Key result
Multimodal signal fusion improves stress detection accuracy to ~76% over single-modality approaches.
Why the study?
Existing multimodal physiological stress detection studies largely separate stress only from baseline, amusement, or workload levels rather than everyday activities, and rely on subject-dependent data splits prone to data leakage, leaving subject-independent performance limited.
Does multimodal fusion of ECG, PPG, and BVP signals using a 1D-CNN improve subject-independent stress task classification accuracy compared to single-modality signals in the GalaxyPPG dataset?
Does multimodal fusion of ECG, PPG, and BVP signals using a 1D-CNN improve subject-independent stress task classification accuracy compared to single-modality signals in the GalaxyPPG dataset?
Combining ECG, PPG, and BVP physiological signals in a 1D-CNN framework improves subject-independent stress detection accuracy during everyday activities compared to single-signal modalities.
Multimodal fusion may aid wearable stress detection; leaves open prospective clinical validation and generalizability.
Prolonged stress can cause cardiovascular disease, depression, and anxiety. Reliable detection is therefore important. Existing studies using multimodal physiological methods have already reported accuracy, up to 99.7% however, two important aspects remain underexplored. First, most studies separate stress only from a neutral baseline, amusement, and different workload levels. Second, subject-independent performance remains limited because most studies evaluate their models using subject-dependent data splits that are prone to data leakage. This proposed study tries to address these gaps. Protocol-defined stress tasks are detected from everyday activities, not from any workload levels. The proposed model, a one-dimensional convolutional neural network (1D-CNN), is evaluated subject independently using leave-one-subject-out (LOSO) cross-validation. This proposed study was conducted on the GalaxyPPG dataset. The dataset includes ECG, PPG, and BVP signals. This enables multimodal fusion for stress detection during everyday activities. Single modalities achieve mean test accuracies of 62.19% for PPG, 67.74% for BVP, and 47.81% for ECG on LOSO. Fusion (ECG+PPG+BVP) achieves the mean test accuracy on LOSO evaluation, 76.47% }14.60%. To the best of the author’s knowledge, this modality’s combination (ECG+PPG+BVP) has not been explored in the literature for GalaxyPPG dataset.
No takes yet. Share an insight, caveat, or question.
Kumar et al. (2026) studied Stress during everyday activities. Multimodal physiological signal fusion (ECG + PPG + BVP) using a 1D-CNN model vs. Single physiological signal modalities (PPG alone, BVP alone, ECG alone) was evaluated on Mean test accuracy for subject-independent protocol-defined stress task classification on leave-one-subject-out (LOSO) cross-validation. Multimodal signal fusion (ECG+PPG+BVP) using a 1D-CNN achieved 76.47% mean test accuracy for subject-independent stress classification vs 67.74% for BVP, 62.19% for PPG, and 47.81% for ECG alone.
Synapse has enriched 5 related papers. Consider them for comparative context: