Review of Deep Learning Model for Hand Gesture Recognition Using EMG Signals

This review presents a comprehensive analysis of deep learning models used for hand gesture recognition based on electromyography (EMG) signals, a key technology for human–computer interaction, prosthetic control, and rehabilitation systems. The paper examines the complete processing pipeline, including EMG signal acquisition, preprocessing, feature extraction, and classification using advanced deep learning architectures such as convolutional neural networks (CNNs), recurrent neural networks (RNNs), long short-term memory (LSTM), and hybrid models. Emphasis is placed on the comparative performance of these models in terms of accuracy, robustness, real-time capability, and adaptability to inter-subject variability. The review also highlights publicly available EMG datasets, evaluation metrics, and major challenges such as signal noise, electrode placement sensitivity, and limited generalization across users. Finally, the study outlines open research directions, including lightweight models for embedded systems, transfer learning, and multimodal fusion, to guide future advancements in EMGbased hand gesture recognition systems.