Key result
Lightweight LSTM-Transformer model achieves ~99% accuracy for biometric identification from wrist-worn PPG signals.
Why the study?
Wrist-worn photoplethysmography offers an unobtrusive biometric modality, but state-of-the-art models rarely address the memory and runtime constraints of embedded hardware.
Effect estimate: EER 0.19%
A lightweight LSTM-Transformer model enables highly accurate biometric identification from wrist-worn PPG signals while significantly reducing model size for edge deployment.
May enable on-device PPG biometrics in wearables; leaves open validation in diverse real-world cohorts.
Wrist-worn photoplethysmography (PPG) offers an unobtrusive biometric modality, yet state-of-the-art models rarely address the memory and runtime constraints of embedded hardware. We conduct a systematic benchmark of CNN, LSTM, Transformer, and hybrid architectures on three public datasets, UBFC-Phys, ECSMP, and EMOGNITION, encompassing 166 unique subjects and over 26,000 PPG windows. Models are evaluated under a unified preprocessing pipeline and then converted to TensorFlow Lite to gauge deployability. A lightweight LSTM-Transformer architecture achieves 9 9 . 1 4 % accuracy and a global Equal Error Rate of 0 . 1 9 %. After TFLite conversion, accuracy drops by less than $0.1 %$ while model size shrinks from 730 kB to under 100 kB. Inference latency on a representative resource-constrained CPU ranged between 13-50 s per 10s window using single-threaded TFLite; we therefore outline quantization and hardware-aware optimization as future work.
No takes yet. Share an insight, caveat, or question.
Kagnici et al. (2025) studied Biometric Identification (n=166). Lightweight LSTM-Transformer architecture vs. CNN, LSTM, Transformer, and hybrid architectures was evaluated on Accuracy and global Equal Error Rate (EER 0.19%). A lightweight LSTM-Transformer architecture for biometric identification from wrist-worn PPG signals achieved 99.14% accuracy and a global Equal Error Rate of 0.19%.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: