Assessment of Proprioception Impairments in Children with Unilateral Spastic Cerebral Palsy Using a Markerless Motion Capture System David Putrino, PT, PhD 1,2 Karen Chin, MA 1,2 Behdad Dehbandi, PhD 1,2 Victor Nwankwo, MD 1,2 Andrew Gordon, PhD 3 Kathleen Friel, PhD 1,2,4 1 Burke Medical Research Institute, White Plains, NY 2 Weill Cornell Medical College, New York, NY 3 Teachers College Columbia University, New York, NY 4 Blythedale Children’s Hospital, Valhalla, NY BACKGROUND / OBJECTIVE d • Proprioception is the ability to sense the position of a limb in space without visual feedback. It is an important contributor to motor control. d • Neurological impairment oſten causes proprioception deficits, which can impede one’s ability to perform motor skills accurately. d • Currently, there are no reliable quantitative scales that can measure proprioception accurately and affordably. d • We developed a low-cost assessment tool using the Microsoſt Kinect 2 to non-invasively and easily quantify proprioception in children with unilateral spastic cerebral palsy (USCP). d METHODS D • 29 children with USCP (6.1-19y), 13 females 6 typically-developing controls (9.2-17.9y), 6 females d1 • Participants were seated 1m in front of the Kinect. d • ey had to perform 3 different tasks while blindfolded, and hold each position for 3 seconds: • • 3 centroids of interest were the elbow, shoulder, and wrist. s • e model arm was always placed in position by the experimenter. RESULTS D D • d DISCUSSION / CONCLUSIONS • d • Proprioception is an important component of motor control, but has been difficult in the past to quantify. D • We focused our analysis only on the contralateral matching task because factors such as attention and memory may have confounded results of the other two tasks. In the future, we plan to administer psychosocial measures to account for such factors. • Correlational analysis showed significant associations between y- and z- coordinate displacement and widely-used clinical measures. More advanced analyses are needed to examine these further. METHODS (cont’d) • e Kinect captured the xyz coordinates of each of the child’s upper body joints. D • • Matlab was used to process the data and quantify differences in joint positions between the model and test arms. Ds • Other measures collected: 1: Ipsilateral Remembered 2: Contralateral Remembered 3: Contralateral Matching Model Arm Test Arm Hand Function and Dexterity • Manual Ability Classification System (MACS) • Jebsen-Taylor Test of Hand Function (JTTHF) • Box and Blocks (BB) Somatosensory • Cooper-Stereognosis (CS) Questions/Comments? Email Karen Chin at [email protected] ✉ Time (frames) z-coord distance from centroid y-coord distance from centroid Time (frames) x-coord distance from centroid Time (frames) y-coordinate displacement z-coordinate displacement y-coordinate displacement z-coordinate displacement Spearman rho 0.48 0.4 0.58 0.4 Sig. (2-tailed) 0.028 0.08 0.005 0.06 Spearman rho 0.48 N.S. 0.46 N.S. Sig. (2-tailed) 0.026 N.S. 0.03 N.S. Spearman rho 0.81 N.S. N.S. N.S. Sig. (2-tailed) 0.0085 N.S. N.S. N.S. Spearman rho 0.74 0.62 0.67 0.61 Sig. (2-tailed) 0.00027 0.005 0.0016 0.0057 Spearman rho N.S. N.S. 0.49 N.S. Sig. (2-tailed) N.S. N.S. 0.02 N.S. N.S. = not significant JTTHF MACS Affected Elbow Affected Shoulder Contralateral Matching Task - Significant Correlations Cooper-Stereognosis Box and Blocks AHA • We found significant differences between CP and controls for y- and z-coordinate displacement scores for elbow and shoulder centroids in the contralateral matching task (p < .001). 0 1 2 3 4 5 6 7 8 Mean Y displacement Mean Z displacement Mean Y displacement Mean Z displacement Elbow Shoulder Control CP Distance from centroid