1. Abstract Previous Design: Initial Results References A MR Compatible Device for Imaging the Hamstrings During Movement and Under Load Eric Bader 1, Arinne Lyman 1, Sarajane Stevens 1, Christopher Westphal 1 Advisor: William Murphy 1, Client: Darryl Thelen 1,2 1 Department of Biomedical Engineering, 2 Department of Mechanical Engineering Hamstring injuries are one of the most common muscle maladies in athletes, especially in runners. Injury usually occurs late in swing phase when the hamstrings is undergoing an eccentric (lengthening) contraction to decelerate the limb prior to foot contact. Re-injury is common among athletes who sustain an initial hamstrings strain injury. An assessment of the residual changes to muscle mechanics and function is therefore important for the development of more effective rehabilitation programs. However, standard muscle imaging techniques are static, and thus cannot provide direct measurements of how the muscle functions under load and during movement. We have developed a MR compatible device that will inertially load the hamstrings in a manner similar to that seen during the swing phase of running. Our first prototype used EMG and motion capture to validate the inertial loading concept. The second generation prototype consists of a compact leg support assembly and coupled inertial loading system. The loading system has an overall 10:1 gear ratio, such that a set of disks rotates 10 times faster than the leg itself. This allows the limited knee range of motion in the MR magnet to generate sufficient accelerations to load the lengthening hamstrings. We validated our second prototype in the motion analysis lab by comparing our loading system to a simple elastic spring system, which has been used by previous researchers. In addition, we also confirmed that our device works in a standard MR magnet, and is compatible with a an extremity wrap coil and plethysmograph trigger device and. The next test will be to acquire dynamic images in the MR magnet. Conclusions/Future Work Problem Definition Motivation : Current musculoskeletal imaging techniques are primarily static--do not provide a direct assessment of dynamic movement and loading. Dynamic imaging allow one to visualize tissue and joint behavior during movement—provide improved diagnosis of impairments and assessment of clinical outcomes. Requires a device that can guide movement and load musculoskeletal structures within the constraints of imaging hardware and software. Device will dynamically load the hamstrings during a lengthening contraction, which is the type of condition that can induce musculotendon injury during running. To be used with phase contrast imaging, which acquires images over many cycles of repeated motion to determine tissue velocity at every pixel within the imaging plane. Velocity data can be integrated to estimate displacements and strain distribution in the muscle tissue. [1] Asakawa DS, Pappas GP, Blemker SS, Dracce JE, Delp SL. Cine phase-contrast magnetic resonance imaging as a tool for quantification of skeletal muscle motion. Seminars in Musculoskeletal Radiology. 2003; 7(4):287-295. [2] Patel VV, Hall K, Ries M, Lindsey C, Ozhinsky E, Lu Y, Majumdar S. Magnetic resonance imaging of patellofemoral kinematics with weight-bearing. Journal of Bone and Joint Surgery. 2003; 85:2419-2424. [3] Thelen DG, Chumanov ES, Sherry MA, Heiderscheit BC. Neuromusculoskeletal models provide insights into the mechanisms and rehabilitation of hamstring strains. Exercise and Sports Science Reviews. 2006; 34(3): 135-141. Acknowledgments Authors would like to thank Advisor William Murphy and Client Darryl Thelen. Also, special thanks to Amy Silder, Kelli Hellenbrand, Scott Reeder, Sylvia Blemker, Frank Korosec and Wally Block for all their help making this project go forward. Figure 3: This graph shows the superposition of a hamstring stretch- shortening cycle and EMG data of muscle activation collected in a motion capture lab. The desired outcome has the hamstring active when the muscle is stretched in an eccentric contraction. Rationale for inertial loading : In the running gait cycle, the hip flexes and knee rapidly extends during mid-swing. Deceleration of the limb occurs during late swing prior to foot contact. The deceleration of the shank induces an inertial torque about the knee. The hamstrings are active and lengthening during late swing to counteract the inertial loading.  Active, lengthening muscle contractions under large load are the type of conditions that can induce muscle injury. Mimic type of loading associated with injury in an MR magnet  use an inertial load Final Design: Results Figure 5: The gear box of the inertia loading system translates the motion of the shank to the inertial disks to create a substantial torque at the knee. To further increase the gearing ratio and provide more torque, spur gears are implemented with a ratio of 2.66:1 to give an overall gearing ratio of 10:1. Figure 4: The leg support assembly consists of a leg brace to guide the shank and prevent translational motion. The sprocket at the knee connects a plastic roller chain to a sprocket at the inertial loading system and increases the gear ratio by 3.75 to allow for added torque to be placed on the knee. Final Design: Components Problem Statement: Develop a MR-compatible device that will induce repeatable, lengthening, low-load contractions of the hamstring muscles. The device must be non-ferrous, fit within the bore of a standard MRI machine and allow for a flexible MR coil to be placed around the fixed thigh. A trigger will be used to signal the start and end of each cycle of motion. Figure 6: The plethysmograph trigger is seen next to the ankle. It will be used as a triggering device to tell the computer which image starts each cycle of motion. The motion of the leg will block and unblock the light emitted from the trigger and send the signal to the computer to mark the start of each motion cycle. Figure 1: The running gait cycle. We are focusing on the 2 nd half of swing phase, approximately 70-100% of the gait cycle [3]. Figure 11: Our device in the motion capture lab: The subject is performing flexion/extension under an elastic load to compare with our inertial system of loading the hamstrings. EMG and kinematic data were collected to assess repeatability. Figure 12: Our device in an MRI scanner: The subject is undergoing flexion/extension motion inside the constraints of the bore. MR compatibility and trigger compatibility was assessed. Have achieved: Physiologically loads hamstring MR compatible (size, materials, coil) Compatible with trigger Repeatability Applications of device: Assessment of rehabilitation Prevention of re-injury Market competition: Only device to use inertia True simulation of muscle action Future work: Obtain CINE-PC pulse sequence Dynamic imaging in MRI Obtain data from previously injured subjects Figure 2: Our first generation prototype consisted of an open loop design with a counterweight. It was used to validate the inertial loading conditions, but was rebuilt to be more stable, compact, and with a closed loop design that allowed us to increase the load. Figure 7: The repeatability of the knee angle over 12 cycles: Average range of motion was 45.5°+1.04 (max) and 14.6 °+0.85 (min). Figure 8: The torque exerted on the knee over 12 motion cycles: Average torque was 17.5+0.9 N-m (max) and -17.6+0.67 N-m (min). Figure 9: This graph shows the EMG data of hamstring activation superimposed with the cycle of motion during inertial loading. The hamstrings become most active when the knee is fully extended. This activity shows the muscle performing a lengthening contraction. Figure 10: This graph shows the EMG data of hamstring activation superimposed with the cycle of motion during elastic loading. The hamstrings become most active when the knee is undergoing flexion. This activity shows the muscle performing a shortening contraction.