Key result
Synthetic-data LSTM framework denoises optical waveforms, cutting heart rate error ~75% versus standard filters.
Why the study?
Optical monitoring of cardiac pulsations using NIRS, PPG, and DCS is frequently hindered by motion artifacts and noise.
A synthetic-data-trained LSTM model effectively denoises optical pulsatile signals and accurately extracts heart rate, outperforming standard filtering methods.
May enable motion-robust optical cardiac monitoring; leaves open prospective clinical validation before adoption.
Optical monitoring of cardiac pulsations using near-infrared spectroscopy (NIRS), photoplethysmography (PPG), and diffuse correlation spectroscopy (DCS) is often hindered by motion artifacts and noise. We introduce a synthetic-data-driven framework using a long short-term memory (LSTM) network to trace and denoise pulsatile optical waveforms without reliance on annotated clinical datasets. Physiologically realistic pulsatile signals are generated, corrupted with parameterized artifacts, and used to train the LSTM model. Applied to experimental NIRS, PPG, and DCS signals, the model recovered beat-to-beat morphology more effectively than widely used wavelet and temporal derivative distribution repair (TDDR) filters. Heart rate (HR) extraction from LSTM-processed signals closely matched ECG-derived measurements (mean absolute error = 0.59 bpm, root mean square error = 0.74 bpm). This flexible approach shows potential for rapid adaptation across various devices and noise conditions.
No takes yet. Share an insight, caveat, or question.
Wu et al. (2025) studied Healthy (optical signal monitoring). Synthetic-data-driven LSTM framework vs. Raw signals, DWT, and TDDR filters was evaluated on Mean absolute error of heart rate extraction compared to ECG (bpm). The synthetic-data-driven LSTM framework effectively denoised pulsatile optical waveforms, yielding heart rate measurements that closely matched ECG-derived ground truth with a mean absolute error of 0.59 bpm.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: