Kianoush Nazarpour

This paper reports a pilot study and computational model showing low-frequency tibial neuromodulation alters bladder activity and urge perception, supporting a frequency-dependent mechanism.

This paper explores the scalability of random forest models in myoelectric control, showing they can be reduced in size with minimal accuracy loss and benefit from higher input dimensions.

This paper introduces a self-calibrating random forest model for myoelectric control that improves performance over time with minimal data and user adaptation.

The paper presents an open-source neurodynamic model of the bladder that improves upon existing models by incorporating a circadian rhythm and neural interventions, showing high correlation with animal data.

This paper develops a one-shot random forest model calibration method for hand gesture decoding using minimal user-specific EMG data, achieving higher accuracy than benchmark models.

This paper shows that reducing motor variability through training improves myoelectric control robustness across different limb positions.

This study characterizes the recovery dynamics of neural function after high-frequency alternating current (HFAC) nerve block, showing that recovery time and completeness depend on signal frequency and amplitude.