Illustrated portrait
MASc researcher
Liam McPherson
Liam applies modern sequence models to inertial-measurement data, connecting body motion, sensor signals, temporal structure, and interpretable system decisions.
- Program
- MASc
- Supervision
- Leonard MacEachern
- Co-supervision
- Andrew Spiers, Mechanical Engineering
- Last updated
- 22 July 2026
Place in MOSAIC
The connected research tiles
Research overview
Liam researches language-model and sequence-model interpretation of inertial-measurement-unit time-series data for human motion. The work is jointly supervised by Leonard MacEachern and Andrew Spiers in Mechanical Engineering.
IMUs produce continuous, multidimensional streams from accelerometers, gyroscopes, and sometimes magnetometers. The research asks how modern sequence models can recognize and interpret nuanced motion without losing the relationship between a model’s conclusion and the physical measurements that produced it.
Place in MOSAIC
Liam’s work renews the group’s long-running wearable and body-signal program with modern machine intelligence. It connects:
- Physical human movement
- Multimodal inertial sensing
- Time-series representation
- Sequence modelling and interpretation
- Validation across people, activities, and changing conditions
Current direction
The profile will grow with the motion vocabulary, dataset and instrumentation, modelling choices, training and validation protocol, interpretability strategy, and project results.