Balance impairment is common in stroke survivors. Balance is the ability to keep the body’s center of gravity over the base of supporting surface without losing stability. Maintaining a stable seated position requires good trunk control and is essential for many activities of daily living (ADL). The biomechanical parameters related to postural balance and control include quantifying the sway of the Center of Pressure (COP) and the asymmetry of forces and moments in static and dynamic tasks. The purpose of this study is to introduce such quantitative measurements into evaluation of postural balance and trunk control in sitting posture in subjects who have suffered a stroke.
Chair The seat height and seat depth of the chair are adjustable (Fig. 1). Under the seat, a 6-DOF load cell (JR3 Inc., Woodland, CA) was mounted to measure the forces and moments when the subject performed all tasks. Force Platforms A force platform (www.amtiweb.com) was placed under the 4 legs of the chair (Fig. 1). Forces in anterior, right, and inferior directions and the moments around these axes were recorded. Another such force platform was placed under the footrest. The locations of the chair and footrest were fixed on the upper surface of the force platforms.
Motion Capturing System A ViconPeak motion capturing system (8 Camera MCam2, Vicon Peak, Centennial, CO) was used to record the movement of the trunk, head, and upper limb of the seated individual (Fig. 1). Reflective markers were placed on the subject based on the standard placement of the VICON Plug-In-Gait (PIG) model.
Figure 1. Experimental setup. The experimental chair and the location of the 6-DOF load cell (JR3) are shown, as well as the force platforms under the chair,the footrest and the VICON reflective markers on the subject.
Static Sitting Trials These tasks were designed for assessing the subject’s ability to maintain an unsupported upright sitting posture (Fig. 2). Trials were tested with and without the footrest. In static sitting, the subject is asked to sit still in the chair for 1-minute. Three conditions were tested, "Eyes Open", "Eyes Closed" and "Target Staring" (staring at a target which is 150cm away in the front at the eye level). COP sway was the major physiological measure and it was characterized in the area (mm2), the COP quadrant distribution (%), and COP maximal displacement in coronal and sagittal planes. Fig. 3 shows two representative results of COP sway recorded in static sitting trial with "Eyes Open" from one stroke subject and one healthy control.
Figure 2. Static sitting trial. Subjects were asked to maintain unsupported seating posture for one minute either with eyes open, eyes closed, or while staring at a target. Trials were repeated both with and without footrest.
Figure 3. COP sway during static sitting. Representative data are from 1 healthy control and one stroke patient.
Dynamic Sitting Trials
Dynamic sitting tasks test the subject’s ability to move efficiently his/her center of mass in a voluntary reaching attempt and the ability to flex the trunk in forward and lateral directions, which are directly related to the balance and trunk control impairment. Trials were tested with and without the footrest. For forward reaching, the subject reaches out forward towards a distant target as much as possible with both arms stretched out (Fig. 4A). For lateral reaching (both to the left and to the right), the subject was asked to reach towards a distant target (Fig. 4B&C) Task efficiency (TE) was determined by calculating moment coupling from the moment data recorded using the 6-DOF load cell underneath the seat. It is the ratio of the squared value of the task moment (Mtask ) to the squared value of the magnitude of the resultant moment (ΣMi) during an active task. It measures how much of the effort is dedicated to the task, i.e. the efficiency. Since each of the designed tasks is a movement in a specific body anatomical plane, the moment recorded during the task should be expected in the corresponding plane with only small amount of coupled moment in other planes. 
Figure 4. Dynamic sitting trials. (A) Subject was asked to reach forward as much as possible towards distant target with both arms fully outstretched. (B) Subject was asked to reach towards distant target with non-affected side. (C) Subject was asked to reach towards distant target with affected side.
Static Trials (Fig. 5)
--The stroke subject had significantly larger sway area and an asymmetrical sway area distribution during upright sitting than that of the healthy controls. (Fig. 5). This finding is consistent with data in the literature (Harley et al. 2006) and suggested that individuals post stroke have impaired postural balance in a seated posture.
--When the footrest was used, stroke subjects swayed significantly larger area than the healthy controls.
--There was a significantly larger maximal coronal displacement in stroke subjects in all 3 sitting conditions. It implies that the impaired sitting balance may be mainly in the postural control of side-wise body movement.
Figure 5. COP area and displacement in static sitting. Data are given for "Target staring" condition in this plot
Dynamic Trials (Fig. 6)
--In all reaching tasks, the healthy controls achieved almost 100% of task efficiency and always were higher than that of stroke individuals.
--When the footrest was not used, this difference was significant only when the stroke individuals reached to their Non-Affected side.
--When the footrest was not used, the individuals with stroke showed significantly lower task efficiency in all 3 reaching directions.
Note: Kinematics of preliminary data are not reported 
Figure 6. Task efficiency (%) in dynamic reaching tasks in seated individuals.
Lateral Trunk Performance Apparatus In order to correlate trunk performance findings with previously validated clinical tests, such as the Trunk Impairment Scale* (TIS), an apparatus has been constructed to help normalize and quantify lateral trunk performance (Fig. 7). Using this device, the subject will be asked to touch his/her elbow to the surface that has an XSensor (www.xsensor.com) pressure pad attached to it (Fig. 8). The Kinematics (Vicon), Kinetics (JR3 and Force Platform), as well as the coordinates of where the subject was able to make contact with the surface (XSensor) will be analyzed to better understand the subject’s ability, or lack thereof, to properly extend and flex opposite sides of trunk.
Figure 7. Lateral Trunk Performance Apparatus. The apparatus seen here will be used for subjects to touch his/her elbow to each surface on each side of the seat. The pressure sensing sheets on each surface record the location of contact.
Figure 8. Lateral trunk performance trials. Demonstration of control subject touching elbow to surface. The goal is to extend and flex opposite lateral trunk muscles in order to achieve an isolated lateral rotation in the coronal plane as much as possible. Also shown is an example of kinematics data seen on Vicon workstation.
Trunk Co-ordination Trials Trunk co-ordination tasks are designed to assess the subject’s ability to axially rotate his/her trunk in both left and right directions. It consists of 2 parts, upper and lower trunk. For upper trunk, the subject is asked to rotate his/her upper trunk in the transverse plane to bring forward each of his/her shoulders in turn at a comfortable speed (Fig. 9). For lower trunk, the subject is asked to rotate his/her lower trunk and pelvis in the transverse plane to bring forward each of his/her knees in turn at a comfortable speed (Fig. 10). Each of the trials last for 1 minute. 
Figure 9. Upper Trunk Co-ordination trials. Control subject demonstrates upper trunk co-ordination by rotating upper trunk about superior-inferior axis.
Figure 10. Lower trunk co-ordination. For lower trunk, the subject is asked to rotate his/her lower trunk and pelvis in the transverse plane to bring forward each of his/her knees in turn at a comfortable speed.
Portable Backrest Due to some discomfort which my occur from constant unsupported sitting during the test, a ortable backrest has been constructed to provide support between the trials (Fig. 11) 
Figure 11. Implementation of portable backrest. Due to discomfort that may arise during extended periods of (A) unsupported seating, a (B) portable backrest was designed to quickly be (C) implemented in between trials to provide temporary support for subject.
*Verheyden G (2004) The Trunk Impairment Scale: a new tool to measure motor impairment of the trunk after stroke. Clin Rehabil 18: 326-334 | 
