EMaGer: A Wearable Full-Circumference HD-EMG Sensor and Data Augmentation Method for Robust Hand Gesture Recognition.

This work presents EMaGer, a new 360° 64-channel high-density electromyography (HD-EMG) bracelet combined with an original data augmentation method for improved robustness in gesture recognition. By leveraging homogeneous electrode density and powerful deep learning techniques, the sensor is capable of rotation invariance around the arm axis, thus increasing gesture recognition robustness to electrode movement and inter-session evaluation. The system is made of a 4x16 electrode array covering the full circumference of the limb, and uses a sampling frequency of 1 kHz and a 16-bit resolution. The sensor's uniform and adjustable geometry paired with an array barrel shifting data augmentation (ABSDA) technique allows a convolutional neural network to maintain a 76.98% inter-session classification accuracy for a 6 gestures dataset, from a baseline intra-session accuracy of 93.75%. High inter-session classification accuracy decreases the training burden for users of EMG control systems such as myoelectric prostheses by minimizing calibration requirements. The same methods applied with different state-of-the-art sensors are demonstrated to be less effective. Thus, this work evidences the importance of co-designing the EMG sensor system with the gesture inference algorithms to leverage synergistic properties and solve state-of-the-art challenges.Clinical relevance- This paper establishes a method that alleviates clinical manipulations in setting up and calibrating myoelectric prosthetic devices.

Novel Wearable HD-EMG Sensor With Shift-Robust Gesture Recognition Using Deep Learning.

In this work, we present a hardware-software solution to improve the robustness of hand gesture recognition to confounding factors in myoelectric control. The solution includes a novel, full-circumference, flexible, 64-channel high-density electromyography (HD-EMG) sensor called EMaGer. The stretchable, wearable sensor adapts to different forearm sizes while maintaining uniform electrode density around the limb. Leveraging this uniformity, we propose novel array barrel-shifting data augmentation (ABSDA) approach used with a convolutional neural network (CNN), and an anti-aliased CNN (AA-CNN), that provides shift invariance around the limb for improved classification robustness to electrode movement, forearm orientation, and inter-session variability. Signals are sampled from a 4×16 HD-EMG array of electrodes at a frequency of 1 kHz and 16-bit resolution. Using data from 12 non-amputated participants, the approach is tested in response to sensor rotation, forearm rotation, and inter-session scenarios. The proposed ABSDA-CNN method improves inter-session accuracy by 25.67% on average across users for 6 gesture classes compared to conventional CNN classification. A comparison with other devices shows that this benefit is enabled by the unique design of the EMaGer array. The AA-CNN yields improvements of up to 63.05% accuracy over non-augmented methods when tested with electrode displacements ranging from -45 ° to +45 ° around the limb. Overall, this article demonstrates the benefits of co-designing sensor systems, processing methods, and inference algorithms to leverage synergistic and interdependent properties to solve state-of-the-art problems.