Joseph A. Gordon III¹, Nicholas M. Remillard¹, Luke R. Arieta¹, Rajakumar M. Nagarajan^1,2, Bruce M. Damon^3,4, and Jane A. Kent^1,2
1Kinesiology, University of Massachusetts Amherst, Amherst, MA, United States, ²Institute for Applied Life Sciences, University of Massachusetts Amherst, Amherst, MA, United States, ³Carle Clinical Imaging Research Program, Carle Health, Urbana, IL, United States, ⁴Stephens Family Clinical Research Institute, Carle Health, Urbana, IL, United States

The purpose of this work was to investigate the associations between skeletal muscle function and markers of fat content, diffusivity, and architecture. Combining MR techniques that measure muscle structure in vivo with muscle strength measures can elucidate mechanisms other than muscle size that may contribute to functional impairments. This study identified variables (e.g., fascicle length, muscle curvature, pennation angle, fat fraction) that are worthy of investigation in conjunction with functional parameters in future research, and highlighted assessment of muscle function across various contraction types and velocities.

Donnie Cameron^1,2, Thom T.J. Veeger¹, Esther J. Schrama³, Celine Baligand⁴, Lydiane Hirschler¹, Matthias J.P. van Osch¹, and Hermien E. Kan^1
1C.J. Gorter Center for High Field MRI, Department of Radiology, Leiden University Medical Center, Leiden, Netherlands, ²Norwich Medical School, University of East Anglia, Norwich, United Kingdom, ³Department of Neurology, Leiden University Medical Center, Leiden, Netherlands, ⁴Université Paris-Saclay, CEA, CNRS, MIRCen, Laboratoire des Maladies Neurodégénératives, Fontenay-aux-Roses, France

To develop perfusion applications in neuromuscular diseases, we examined the SNR and noise sensitivity of pulsed ASL for estimating microvascular reactivity and perfusion in skeletal muscle. We determined muscle blood flow and arterial transit time in volunteers after exercise and used these to inform simulations to establish the minimum ASL SNR required for accurate and precise perfusion parameter estimation. We observed a peak SNR of 10 in the tibialis anterior muscle. Simulations indicated that SNR>6 is sufficient for accurate parameter estimation, while SNR>12 is preferred for good precision. These findings will inform future exercise ASL protocol development in patient cohorts.  

Andrew M Schmidt¹, Elka Rubin¹, Michael Ko¹, Lauren E Watkins², Marco Barbieri¹, Laurel Hales³, Garry Gold^1,2, Scott Delp^2,4,5, Feliks Kogan¹, Valentina Mazzoli¹, and Akshay S Chaudhari^1,6
1Radiology, Stanford University, Stanford, CA, United States, ²Bioengineering, Stanford University, Stanford, CA, United States, ³Electrical Engineering, Stanford University, Stanford, CA, United States, ⁴Mechanical Engineering, Stanford University, Stanford, CA, United States, ⁵Orthopaedic Surgery, Stanford University, Stanford, CA, United States, ⁶Biomedical Data Science, Stanford University, Stanford, CA, United States

Current evaluation methods of rehabilitation following acute musculoskeletal injuries are largely qualitative. MRI and biomechanics tools can provide sensitive, quantitative measures of knee joint and lower extremity muscle changes, but the relationship between MRI and gait markers is not well characterized. We combined an MRI protocol with wearable sensors in healthy participants to characterize the relationship between gait kinematic asymmetries and thigh muscle and cartilage morphology and composition. We show that vastus lateralis (VL) muscle microstructure assessed via Diffusion Tensor Imaging (DTI) may be sensitive to gait variations. Future work may further explore these correlations in patients with musculoskeletal injuries.

Ravi Prakash Reddy Nanga¹, Mark Elliott², Neil Wilson², Sophia Swago³, Walter Witschey², and Ravinder Reddy^2
1Perelman School of Medicine at The University of Pennsylvania, Philadelphia, PA, United States, ²Radiology, Center for Advance Metabolic Imaging in Precision Medicine, Perelman School of Medicine at The University of Pennsylvania, Philadelphia, PA, United States, ³Bioengineering, University of Pennsylvania, Philadelphia, PA, United States

¹H downfield MRS is an emerging technique for quantification of in vivo NAD+ concentrations in human tissue. Though NAD⁺ can be detected in the skeletal muscle downfield spectrum (>4.7 ppm) despite its low concentration (<1 mM), quantification of [NAD⁺] in muscle disorders is further challenged by depleted NAD⁺ and long scan times. To address this challenge, we developed and optimized a downfield MRS technique for rapid measurement of NAD⁺ targeting calf skeletal muscle in a large voxel approach. This optimized technique has allowed us rapid measurement of NAD⁺ and shortened scan time considerably compared to previously published MRS techniques.

