Key result
In-ear PPG system detects cardiac arrhythmias with an AUC of ~0.81.
Why the study?
Space missions require reliable, safe, and continuous physiological monitoring systems due to microgravity effects on cardiovascular function and heart rhythm stability.
Does a smart in-ear PPG system with a hybrid deep learning model accurately detect arrhythmias compared to standard ECG?
Cross-Sectional (n=70)
No
Does a smart in-ear PPG system with a hybrid deep learning model accurately detect arrhythmias compared to standard ECG?
Effect estimate: AUC 0.8078 ± 0.0492
A novel in-ear PPG system utilizing a hybrid deep learning model demonstrates an AUC of 0.80 for arrhythmia detection, indicating potential for continuous cardiovascular monitoring in extreme environments like space missions.
Supports in-ear PPG feasibility for arrhythmia detection; leaves open prospective ECG validation before clinical or space use.
Space missions require reliable and safe physiological monitoring systems, as well as capable of continuous operation, as microgravity affects cardiovascular function and heart rhythm stability. This study develops and validates an intelligent in-ear photoplethysmography (PPG) system for continuous cardiac rhythm monitoring and real-time arrhythmia detection. A total of 2,926 PPG signal segments were collected, covering both normal and arrhythmic conditions. Signals were recorded at 75 Hz and upsampled to 256 Hz, then preprocessed using a 0.56 Hz Butterworth filter, artifact detection, and segmentation into 2,560 data points (≈10 s). A hybrid ResNet–LSTM–Attention model was developed to extract morphological and temporal features, while performance was evaluated using 5-fold crossvalidation. The results demonstrated consistent and robust performance with accuracy 0.7138 ± 0.1115, AUC 0.8078 ± 0.0492, and F1-Score 0.7198 ± 0.0788, confirming the model’s capability to detect cardiac rhythm abnormalities from in-ear PPG signals. These findings highlight the potential of this technology as a continuous cardiovascular monitoring platform, suitable for integration into wearable devices and mobile health applications, as well as for use in extreme environments such as space, where conventional electrocardiography is less practical.
No takes yet. Share an insight, caveat, or question.
Azudin et al. (2026) conducted a cross-sectional in Arrhythmia (n=70). Smart In-Ear PPG-Based System with ResNet-LSTM-Attention model vs. Standard Electrocardiogram (ECG) was evaluated on Arrhythmia detection performance (AUC 0.8078 ± 0.0492). The intelligent in-ear photoplethysmography (PPG) system detected cardiac rhythm abnormalities with an accuracy of 0.7138 ± 0.1115 and an AUC of 0.8078 ± 0.0492.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: