The observation of human joint movement

Abstract

Recently Freeman e Pinskerova stated: “It may be hoped that the next few years will see a resolution to the question: how does the healthy and diseased knee move in the activities of daily life?”. To contribute to their auspice and to pursue the relevant scientific objectives, joint kinematics must be estimated with a higher resolution than presently available. To this purpose, the following issues must be tackled: 1) optimal estimation of the instantaneous pose of bones in a laboratory system of reference during the execution a physical exercise using the instantaneous position or pose of skin markers; 2) optimal estimation of the subject-specific bone morphology and determination of the relevant anatomical axes.

The first issue entails the development of an estimator of bone pose which incorporates a mathematical model of the displacement of the markers relative to the underlying bone identified using a non-invasive, subject-specific, approach. At present, when subject-specific bioimages are not available, anatomical data are obtained through a low resolution anatomical calibration which entails the determination, through stereophotogrammetry, of the position of the few bony landmarks that are identifiable through palpation. This procedure, although universally used, provides totally unsatisfactory results. Thus, the second issue calls for the development of a procedure which allows for an estimate of subject-specific bone digital models and their registration with the movement data. In this way, the manual identification of anatomical landmarks may be avoided, and, since the entire bone is available instead of a few anatomical landmarks, the identification of anatomical axes would enjoy more flexibility and results would be more accurate and repeatable.

This paper presents the state of the art of knowledge regarding the above-mentioned issues.