10:00 - 10:20 amSaturday, September 6
Plenary Hall
Development of a Microsoft kinect-based screening tool and video game intervention to decrease risk of ACL tears in female athletes
Plenary Hall
Development of a Microsoft kinect-based screening tool and video game intervention to decrease risk of ACL tears in female athletes
Family and Sports Medicine Physician, University of Missouri
BackgroundCompared to male athletes, female athletes who participate in high-risk sports that involve cutting, jumping, and pivoting motions suffer injuries of the anterior cruciate ligament (ACL) at... Read more

Description

Background

Compared to male athletes, female athletes who participate in high-risk sports that involve cutting, jumping, and pivoting motions suffer injuries of the anterior cruciate ligament (ACL) at rates 4 to 6 times greater than male athletes.  A number of biomechanical measures about the knee – including knee abduction angle, frontal plane knee angle, and knee-to-ankle separation ratio – captured during the Drop Vertical Jump (DVJ) test have been studied as indicators of future ACL injury risk.  However, capturing these measures is not easy, as it generally involves the use of expensive equipment, such as 3D motion capture systems found in performance labs, or manual joint identification from the recordings of a calibrated camera system.

A fast, low cost, portable system for capturing these biomechanical measures would facilitate more widespread screening of female athletes for elevated risk of ACL injuries, and, thus, allow for more targeted use of injury prevention programs. With the release of the Microsoft Kinect sensor device and corresponding Software Development Kit (SDK), a low-cost method for obtaining a three-dimensional skeletal model (joint positions) of an individual without the need for wearable markers or manual joint identification became available.

Methods

The Kinect skeletal model was evaluated for use in capturing four biomechanical measures during the DVJ test which have proven to be useful in screening for future ACL injuries. Specifically, these were knee valgus motion as measured from initial contact (IC) to the point of peak flexion (PF), frontal plane knee angle at both IC and PF, and knee-to-ankle separation ratio measured at PF.  A Vicon motion capture system was used for ground truth.  Thirteen participants ages 20 to 31 successfully completed between 5 to 7 DVJs for a total of 84 tests.

Results

Comparison of joint tracking data from the Microsoft Kinect and the Vicon system showed frontal plane knee angle and knee-knee:ankle-ankle separation ratio.  Both had an intra-class correlation coefficient value of 0.89, which indicates excellent agreement between measures.  Frontal plane knee angle at initial contact had the lowest standard deviation of 1.88 (left) and 1.69 (right).  This gives a 95% confidence
interval of being within ~ +/- 3.5 degrees for a single jump/landing test.

Conclusion 

The Microsoft Kinect’s accuracy of skeletal tracking offers biomechanical screening to identify female athletes with high risk of ACL tear in a fast, low cost, portable screening tool.  Increasing the number of measured jump/landing tasks during screening could increase the accuracy.  If an athlete jumped 3 times, they would theoretically decrease the standard deviation to approximately +/- 2 degrees.  At that level, it would be acceptable for a screening test to identify female youth athletes at increased risk of ACL tear. Future studies should investigate preventative strengthening exercises and biofeedback of landing utilizing the Microsoft Kinect motion sensor device.

 

Aaron Gray, M.D. is a family physician and sports medicine physician at the University of Missouri Hospital and Clinics. He has a joint faculty appointment as an assistant professor in the Department of Family Medicine and the Department of Orthopaedics. He is board certified from the American Board of Family Medicine and has a Certificate of Added Qualifications in sports medicine. His research interests include using technology to prevent, diagnose and treat medical and orthopaedic diseases. He has co-developed software that utilizes the Microsoft Kinect for Windows motion sensor to detect female athletes who have increased risk of tearing the ACL ligament of the knee. In 2013 he became a member of the University of Missouri Health Care Physician EMR Council. His sports medicine duties include being a team physician for University of Missouri athletes and team doctor for Mizzou Baseball.  He is a team physician for US Soccer youth national teams.

He is a Techstars mentor for the Sprint Mobile Health Accelerator in Kansas City Spring '14 class.

He graduated from Tennessee Technological University in Cookeville, TN with a BS in Chemistry and a minor in Biology. Dr. Gray completed medical school at the University of Tennessee Health Science Center-College of Medicine. He completed a residency in Family Medicine at the University of Missouri. He also completed a primary care sports medicine fellowship at the University of California-Los Angeles. While at UCLA, he was the team physician for men's soccer, baseball, men's volleyball, women's golf and assistant team physician for football.

Contact Us

We're not around right now. But you can send us an email and we'll get back to you, asap.

Not readable? Change text. captcha txt

Start typing and press Enter to search