Automatic discovery of resource-restricted Convolutional Neural Network topologies for myoelectric pattern recognition

• Published in: Computers in biology and medicine Vol. 120; p. 103723
• Main Authors: Olsson, Alexander E., Björkman, Anders, Antfolk, Christian
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.05.2020; Elsevier Limited
• Subjects: Algorithms; Artificial neural networks; Bioinformatics (Computational Biology); Bioinformatik (beräkningsbiologi); Collating; Computer and Information Science; Computer and Information Sciences; Computer applications; Convolutional neural networks; Data- och informationsvetenskap (Datateknik); Deep learning; Discriminant analysis; Electromyography; Embedded systems; Evolutionary algorithms; Information retrieval; Interfaces; Internal Medicine; Iterative methods; Machine learning; Medical research; Model selection; Muscle-computer interfaces; Muscles; Myoelectric control; Myoelectric pattern recognition; Myoelectricity; Natural Sciences; Naturvetenskap; Network topologies; Neural networks; Other; Pattern recognition; Problem solving; Prostheses; Unstructured data; Myoelectric control; Deep learning; Model selection; Muscle-computer interfaces; Machine learning; Electromyography; Convolutional neural networks; Myoelectric pattern recognition
• ISSN: 0010-4825; 1879-0534; 1879-0534
• DOI: 10.1016/j.compbiomed.2020.103723

Convolutional Neural Networks (CNNs) have been subject to extensive attention in the pattern recognition literature due to unprecedented performance in tasks of information extraction from unstructured data. Whereas available methods for supervised training of a CNN with a given network topology are well-defined with rigorous theoretical justification, procedures for the initial selection of topology are currently not. Work incorporating selection of the CNN topology has instead substantially been guided by the domain-specific expertise of the creator(s), followed by iterative improvement via empirical evaluation. This limitation of methodology is restricting in the pursuit of naturally controlled muscle-computer interfaces, where CNNs have been identified as a promising research avenue but effective topology selection heuristics are lacking. With the goal of mitigating ambiguities in topology selection, this paper presents a systematic approach wherein we apply a novel evolutionary algorithm to search a space of candidate topologies. Furthermore, we constrain the search-space by excluding topologies with excessive inference-time computational complexity, making the obtained results implementable in embedded systems. In contrast to manual topology design, our algorithm requires the user to only specify a relatively small set of intuitive hyperparameters. To validate our approach, we use it in order to create topologies for myoelectric pattern recognition via movement decoding of surface electromyography signals. By collating offline classification accuracies obtained from experiments on a collection of publicly available databases, we demonstrate that our method generates computationally lightweight topologies with performance comparable to those of available alternatives. •A meta-learning algorithm is introduced to generate CNN topologies suitable for EMG gesture classification.•Only classifiers with feasible time complexity and memory footprint are considered.•The method is demonstrated by automatically generating a novel CNN topology.•The resulting network is evaluated on a set of publicly available EMG databases.•Classification performance was comparable to that of costlier CNNs introduced previously.