Innovative research methods in biomechanics
The development and dissemination of analysis tools ranging from basic data processing for undergraduate students to advanced vector field analysis based on statistical parametric mapping techniques.
Advanced statistics
Vector continuum schematic, depicting a mean two-muscle EMG waveform in blue along with inter-muscle dependence (EMG1-EMG2 covariance) and time-dependence (TIME-EMG smoothness).
Biomechanical data range from discrete 0D scalars to complex 3D vectors (and beyond). Traditional analyses of biomechanical data typically reduces the complexity of the dataset by extracting “key scalars” to analyse with standard statistical analysis techniques. The problem with this approach is that the selection of the “key scalars” is inherently biased unless a hypothesis directly pertained to that key scalar at that instance in time.
Our research involves the analysis of n-D biomechanical data using statistical parametric mapping. We use traditional general linear models, for example t-test, ANOVA, regression to analyse biomechanical data but with the maintenance of time within the analysis fundamentally differentiating it from traditional analyses.
Primary contact
- Dr Mark Robinson
Artificial neural networks
A set of unique methods to quantify deviation of movement from normality to support clinical decision-making using artificial intelligence. Software tools developed in our project are available as free downloads.
Characteristics of abnormal gait can be captured by quantifying the three dimensional joint angles, moments and powers of the lower limbs but the complexity of the resultant high dimensional data space makes data interpretation prone to bias. Self-organising neural networks can be used to map complex gait data onto a two dimensional topographic map, thereby improving the efficiency of decision-making in clinical gait analysis.
We have developed the Movement Deviation Profile which shows the deviation of an individual’s movement from normality in a single curve and a single number.
Our staff
- Prof Gabor Barton
- Dr Malcolm Hawken
- Dr Mark Robinson
- Jacob Beesley