Hyeong-Yeong Park

Papers

Total Citations

H-Index

About

Hyeong-Yeong Park is a leading researcher at the intersection of brain-computer interfaces (BCIs) and neurorehabilitation, with a primary focus on decoding motor imagery (MI) for stroke recovery. Their work addresses the critical challenge of cross-subject variability in EEG classification, where they pioneered multi-task heterogeneous ensemble learning frameworks that significantly improve the generalizability of BCI systems across diverse patient populations. Park's most impactful study (2024, 22 citations) demonstrates how robot-assisted motor training integrated with MI-based BCIs can offer real-time assistance to stroke patients facing movement challenges, bridging the gap between neural decoding and practical rehabilitation tools. Their latest contribution, DeepSMR (2025), introduces subject-dependent multi-feature refinement within deep convolutional networks, enabling high-complexity MI decoding with unprecedented precision. By tackling the fundamental limitations of traditional single-subject models, Park's research accelerates the clinical translation of non-invasive BCIs, offering scalable solutions for personalized neurorehabilitation. Their work is particularly notable for its emphasis on heterogeneous ensemble learning—a methodological innovation that combines multiple neural network architectures to robustly handle the noisy, non-stationary nature of EEG signals. Park's contributions are shaping the next generation of adaptive, patient-specific BCI systems for motor recovery.