Although current gold standard gait analysis tools, such as stereophotogrammetry and force plates, provide high quality kinematic data, these systems also have many pitfalls such as their cost, their setup time and the fact they are confined to the camera defined collection space. On the contrary, inertial sensors are inexpensive, easy to use when combined with processing software and are not confined to a small collection space [3,4]. Instead of gait analysis being confined to complex biomechanics laboratories, as with traditional gait analysis tools, gait analysis with inertial sensors could take place during regular clinical check-ups, in the home and ubiquitously as people go about their daily lives [5]. Such inertial sensor use could drastically increase the amount of gait data that clinicians and researchers can obtain.

The subsequent expansion of available databanks would also allow for enhanced gait rehabilitation programs and more ecologically valid research [6].Despite the current depth of research in the inertial sensor gait area, the sensors are not used extensively outside of the laboratory environment, save for example as a step counter. This may be partly due to the fact that inertial sensor gait analysis use has been primarily confined to replication of traditional gait analysis metrics; such as temporal, spatial and kinematic data. There is therefore a need for research to look toward using inertial sensor data in innovative ways to provide new clinically meaningful metrics.

Despite the availability of the theoretical framework in which inertial sensor data can be integrated to obtain position, it is a well-documented phenomenon that this process results in significant errors [7,8]. It is hypothesized that the use of the raw inertial sensor signal would yield more valid and reliable results as opposed to the error ridden integration to position.Previous inertial sensor gait research on Cilengitide patient populations has used temporal or spatial outputs from inertial sensors to identify pathological gait patterns, however the pathologies studied have been characterized by large gait deviations [9,10]. Observed differences ranged from 0.5 m in step length for Parkinson patients [10] to 0.6 m in step length for stroke patients [9]. Some pathological conditions, such as anterior crucitate ligament reconstruction or early on-set knee osteoarthritis (OA), may result in gait deviations that do not alter temporal or spatial parameters, but do alter joint angular kinematics and kinetics and can consequently lead to the premature development of degenerative joint disease such as OA.