Motion tracking and modelling of the forefoot for clinical gait analysis

Thesis

Multi-segment foot models (MSFMs) are often used in clinical gait analysis (CGA) for the diagnosis and treatment planning for patients with foot deformities. Thus far, close to 40 MSFMs have been proposed, and each differs in terms of segmentation, marker placement protocol, tracking technology, and joint motion. By representing individual segments as a series of linked rigid body segments, the computation and interpretation of foot biomechanics can be greatly simplified. While this approach may be acceptable for lower-limb segments such as the thigh, shank, or hindfoot, the assumption that this is also valid for the forefoot has been met with criticism. Out of all the segments, there is least agreement on how the forefoot should be represented in an MSFM. In general, Whether or not the flexibility of the forefoot is clinically meaningful is controversial.

Therefore, the aims of this thesis were to investigate the most appropriate way of segmenting the forefoot based on a rigidity criteria and to develop a clinically relevant and practical forefoot model based on the recommended segmentation. It was hypothesised that minimising the violation of rigid body assumption in the proposed novel MSFMs will produce the most relevant forefoot representation that is both practical and useful for clinical gait analysis.

This study first validated the use of sigma_RBE to quantify MSFM segment rigidity through a Monte-Carlo based global sensitivity analysis, which showed that for a typical forefoot segment the sigma_RBE represented 86.7% of the deformation magnitude, with a linear relationship that can be predicted using a simple regression model. Experimental data were used to compute rigidity and track deformation of the Oxford Foot Model segments and the metatarsal markers (attached to the heads and bases of all metatarsals) over the full gait cycle. These data were used to determine appropriate segmentation and marker-placement strategies in the design of novel MSFMs.

Several novel forefoot models were proposed and tested through comparison of visual and descriptive statistics of kinematic output and repeatability analysis. A single novel forefoot model was selected based on practicality and kinematic advantages, and piloted in five individual clinical case studies. The research has contributed to the evidence that the forefoot should not be treated as a single segment, and a new forefoot model suitable for CGA has been proposed.

Authors: Chan, PH

• Type of award: DPhil
• Level of award: Doctoral
• Awarding institution: University of Oxford
• Language: English
• Keywords: Biomedical engineering; Forefoot; Gait analysis; Human motion analysis; Multi-segment foot model; Biomechanics; Engineering
• Subjects: Biomedical engineering; Biomechanics; Engineering