## Motion Analysis Research

#### Introduction

Motion analysis methods have been widely used to measure and model human movements. Traditionally motion analysis has been only qualitative, e.g., interpretation of video recorded during a gait. With help of force plates and kinematics originated from photogrammetry motion analysis can nowadays also be considered as a quantitative method. Applications of motion analysis consist of diagnosis of neurophysiological diseases, monitoring of rehabilitation after surgery and studying forces inside a knee joint in case of arthopathic patients, for example.

#### Ground reaction force analysis

With the force plates ground reaction forces and moments can be measured in a three-dimensional coordinate system. Furthermore, with these measurements it is possible to obtain derived quantities, such as dynamical center of pressure (COP), resultant forces, and power during movements. The ground reaction force analysis consists of determining several common parameters, for example the temporal location of maxima of forces, loading rate, angles and impulses of acceleration and deceleration, and power and work over the contact.

#### Correlation analysis

One of our interests is to utilize mobile data loggers with the motion analysis in the field. Our approach applies principal component analysis (PCA) with multisignal data acquired at the motion analysis laboratory to model the correlations between various signals. By using this information our aim is to estimate and predict, with a portable device, the same essential parameters which could be measured with help of full arsenal of instruments at the motion laboratory. Such a mobile device could make use of accelerometers, goniometers and electromyography.

#### Biomechanical modeling

Biomechanical modeling and photogrammetry have been widely used as a basis of motion analysis. Three-dimensional locations of reflective markers are tracked by using high speed cameras. Those markers are typically placed on the joint points of body so that dynamical stick figure could be constructed.

Our research will concern appending 3D-marker data to a biomechanical model in which masses of segments are taken into account. Further ahead, our goal is to add other quantities to the model, such as ground reaction forces.

The biomechanical model can be used to obtain several mechanical parameters of body and body parts, for example joint angles, velocities of body parts, accelerations, forces, and moments. The better model the more accurate estimates will be obtained, e.g. about the forces inside knee joint.

#### Analysis of other signal types

Besides photogrammetry and force plates we are also applying accelerometry and some other signal types such as EMG and goniometry. Preprocessing tools have to be developed in order to remove acceleration component originating from gravitation. For further research, inverse dynamics by using accelerometry and goniometry will be explored.