Deep Learning models, particularly Convolutional Neural Networks, were reported as the best performing model in 85 studies achieving classification accuracies exceeding 90% in many cases, outperforming traditional models like LDA and SVM, which also showed strong performances with many results exceeding 95%.
Systematic Review (n=301)
While deep learning is increasingly replacing traditional machine learning for upper-limb motion classification, there is a significant underutilization of hybrid EEG-EMG signal fusion.
This systematic review summarizes the application of artificial intelligence (AI) in classifying upper-limb motion using Electroencephalogram (EEG) and Electromyogram (EMG) signals, focusing on the field’s progression from Traditional Machine Learning (TML) to Deep Learning (DL) architectures. Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, a search of PubMed, IEEEXplore, and Web of Science yielded 301 eligible studies published up to June 2025. The results indicate a change from classical classifiers like Linear Discriminant Analysis (LDA) and Support Vector Machines (SVMs) toward DL approaches. While Convolutional Neural Networks (CNNs) remain the most frequently implemented, emerging architectures, including Long Short-Term Memory (LSTM) networks and Transformers, have demonstrated remarkable performance. Despite the rise of DL, classical models remain highly relevant due to their robustness and efficiency. This review also identifies a heavy reliance on EEG-only modalities (60%), with only 7% of studies utilizing hybrid EEG-EMG systems, representing a potential missed opportunity for signal fusion.
Bandes et al. (Thu,) conducted a systematic review in Human participants with non-pathological EEG and/or EMG signals used for upper-limb movement intent classification (n=301). Artificial intelligence models (Traditional Machine Learning and Deep Learning) for classification of upper-limb motion using EEG and/or EMG signals vs. Classical classifiers (e.g., LDA, SVM) versus Deep Learning models (e.g., CNN, LSTM, Transformers, Autoencoders) was evaluated on Classification accuracy of upper-limb motion intent from EEG and/or EMG signals. Deep Learning models, particularly Convolutional Neural Networks, were reported as the best performing model in 85 studies achieving classification accuracies exceeding 90% in many cases, outperforming traditional models like LDA and SVM, which also showed strong performances with many results exceeding 95%.