Shriram is a postdoctoral fellow who joined the BLINC lab in 2025. He develops and applies machine learning models to improve the performance and reliability of myoelectric devices and other human-machine interfaces. His work focuses on bridging the gap between algorithmic capabilities and the practical demands of real-world applications.

During his PhD in Electrical Engineering at University of New Brunswick’s Institute of Biomedical Engineering, Shriram addressed the classification errors that arise when users transition between muscle contractions in myoelectric control. To mitigate these errors, he trained deep temporal models, such as LSTMs and Transformers, on data containing these transitional states to improve system robustness. He also holds an MScE from University of New Brunswick, where his research involved improving and validating an EMG simulation tool.

Shriram’s research interests lie at the intersection of applied machine learning and biomedical engineering. His primary technical interests include deep learning, representation learning, and self-supervised learning for robust pattern recognition. He is also exploring how methods from explainable AI (XAI) can be used to help a person better understand, interpret, and adapt to the behavior of the system they are using.

In his spare time, Shriram enjoys long walks, cycling, hiking, going to the gym, playing board games, playing video games, and watching TV shows. He also enjoys reading about latest PC hardware, software, and AI related content.

Selected Publications:

• S. T. P. Raghu, D. T. MacIsaac, and E. J. Scheme, ‘Self-supervised representation learning with continuous training data improves the feel and performance of myoelectric control’, Computers in Biology and Medicine, vol. 197, p. 111029, Oct. 2025, doi: 10.1016/j.compbiomed.2025.111029.
• S. T. P. Raghu, H. E. Williams, and E. Scheme, ‘Efficient Multi-Positioned Training for Regression-Based Myoelectric Control: Exploring Transformer Models with Transfer Learning’, in 2025 International Conference On Rehabilitation Robotics (ICORR), May 2025, pp. 1597–1603. doi: 10.1109/ICORR66766.2025.11063067.
• S. T. P. Raghu, D. T. MacIsaac, and E. J. Scheme, ‘Self-supervised learning via VICReg enables training of EMG pattern recognition using continuous data with unclear labels’, Computers in Biology and Medicine, vol. 185, p. 109479, Feb. 2025, doi: 10.1016/j.compbiomed.2024.109479.
• S. Tam, S. T. P. Raghu, E. Buteau, E. Scheme, M. Boukadoum, A. Campeau-Lecours, and B. Gosselin., ‘Towards Robust and Interpretable EMG-based Hand Gesture Recognition using Deep Metric Meta Learning’, Apr. 17, 2024, arXiv: arXiv:2404.15360. doi: 10.48550/arXiv.2404.15360.
• S. T. P. Raghu, D. MacIsaac, and E. Scheme, ‘Decision-Change Informed Rejection Improves Robustness in Pattern Recognition-Based Myoelectric Control’, IEEE Journal of Biomedical and Health Informatics, vol. 27, no. 12, pp. 6051–6061, Dec. 2023, doi: 10.1109/JBHI.2023.3316599.
• S. T. P. Raghu, D. T. MacIsaac, and A. D. C. Chan, ‘Automated Biomedical Signal Quality Assessment of Electromyograms: Current Challenges and Future Prospects’, IEEE Instrumentation & Measurement Magazine, vol. 25, no. 1, pp. 12–19, Feb. 2022, doi: 10.1109/MIM.2022.9693438.