This is the authors' abstract. We don't add key points for this paper.
This study details the design and implementation of a real-time early detection system for stress levels among competitive mobile gamers of Mobile Legends: Bang Bang (MLBB), with a specific comparative focus on children and adolescents. Employing an experimental quantitative approach, multimodal physiological data comprising Heart Rate (HR), Heart Rate Variability (HRV), and Galvanic Skin Response (GSR) were acquired using an Internet of Things (IoT)-based wearable device powered by an ESP32 microcontroller. The captured time-series data were preprocessed and classified into three stress levels (low, moderate, and high) using a hybrid CNN-LSTM deep learning architecture evaluated with an 80:20 cross-subject validation scheme. The model achieved a training accuracy of 84.62% and a validation accuracy of 61.54%, with macro-averaged precision, recall, and F1-score of 66.67%, 60.00%, and 57.78%, respectively. The findings demonstrate that the spatial feature extraction of CNN combined with the temporal processing of LSTM effectively recognizes biosignal patterns, while revealing significant variations in autonomic response characteristics between children and adolescents attributed to differences in psychophysiological regulatory maturity
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Ridho et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: