Ivan Vujaklija

Effects of Spatial and Signal-Imposed Noises on Motor Unit Decomposition

This study evaluates how different types of noise affect motor unit decomposition using HD-sEMG, finding that global noise severely impacts results while localized noise has minimal effects.

Sensitivity of Non-Invasive Motor-Unit-Based Gesture Recognition to Signal Degradation

This study evaluates how signal degradation affects motor-unit-based gesture recognition using HMI, finding that LDA and DNN perform differently under noise and spatial perturbations.

Impact of Noise on Deep Learning-Based Pseudo-Online Gesture Recognition with High-Density EMG

This study evaluates how different noises affect a DNN-based gesture recognition system using HD-sEMG, finding that feature type and noise type influence performance degradation.

Adaptive HD-sEMG decomposition: Towards robust real-time decoding of neural drive

This study develops an adaptive real-time algorithm for decoding motor unit activity from HD-sEMG and shows it outperforms static methods under changing contraction conditions.

Non-invasive real-time access to the output of the spinal cord via a wrist wearable interface

This paper presents a non-invasive wrist wearable interface that decodes spinal motoneuron activity to predict finger movements with high accuracy in real-time.