Keshi He

Papers

Total Citations

110

H-Index

About

Keshi He is a leading researcher in biomedical signal processing and human-machine interaction, with a focus on advancing non-invasive sensing technologies for hand and finger motion recognition. His seminal work, "Ultrasound-Based Sensing Models for Finger Motion Classification" (2017, 110 citations), addresses the limitations of traditional surface electromyography (sEMG) by leveraging ultrasound imaging to capture the complex spatial and temporal coordination of forearm muscles and tendons. This pioneering approach significantly improves classification accuracy for intricate finger movements, offering a robust alternative for prosthetic control and rehabilitation systems. He’s contributions have been widely recognized, with his research bridging the gap between biomechanics and machine learning to enable more natural and intuitive interfaces. By tackling the inherent challenges of sEMG—such as signal variability and limited specificity—his work has laid a foundation for next-generation wearable devices. With over 110 citations on this key paper alone, He’s impact is evident in both academic and applied contexts, inspiring further exploration into ultrasound-based sensing for fine motor control. His achievements underscore a commitment to transforming how we decode human intent through advanced sensing modalities.