Wendy M. Murray
· Professor of Biomedical EngineeringVerifiedNorthwestern University · Chemical Engineering
Active 1957–2024
About
Wendy M. Murray is a Professor of Biomedical Engineering and a Professor of Physical Medicine and Rehabilitation at Northwestern University. Her research focuses on developing biomechanical models that accurately represent the mechanical actions of the upper extremity muscles. She has shared these models and anatomical databases with the scientific community, which have been cited hundreds of times. Her main research application is to better understand and improve the function of the disabled upper limb, with relevance to basic motor control, control systems for exoskeletons and upper limb prosthetics, restoration of hand and arm function following cervical spinal cord injury, rehabilitation after stroke, orthopaedic interventions, and injury prevention in baseball pitching.
Research topics
- Computer Science
- Artificial Intelligence
- Human–computer interaction
- Simulation
Selected publications
Biomechanical Arm and Hand Tracking with Multiview Markerless Motion Capture
2024-09-01 · 6 citations
articleHuman arm and hand function is extremely complex with many degrees of freedom. It is also a common target for clinical interventions. However, precisely measuring upper extremity movement in both clinical and research settings is logistically challenging. We overcame this challenge through a novel approach to reconstructing arm biomechanics from markerless motion capture from multiple synchronized videos. Our approach directly opti-mizes the kinematics of an accurate biomechanical arm and hand that allows end-to-end minimization of the errors between the reconstructed movements and keypoints detected by computer vision. Key to this is an implicit function that maps from time to joint kinematics, which provides a learnable trajectory representation that can be differentiated through the biomechanical model, and supports GPU acceleration using MuJoCo-MJX. This optimization solves for the inverse kinematic solution consistent with the measured keypoints, consistent with biomechanical constraints, in addition to scaling the model while solving for the kinematics. We compare different hand keypoint detectors and find the best produces a fit with only several millimeters of reconstruction error. We also find that end-to-end optimization outperforms a two-stage fitting procedure, equivalent to more traditional biomechanical pipelines, where we first compute 3D marker trajectories and then perform inverse kinematics fitting in OpenSim. We anticipate this framework will reduce the barriers to biomechanical analysis of the arm and hand in both clinical and research settings.
Soft, 3D printed muscle ultrasound phantom with structurally tunable B-mode echo intensity
bioRxiv (Cold Spring Harbor Laboratory) · 2024-12-03
preprintOpen accessSenior authorCorrespondingABSTRACT OBJECTIVES Imaging phantoms for training and validation are vital to improving the performance and adoption of ultrasound imaging modalities in clinical and pre-clinical applications, and the goal of this study was to assess the viability of 3D printed muscle ultrasound phantoms to meet this need. METHODS We used a soft stereolithography resin to 3D print phantoms that mimicked the fascicle- and perimysium-scale structure of skeletal muscle and compared the long axis B-mode imaging quality and pattern of the phantom to that of healthy, adult Biceps brachii. We used a pulse-echo, time-of-flight method to measure the acoustic impedance of the resin for comparison to skeletal muscle and common soft tissue mimicking materials. We analyzed the echo intensity (EI) of muscle images to establish a physiological range and compared the EI of different phantom designs to assess the ability to control imaging brightness through structural modification. RESULTS A linear, striated hyper-/hypo-echoic B-mode imaging pattern mimicking long axis Biceps brachii muscle images was achieved with two 3D structure paradigms, rod and honeycomb. Acoustic impedance of Elastic 50A resin is higher than skeletal muscle in bulk, but appears suitable for use in a 3D structured phantom. EI measured in the Biceps images were found to vary both within and across images with an overall mean ± SD of 87 ±13 AU. EI measured in honeycomb phantoms (55 ±15 AU) was higher than in rod phantoms (42 ±13 AU), and a latticed honeycomb further increased EI (90 ±11 AU). CONCLUSIONS This study serves as proof-of-concept for soft, 3D printed phantoms that replicate the characteristic muscle ultrasound imaging pattern with the ability to tune clinically relevant EI values via structural design.
