The use of methods for human motion capture in clinical environments is motivated by the need to understand normal and pathologic movement. Restoration of function is the goal of most treatments for musculoskeletal injuries or degenerative conditions. Motion analysis provides a unique opportunity to obtain objective functional information that may enhance diagnosis, treatment evaluation, and design of new interventions to address orthopaedic and sports medicine problems. In a research setting, motion capture has been used for the evaluation of the pathomechanics related to musculoskeletal injuries or diseases and the evaluation of new rehabilitive and preventative interventions. Results from motion analysis studies have provided valuable insight for joint replacement design, anterior cruciate ligament surgery technique, and the evaluation of arthroscopic hip surgery, and they have informed the role of neuromuscular mechanics in osteoarthritis initiation and progression. However, the clinical potential of these methods have not yet been realized.
A large number of parameters can be used to describe the kinematic (i.e., the motion of joints or body segments) and kinetic (i.e., the relationship between motion and applied forces) aspects of healthy or pathologic motion; however, a few key parameters that are particularly relevant for understanding the pathomechanics of human movement have been identified. Three categories (time-distance, kinematics, and kinetics) of functional outcome metrics can provide insightful objective functional information. This chapter reviews the basic techniques for motion analysis data collection and analysis. In addition, the clinical applications of these techniques and the potential benefits of functional outcome measures for the evaluation of pathologic movement and treatment outcomes are reviewed.
Functional Outcome Metrics
Simple time-distance metrics of the gait cycle over a range of walking speeds can be a useful indicator of the relative state of lower extremity pathologies. Steady-state walking speed is typically measured over an 8-m walkway at the patient’s normal or most comfortable walking pace, slightly faster than normal, and slightly slower than normal. One gait cycle is considered to be the period between foot to ipsilateral foot contact. The stance phase for walking constitutes approximately 62% of the entire gait cycle and has two double support phases separated by a single leg stance phase. The swing phase constitutes the remaining 38% of the gait cycles and denotes the time when the foot is in the air. Stride length is measured from initial foot contact to ipsilateral foot contact, whereas step length is measured from initial foot contact to initial contact of the contralateral foot.
Clinical Relevance of Time-Distance Measurements
Patients with knee disabilities tend to walk at slower self-selected walking speeds, and the step-length–walking speed relationship differs compared with healthy persons. Patients with knee disabilities typically have shorter stride lengths and higher step cadence compared with healthy persons when they are walking at a similar speeds. Clinical improvement after treatment is consistent with changes in these gait parameters. However, the objective nature of gait parameters for assessment of clinical status or function can be used to assess subtle improvements in status with limited influence of the placebo effects.
Walking speed and step length are sensitive to slight changes in the severity of symptoms related to treatment. In a recent randomized, single-blind washout, double-blind treatment, double-dummy crossover trial using three treatment arms—placebo, opioid (oxycodone), and nonsteroidal antiinflammatory drug (NSAID; celecoxib)—in patients with medial compartment knee osteoarthritis (OA), it was found that walking speed increased in response to both active treatments compared with the placebo treatment. However, in this study the increase in step length differed between the two treatments (NSAID vs. opioid), suggesting that this metric is also sensitive to differences in the mechanism of action of the treatment.
Three-Dimensional Joint Kinematics
The patterns of joint movement in healthy young persons over the stride period are quite invariant, which remains true over a range of walking speeds and different cadences. The sagittal plane joint angles for the hip, knee, and ankle for walking at a self-selected normal speed are illustrated in Figure 8-1 . Age, joint injury, pain and obesity are just a few of the known factors that influence these patterns of movement.
Methodology for Kinematic Data Collection
This section will address the fundamental techniques of clinical motion analysis and the important assumptions and limitations of the commonly applied techniques. More detailed reviews of the mathematics underlying this procedure can be found elsewhere.
A marker-based stereophotogrammetric technique is typically used to provide kinematic information about human body movement. Two types of motion capture systems exist. One type of system uses retroreflective (passive) spherical markers and high-speed active infrared (IR) cameras. These cameras emit and sense IR light. The second type of system uses light-emitting diode markers (active) and IR signal-sensing cameras. The techniques for calculating rigid body motion and relative segment movements are the same for both systems.
The primary assumption in the field of motion analysis is that the limb segments act as rigid bodies. In reality, skin and soft tissue overlying the bone do move and deform, resulting in a known source of error. However, this simplifying assumption permits the use of simple rigid body dynamics equations. Many techniques have been developed to minimize the effect of soft tissue movement on the calculation of rigid body motion.
Rigid Body Dynamics
Beyond the basic principles of linear and angular motion, a few additional concepts are needed to describe the position and orientation of a rigid body in space and for calculation of relative motion between two bodies.
