The lab is headed by Dr. Andrew Hamilton-Wright. I use machine-learning and knowledge discovery techniques in clinically relevant biomedical applications based on time-series data. I am interested in the following types of problems:
- machine learning and pattern recognition on time-series data
- information content in templates found within time-series signals
- clinical decision support using electromyographic data
- visualization for risk analysis
- knowledge discovery and representation
My research focus is the application of computer based decision support tools for important human decision making problems, including the improvement of human understanding of electrophysiological data, in particular regarding postural sequences and electromyography (EMG).
- electromyographic data analysis and characterization
- assessment of stress and control using metabolic markers (electrodermal activity, heart rate, breathing rate and others)
- dynamic postural data risk characterization
- attentional state in driving tasks using indirect signals
Quantitative Electromyography
EMG data is time-series data acquired from contracting muscles. This data contains information about muscle control structure and disease state.
More information on quantitative electromyographic projects is available here.
Postural State Risk Analysis
Postural sequence analysis is based on observing the position of body segments (head, upper back, lower back, etc) and discovering patterns relating postural segment position with adverse postural events (pain, fatigue, etc). This project is based on collaboration with Dr. Nancy Black of the Université de Moncton.
More information on postural risk projects is available here.
Drive Lab
The University of Guelph DriveLab houses a full car driving simulator. Using this simulator, many driving scenarios can be safely created to study driver response. Currently, studies are ongoing based on predicting driver mental state through machine-learning derived markers found in car control state (throttle, lane drift, etc).
Software
Please see discussion on software on this page.
People
The lab is run by Dr. Andrew Hamilton-Wright. Please see my:
- Profile page in the School of Computer Science
- Bio page on this website
Current Students
John Beninger : M.Sc. Student
Kassy Raymond : M.Sc. Student