Biceps Tenotomy and Tenodesis Surgeries Under-Tension Muscle: A Simulation Study
SSRN Electronic Journal · 2023-01-01
preprintOpen accessSenior authorJournal of Biomechanics · 2023-02-22 · 5 citations
articleOpen accessSenior authorCorrespondingMagnetic Resonance in Medicine · 2023-10-10 · 6 citations
articleOpen accessPURPOSE: To determine the sensitivity profiles of probabilistic and deterministic DTI tractography methods in estimating geometric properties in arm muscle anatomy. METHODS: Spin-echo diffusion-weighted MR images were acquired in the dominant arm of 10 participants. Both deterministic and probabilistic tractography were performed in two different muscle architectures of the parallel-structured biceps brachii (and the pennate-structured flexor carpi ulnaris. Muscle fascicle geometry estimates and number of fascicles were evaluated with respect to tractography turning angle, polynomial fitting order, and SNR. The DTI tractography estimated fascicle lengths were compared with measurements obtained from conventional cadaveric dissection and ultrasound modalities. RESULTS: The probabilistic method generally estimated fascicle lengths closer to ranges reported by conventional methods than the deterministic method, most evident in the biceps brachii (p > 0.05), consisting of longer, arc-like fascicles. For both methods, a wide turning angle (50º-90°) generated fascicle lengths that were in close agreement with conventional methods, most evident in the flexor carpi ulnaris (p > 0.05), consisting of shorter, feather-like fascicles. The probabilistic approach produced at least two times more fascicles than the deterministic approach. For both approaches, second-order fitting yielded about double the complete tracts as third-order fitting. In both muscles, as SNR decreased, deterministic tractography produced less fascicles but consistent geometry (p > 0.05), whereas probabilistic tractography produced a consistent number but altered geometry of fascicles (p < 0.001). CONCLUSION: Findings from this study provide best practice recommendations for implementing DTI tractography in skeletal muscle and will inform future in vivo studies of healthy and pathological muscle structure.
SSRN Electronic Journal · 2022-01-01
articleOpen accessSenior authorFrontiers in Physiology · 2022-02-08 · 8 citations
articleOpen accessSenior authorCorrespondingThe lengths of a muscle’s sarcomeres are a primary determinant of its ability to contract and produce force. In addition, sarcomere length is a critical parameter that is required to make meaningful comparisons of both the force-generating and excursion capacities of different muscles. Until recently, in vivo sarcomere length data have been limited to invasive or intraoperative measurement techniques. With the advent of second harmonic generation microendoscopy, minimally invasive measures of sarcomere length can be made for the first time. This imaging technique expands our ability to study muscle adaptation due to changes in stimulus, use, or disease. However, due to past inability to measure sarcomeres outside of surgery or biopsy, little is known about the natural, anatomical variability in sarcomere length in living human subjects. To develop robust experimental protocols that ensure data provide accurate representations of a muscle’s sarcomere lengths, we sought to quantify experimental uncertainty associated with in vivo measures of sarcomere lengths. Specifically, we assessed the variability in sarcomere length measured (1) within a single image, along a muscle fiber, (2) across images captured within a single trial, across trials, and across days, as well as (3) across locations in the muscle using second harmonic generation in two upper limb muscles with different muscle architectures, functions, and sizes. Across all of our measures of variability we estimate that the magnitude of the uncertainty for in vivo sarcomere length is on the order of ∼0.25 μm. In the two upper limb muscles studied we found larger variability in sarcomere lengths within a single insertion than across locations. We also developed custom code to make measures of sarcomere length variability across a single fiber and determined that this codes’ accuracy is an order of magnitude smaller than our measurement uncertainty due to sarcomere variability. Together, our findings provide guidance for the development of robust experimental design and analysis of in vivo sarcomere lengths in the upper limb.