Global Coordinate System
To describe the position and orientation of a rigid body in space, a stationary Cartesian coordinate system first must be defined. Such a system is referred to in the literature as a global, laboratory, inertial, or Newtonian reference frame or coordinate system . In this text we will use the term global coordinate system (GCS) . To describe the position of the rigid body in space, we first need to define or describe the location of a specific point with that body ( Fig. 8-2 ). Because this body has a shape and volume, the orientation relative to the GCS also must be described, which means that we need to define a second Cartesian coordinate system that has its origin and axes fixed in the segment of interest (see Fig. 8-2 ), which is referred to as the local or segment coordinate system (SCS).
Segment Coordinate System Definitions
For the appropriate, clinically relevant interpretation of kinematic measurements, a precise and physically meaningful definition of the SCS is needed. Thus the definition or selection of an SCS is an important step for the reconstruction of the motion of the limb segment or body of interest. With the Cartesian SCS embedded in the distal and proximal segments of a joint, the orientation of the distal segment relative to the proximal segment is defined by a 3 × 3 rotation matrix. The joint angles can be determined from this rotation matrix with Cardon/Euler angles. Euler angles are a defined as a set of three finite rotations assumed to take place in sequence to achieve the final orientation from an initial reference orientation. Thus a set of three independent angles is obtained by an ordered sequence of rotations about the axes of one of the SCSs (proximal) to find the orientation of the second (distal) SCS. The magnitude of the independent angles is influenced by the how the axes are assigned to the ordered sequence. Differences in the definition of the axes or errors in the definition of the axes will also result in changes in axis orientation about which the rotation and translations are calculated, which will affect the magnitude of the measured joint angles for any given activity.
To define an SCS on the segment, the three-dimensional position of at least three noncolinear (i.e., not in one line) markers are needed. Current methods to define the origin (joint centers) and orientation of the axes for the SCS are based primarily on palpation of anatomic landmarks, precise placement of skin markers on these landmarks, and established distance correction factors. An example for the knee is provided in Figure 8-3 .
Markers placed at the joint margin that typically are used to define the SCSs can be tracked/reconstructed during movement to find the position and orientation of the limb segment and determine joint angles. However, significant technical challenges exist with use of these markers, particularly for the thigh and shank segments, including substantial skin motion artifacts and marker obstruction during movement from contralateral limb or upper exterminates during movement. Therefore additional markers are often fixed to the segment to facilitate data collection ( Fig. 8-4 ). These additional markers are used to create a second coordinate system fixed to the limb segments. The orientation and position of this moving coordinate system relative to the SCS is determined while the subject is standing in the anatomic position. The relative position measured during this neutral trial can then be compared with the measured positions during movement to determine the segment movement. The least squares method, singular value decomposition method, or point cluster technique can be used for three or more markers.
Overview of Joint Coordinate Systems
The description of joint motions is simply the characterization of how the relative positions of two segment coordinate systems change with time. As a general rule, one needs to have caution when comparing results between different researchers and techniques because the choice of coordinate system influences the magnitude of rotations and displacements. The International Society of Biomechanics has proposed methods for defining the segment coordinate systems and for calculating and reporting joint motion in clinically relevant terms to enable and facilitate comparison of results in the literature. First proposed by Grood and Suntay in 1983 as a standard for the knee, the Joint Coordinate System has been expanded to other joints, including the ankle, hip, and spine.
To create a joint coordinate system (JCS), first, as previously described, a coordinate system is established for each of segments adjacent to the joint of interest. A common origin of both axis systems is used as the point of reference for the calculation of linear translation occurring in the joint. A JCS is established at the joint based on the two SCSs. Two of the JCS axes are body fixed and the third is a “floating” axis and is not a fixed segment. Joint motion (rotation and translation) is determined on the basis of the JCS. Although this method provides a temporally independent method for describing the motion in clinically relevant terms, some ambiguity remains because there is flexibility in choosing the anatomic reference points that are used to define the body fixed axis. A standardization that has been proposed for JCS use suggests that the first body fixed axis, chosen as the “axis of flexion,” is oriented in the mediolateral direction; the second axis, the longitudinal axis, which is chosen as the axis of the SCS, is oriented lengthwise along the segment; and the third axis is calculated as the cross-product of the longitudinal axis and the axis of flexion.
Special Considerations: Six Degree of Freedom Knee Kinematics
The in vivo function of the knee is an area of particular interest given its predisposition to injury and degeneration. The functional motion of the knee during daily movements involves a complex set of coupled rotations and translations. The complex motion of the knee depends on the activity and passive constraints provided by the anatomy of the knee and the muscle firing patterns. Measurement of the small secondary rotations and translation is challenging because of the poor signal-to-noise ratio. However, these measures are important to consider when studying the role of knee ligament injury or reconstruction, the primary function of which is to constrain the joint in rotation and anterior-posterior (AP) translation on changes in in vivo joint function.
A new method described as the point cluster technique has been designed to minimize the effect of random skin movement on the measurement of knee joint motions and reduce the ambiguity present in the JCS by prescribing a sequence of rotation and extracting projection angles from the rotation submatrix. The additional benefit of this method is that it avoids tightly strapping fixtures to the limb segments or requiring the subject to move in a limited fixed field of view as required by fluoroscopic, radiographic, or magnetic resonance imaging techniques.
