GPT-PPG: A GPT-based Foundation Model for Photoplethysmography Signals
This work addresses the need for advanced signal processing in healthcare by providing a domain-specific foundation model for PPG analysis, though it is incremental as it adapts an existing GPT architecture to a new data type.
The study tackled the problem of analyzing photoplethysmography (PPG) signals by introducing a GPT-based foundation model, achieving performance comparable to or surpassing state-of-the-art methods in tasks like atrial fibrillation detection and effectively performing generative tasks such as signal denoising without fine-tuning.
This study introduces a novel application of a Generative Pre-trained Transformer (GPT) model tailored for photoplethysmography (PPG) signals, serving as a foundation model for various downstream tasks. Adapting the standard GPT architecture to suit the continuous characteristics of PPG signals, our approach demonstrates promising results. Our models are pre-trained on our extensive dataset that contains more than 200 million 30s PPG samples. We explored different supervised fine-tuning techniques to adapt our model to downstream tasks, resulting in performance comparable to or surpassing current state-of-the-art (SOTA) methods in tasks like atrial fibrillation detection. A standout feature of our GPT model is its inherent capability to perform generative tasks such as signal denoising effectively, without the need for further fine-tuning. This success is attributed to the generative nature of the GPT framework.