# Rapid Adaptation of SpO2 Estimation to Wearable Devices via Transfer Learning on Low-Sampling-Rate PPG

arXiv: https://arxiv.org/abs/2509.12515 · HTML: https://mlsanity.com/paper/2509.12515/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 0 of its 13 references that are in the map.
- First posted on arXiv 15 Sep 2025.

## Built on by (most cited first, 2 of 2)

- [SpO2 Predictor-Guided Stage-Wise Time-Frequency Reconstruction of Low-Quality Dual-Wavelength PPG for Oxygen Saturation Estimation](https://mlsanity.com/paper/2607.07996/) (2026)
- [Unit-Independent Low-Rate Wrist GSR Processing for Stress Detection Using Phasic nSCR Features](https://mlsanity.com/paper/2607.08007/) (2026)

## Abstract (arXiv)

Blood oxygen saturation (SpO2) is a vital marker for healthcare monitoring. Traditional SpO2 estimation methods often rely on complex clinical calibration, making them unsuitable for low-power, wearable applications. In this paper, we propose a transfer learning-based framework for the rapid adaptation of SpO2 estimation to energy-efficient wearable devices using low-sampling-rate (25Hz) dual-channel photoplethysmography (PPG). We first pretrain a bidirectional Long Short-Term Memory (BiLSTM) model with self-attention on a public clinical dataset, then fine-tune it using data collected from our wearable We-Be band and an FDA-approved reference pulse oximeter. Experimental results show that our approach achieves a mean absolute error (MAE) of 2.967% on the public dataset and 2.624% on the private dataset, significantly outperforming traditional calibration and non-transferred machine learning baselines. Moreover, using 25Hz PPG reduces power consumption by 40% compared to 100Hz, excluding baseline draw. Our method also attains an MAE of 3.284% in instantaneous SpO2 prediction, effectively capturing rapid fluctuations. These results demonstrate the rapid adaptation of accurate, low-power SpO2 monitoring on wearable devices without the need for clinical calibration.