The point cluster technique method provides accurate measurements of six degree of freedom movement at the knee. This method uses clusters of nine and seven markers distributed evenly on the thigh and shank, respectively, to predict the movements of the underlying femur and tibia ( Fig. 8-4 ). The center of mass and the inertia tensor of the cluster of points are calculated. The eigenvectors and eigenvalues of these inertia tensors are principal moments of inertia and the principal axes of the cluster (moving coordinate system). These establish the transformation between the cluster coordinate system and the GCS. During a static reference trial, markers placed on bony landmarks in addition to the clusters of markers are used to establish the tibial, femoral, and pelvic SCSs and the rotation matrix relating the cluster coordinate system to the SCS. The orientation of the cluster marker axes are determined by the eigenvectors of the marker clusters at each instant of the gait cycle. The motion of the knee is determined by relating the motion of the marker clusters to the SCSs. Translations at the knee joint are expressed as the displacement of the origin of the thigh coordinate system relative to the shank system. AP motion is the thigh displacement projected on the AP axis of the shank, whereas medial-lateral and inferior-superior motion are determined from the projection of the corresponding shank axis. The use of extra, redundant markers minimizes the effect of skin movement artifacts from any given marker during dynamic movements.
Kinematic Functional Outcomes Metrics for Knee Pathologies
Aging and joint injury are two factors that can contribute to changes in the primary (flexion) motion of the knee joint. Disruption of the anterior cruciate ligament (ACL) has been associated with changes as a result of (internal/external) knee joint rotations and translations (AP). After injury, the normal external rotation and anterior translation of the tibia that occurs as the knee extends in terminal swing is reduced ( Fig. 8-5 ). This rotational offset compared with normal is maintained throughout the stance phase of the walking cycle. Reconstruction of the ACL can lead to improved AP stability and a restoration of normal AP translation at the knee during walking. However, in recent studies it has been shown that reconstruction of the ACL may not restore the internal/external rotation to match that of the contralateral knee. Initial evidence suggests that the altered joint kinematics during locomotion after an ACL injury and reconstruction may be a significant contributing factor to degeneration of joint tissue by shifting the loading during normal movement to areas of the joint that are not conditioned for those loads.
Special Considerations for Hip Joint Kinematics
For most biomechanical applications, the hip joint is treated as a ball and socket joint with the center of rotation at the center of the hip. Kinematics of the hip joint are described as motions of the thigh coordinate system relative to the pelvic coordinate system. Locating the hip center of rotation presents a greater challenge than for most other joints because only the greater trochanter can be “easily” palpated, and substantial soft tissue overlies this landmark and surrounds the joint. Errors in establishment of the hip joint center location may result in error in the maximum hip adduction moment during walking of up to 30%. Methods for estimating the hip joint center location require data describing the pelvis and/or thigh segment or require data obtained with the subject performing a dynamic movement in a functional range of hip motion.
A variety of methods using landmarks have been proposed, including a method by Andriacchi and Strickland that identifies the mediolateral hip joint center coordinate 2.5 cm inferior to the point halfway between the anterior superior iliac spine (ASIS) and the pubic tubercle, with the AP and proximal-distal coordinate defined by the greater trochanter. An alternate method locates the hip joint center as a percentage of the distance between right and left. Specifically, this method defines the hip joint center as 30% of the inter-ASIS distance distal to the ASIS (vertical position), 26% of the inter-ASIS distance lateral to the pelvis center of mass (medial-lateral position), and aligned with the superior point of the iliac crest (vertical position). In a comparison study it was shown that these methods are in general very similar with respect to the hip joint center locations.
Functional joint center (FJC) methods have been shown to produce more reproducible results compared with landmark methods in patients with excellent range of hip motion. All FJC methods identify a joint center as the functional center of rotation between two adjacent methods. For the hip, markers placed on the thigh and pelvis are tracked while the thigh is actively or passively rotated relative to the pelvis. The algorithms proposed for the FJC location estimation using reconstructed marker position are based on the assumption that the distance between thigh markers and the center of rotation does not vary (i.e., rigid body motion). Details on these algorithms can be found elsewhere.
Kinematic Functional Outcome Metrics for Hip Pathologies
Abnormalities in the hip flexion-extension motion have been identified in patients with end-stage hip OA. These abnormalities present as a reduction in the sagittal plane range of motion and a discontinuity in the normally smooth trajectory of the sagittal plane hip motion at midstance. Similar patterns have also been reported in young athletic patients prior to hip resurfacing surgery for femoroacetabular impingement ( Fig. 8-6 ), which suggests that hip motion during walking may be an early marker of hip OA risk. Sagittal plane hip motion in walking and stair climbing has also been studied as a potential objective functional outcome for evaluating arthroscopic hip resurfacing procedures. In these studies it was found that after hip resurfacing procedures, the normally smooth flexion-extension and internal-external rotation motions are restored in walking but not in stair climbing. These results have potential implications for understanding the pathomechanics of femoroacetabular impingement and may provide an objective functional outcome metric for evaluating different treatment pathways.