 This paper proposes a new approach to decomposing electromyogram, EMG, signals into individual motor units, MUs. The authors use a segment, Wise Convolution Kernel Compensation, CKC, algorithm to identify MUs from multiple motor neuron populations. This allows them to decode more MUs than previously possible, increasing the potential applications of EMG decomposition for neural decoding during multiple motor tasks. This article was authored by Chen Chen, Sherhan Ma, Yang Yu, and others.