New Technology for the Capture, Visualization, and Analysis of Human Motion

Principal Investigator: Thomas P. Andriacchi, PhD

Specific Aims

The principal goal of the proposed five-year effort is to design, develop and evaluate a novel system using multiple video sensors that will efficiently and accurately model and measure human movement. The motion capture and analysis will be completely non invasive since no markers or fixtures placement on the subject’s body will be necessary.

Progress during 2006

Work Accomplished

In the last year a first round of validation of the developed technology has been completed and published. A customizable tracking algorithm has been developed, allowing specifying the degrees of freedom of the model used to extract the kinematics data. For the model generation side, a functional joint center method has been developed and validated on a first group of subjects. A statistical method for the generation of subject specific models has also been developed, allowing the automatic processing of the captured data, and it is currently under validation process. The markerless technology has been applied to a first clinical study involving gait retraining as a preventive intervention for osteoarthritis.

Main Findings

• Kinematic data obtained through the markerless motion capture system have been validated in both virtual and experimental environment, showing good accuracy for movements in the sagittal and frontal plane (deviations on the order of 1-2° for the knee joints in the sagittal and frontal planes).
• The functional method developed for the identification of joint centers has been applied to the hip joint. Results in virtual and experimental environment showed average deviations along the 3 axes of 6mm and 12 mm, respectively. For the experimental comparison the method was compared to marker-based technique.
• The gait retraining study demonstrated how clinically valuable information could be extracted from the analysis of the subject’s motion using markerless method, in particular, the motion of the upper body.

Plan for Subsequent Years

• Improvement of the background/foreground separation algorithm
• Extension and validation of the automatic model generation algorithm
• Implementation of a markerless protocol for posturographic analysis
• Improvement of ease of use and kinematics extraction/reporting tools

Publications

Corazza S, Muendermann L, Chaudhari A, Demattio T, Cobelli C, Andriacchi T: A markerless motion capture system to study musculoskeletal biomechanics: visual hull and simulated annealing approach, Annals of Biomedical Engineering, (Epub ahead of print, May 5, 2006), Jun; 34(6):1019-29, 2006 (figure selected as manuscript cover illustration).

Corazza S, Muendermann L, Andriacchi T: Model-free markerless motion capture for biomechanical applications: visual hull and laplacian eigenmaps approach. IEEE transactions on Biomed. Eng., submitted for publication February 2006.

Muendermann, L, Corazza, S, Andriacchi, T: The evolution of methods for the capture of human movement leading to markerless motion capture for biomechanical applications. Journal of NeuroEngineering and Rehabilitation, Mar 15; 3(1):6, 2006 (Epub ahead of print), 2006.

Presentations

Andriacchi Invited Speaker: The Interaction of Mechanics and Biology in Knee Joint Restoration and Regeneration. Session Topic: Knee Kinematics and Kinetics, Berlin, Germany, June 29-July 1st, 2006.

Andriacchi Keynote Speaker: 5th World Congress of Biomechanics, Biomechanics of Ligaments and Tendons Symposium. Invited Papers: “In Vivo Knee Kinematics Following ACL Injury and Premature Osteoarthritis” and “A New Era in the Capture of Human Movement: Markerless Capture of Human Movement”, Munich, Germany, July 29-August 4, 2006.

Lars Muendermann, Stefano Corazza, Anne Muendermann, Tina Lin, Ajit Chaudhari, Thomas P. Andriacchi: Gait retraining to reduce medial compartment load at the knee assessed using markerless motion capture, 52nd Annual meeting of the Orthopaedic Research Society, Chicago, IL, March 19-22, 2006.

Stefano Corazza, Lars Muendermann, Ajit Chaudhari and Thomas P. Andriacchi. A framework for the functional identification of joint centers using markerless motion capture, validation for the hip joint, American Society of Biomechanics Meeting, Blacksburg, VA, September 6-9, 2006. Finalist contender for the American Society of Biomechanics Microstrain Award.

Funding Source: NSF