This poster presents my Laidlaw research project on the computational efficiency of surface electromyography (sEMG) motor-unit decomposition. I compared classical blind source separation and autoencoder-based approaches to investigate how algorithmic budget affects motor-unit recovery and processing time, with the aim of informing more energy-efficient wearable neural interfaces.
Hi everyone! I’m Donna, and I have just completed my second year studying Molecular Bioengineering at Imperial College London.
I’m interested in how research can move from scientific ideas into real-world healthcare impact. I enjoy learning about the biology behind disease and technology, but I’m also curious about what happens after a discovery is made: how it can be translated responsibly into something useful, accessible, and meaningful.
This summer, my Laidlaw research project will focus on wearable muscle-signal interfaces for assistive control after spinal cord injury. More broadly, I’m excited by healthcare technologies that are not only technically impressive, but also practical enough to make a difference in people’s daily lives.
Through Laidlaw, I hope to better understand the role leadership plays in helping research move beyond the lab, from shaping ideas thoughtfully to connecting them with people and communities who may benefit from them.
I’m looking forward to learning from people across the Laidlaw community, especially those working in bioengineering, healthcare innovation, neural engineering, or protein de novo design. Please feel free to reach out. I’d be very happy to connect.