A Musculoskeletal Model of the Hand and Wrist Capable of Simulating Functional Tasks
IEEE Transactions on Biomedical Engineering · 2022-11-03 · 48 citations
articleOpen accessSenior authorOBJECTIVE: The purpose of this work was to develop an open-source musculoskeletal model of the hand and wrist and to evaluate its performance during simulations of functional tasks. METHODS: The current model was developed by adapting and expanding upon existing models. An optimal control theory framework that combines forward-dynamics simulations with a simulated-annealing optimization was used to simulate maximum grip and pinch force. Active and passive hand opening were simulated to evaluate coordinated kinematic hand movements. RESULTS: The model's maximum grip force production matched experimental measures of grip force, force distribution amongst the digits, and displayed sensitivity to wrist flexion. Simulated lateral pinch strength replicated in vivo palmar pinch strength data. Additionally, predicted activations for 7 of 8 muscles fell within variability of EMG data during palmar pinch. The active and passive hand opening simulations predicted reasonable activations and demonstrated passive motion mimicking tenodesis, respectively. CONCLUSION: This work advances simulation capabilities of hand and wrist models and provides a foundation for future work to build upon. SIGNIFICANCE: This is the first open-source musculoskeletal model of the hand and wrist to be implemented during both functional kinetic and kinematic tasks. We provide a novel simulation framework to predict maximal grip and pinch force which can be used to evaluate how potential surgical and rehabilitation interventions influence these functional outcomes while requiring minimal experimental data.
bioRxiv (Cold Spring Harbor Laboratory) · 2021-11-18
preprintOpen accessSenior authorCorrespondingAbstract The lengths of a muscle’s sarcomeres are a primary determinant of its ability to contract and produce force. In addition, sarcomere length is a critical parameter that is required to make meaningful comparisons of both the force-generating and excursion capacities of different muscles. Until recently, in vivo sarcomere length data have been limited to invasive or intraoperative measurement techniques. With the advent of second harmonic generation microendosopy, minimally invasive measures of sarcomere length can be made for the first time. This imaging technique expands our ability to study muscle adaptation due to changes in stimulus, use, or disease. However, due to the inability to measure sarcomeres outside of surgery or biopsy, little is known about the natural, anatomical variability in sarcomere length in living human subjects. To develop robust experimental protocols that ensure data provide accurate representations of a muscle’s sarcomere lengths, we sought to quantify experimental uncertainty associated with in vivo measures of sarcomere lengths. Specifically, we assessed the variability in sarcomere length measured 1) within a single image, along a muscle fiber, 2) across images captured within a single trial, across trials, and across days, as well as 3) across locations in the muscle using second harmonic generation in two upper limb muscles with different muscle architectures, functions, and sizes. Across all of our measures of variability we estimate that the magnitude of the uncertainty in in vivo sarcomere length are on the order of ~0.25μm. In the two upper limb muscles studied we found larger variability in sarcomere length within a single insertion than across locations. We also developed custom code to make measures of sarcomere length variability across a single fiber and determined that this codes’ accuracy is an order of magnitude smaller than our measurement uncertainty due to sarcomere variability. Together, our findings provide guidance for the development of robust experimental design and analysis of in vivo sarcomere lengths in the upper limb.
Journal of Biomechanics · 2021-11-21 · 3 citations
erratumOpen accessSenior authorCorresponding
Recent grants
NIH · $1.3M · 2019
Prosthesis Control by Forward Dynamic Simulation of the Intact Biomedical system
NIH · $1.7M · 2011–2016
NIH · $1.3M · 2011
A Comparison of Two Surgical Procedures that Restore Elbow Extension
NIH · 2010–2014
Frequent coauthors
- 59 shared
Michael S. Bednar
Loyola University Medical Center
- 57 shared
Eric J. Perreault
Shirley Ryan AbilityLab
- 56 shared
Jennifer A. Nichols
University of Florida
- 56 shared
Amy N. Adkins
North Carolina State University
- 36 shared
Benjamin I. Binder‐Markey
Drexel University
- 33 shared
Carrie L. Peterson
Virginia Commonwealth University
- 33 shared
Michael W. Keith
MetroHealth Medical Center
- 30 shared
Sarah J. Wohlman
Northwestern University
Education
- 1997
PhD, Biomedical Engineering
Northwestern University
- 1992
MS, Biomedical Engineering
Northwestern University
- 1990
BS, Mathematics
University of Notre Dame
- Resume-aware match score
- Save to shortlist
- AI-drafted outreach
See your match with Wendy M. Murray
PhdFit ranks faculty by your research interests, methods, and publications — grounded in their actual work, not templates.
- Free to start
- No credit card
- 30-second signup