451453 |