Zane Coleman¹, Ayaz Khan¹, Shubo Wang¹, Pat Hanby¹, Donald Wallace¹, Zoltan Patay¹, and Puneet Bagga^1
1Diagnostic Imaging, St. Jude Children's Research Hospital, Memphis, TN, United States

CrCEST MRI indirectly measures creatine as it exchanges amine (-NH₂) protons with free water in the skeletal muscle. When performed on the calf after plantar flexion exercise, creatine recovery data can be obtained in all four major calf muscles. While 2D CrCEST has been performed in adults before, no studies showing its reproducibility have been published. Here, we show that the essential muscles in plantar flexion movement, gastrocnemius muscles, in the calf are candidates for reproducible CrCEST data.

Joshua Ethan McGillivray^1,2, Dinesh Kumbhare^1,2,3, and Michael D Noseworthy^1,2,4,5
1School of Biomedical Engineering, McMaster University, Hamilton, ON, Canada, ²Imaging Research Centre, St. Joseph's Healthcare, Hamilton, ON, Canada, ³Department of Medicine and Biomedical Engineering, University of Toronto, Toronto, ON, Canada, ⁴Electrical and Computer Engineering, McMaster University, Hamilton, ON, Canada, ⁵Department of Radiology, McMaster University, Hamilton, ON, Canada

Lower leg resting-state BOLD images (n=8 males, 4 endurance, 4 power athletes), were acquired after 30min of rest, allowing for blood-flow normalization. BOLD images were motion corrected and the gastrocnemius and soleus manually segmented using an anatomical reference for their differing twitch fibre profiles. Voxel-wise BOLD mono- and bi-fractal dimension were computed using the scaled windowed variance approach, with linear detrending, removing scanner induced low-frequency variations. The bi-fractal dimension was significantly different between endurance and power groups in both muscles. Specific bi-fractal components more readily distinguished soleus and gastrocnemius for the endurance (I) and power (II) group, when at rest.

Jill T Shah¹, Tanay T Shah², Haresh R Rajamohan^3,4, Katherine Medina^3,5, Smita Rao⁶, Cem Deniz^3,5, and Ryan Brown^3,5
1NYU Grossman School of Medicine, New York University, New York, NY, United States, ²College of Arts and Sciences, New York University, New York, NY, United States, ³Center for Biomedical Imaging, Department of Radiology, New York University Grossman School of Medicine, New York, NY, United States, ⁴Center for Data Science, New York University, New York, NY, United States, ⁵Center for Advanced Imaging Innovation and Research (CAI2R), Department of Radiology, New York University Grossman School of Medicine, New York, NY, United States, ⁶Department of Physical Therapy, New York University, New York, NY, United States

Diabetic peripheral neuropathy (DPN) is characterized by increased adiposity implicated in metabolic dysfunction. Proton-based Dixon MRI is an appropriate means to quantify adiposity, but analysis requires time-consuming manual image segmentation. To address this problem, we developed an automated segmentation algorithm based on convolutional neural networks that provided high dice similarity coefficient scores (>0.88) on multiple regions of interest (ROI) within the calf. We utilized the networks to analyze fat fraction trends in individuals with DPN following a 10-week supervised exercise program. We measured decreased adiposity in the combined calf interstitial and muscle space (P<0.1) but not in individual muscle ROIs.

Michael Ko¹, Andrew Schmidt¹, Elka Rubin¹, Lauren Watkins², Marco Barbieri¹, Laurel Hales³, Anthony Gatti¹, Garry Gold^1,2, Feliks Kogan¹, Scott Delp^2,4,5, Valentina Mazzoli¹, and Akshay Chaudhari^1,6
1Radiology, Stanford University, Stanford, CA, United States, ²Bioengineering, Stanford University, Stanford, CA, United States, ³Electrical Engineering, Stanford University, Stanford, CA, United States, ⁴Mechanical Engineering, Stanford University, Stanford, CA, United States, ⁵Orthopaedic Surgery, Stanford University, Stanford, CA, United States, ⁶Biomedical Data Science, Stanford University, Stanford, CA, United States

Gait retraining has been studied as an intervention to improve osteoarthritis symptoms but has a variety of limitations. An alternative approach may be muscle strengthening interventions that have known impacts on gait changes. However, limited studies have examined the quantitative relationships between muscle strengthening and gait. Here, we combine a rapid MRI protocol with wearable sensors to determine that a 12-week exercise intervention induced significant changes in both quadricep muscle morphology and gait kinematics. Morphological changes in different muscles were related to different kinematic changes, which may inform future strengthening interventions that aim to achieve specific changes in gait kinematics.

Mahsa Salimi Majd¹, Markus Lerchbaumer¹, Suchung Kim¹, Tilman Hees¹, Yang Yang¹, Steffen Görner¹, Thomas Fischer¹, Ingolf Sack¹, and Jing Guo^1
1Charité - Universitätsmedizin Berlin, Berlin, Germany

Achilles tendinopathy is a common injury among athletes. It has been reported that the calf muscles are involved and affected by Achilles tendinopathy. In this study, tomoelastography, a multifrequency MR-elastography (MRE) technique, was applied to patients with Achilles tendinopathy for the characterization of the biomechanical properties of gastrocnemius muscle. Preliminary findings showed the stiffness of gastrocnemius muscle increased significantly as a result of tendinopathy.

Constance P Michel¹, Laurent A Messonnier², Benoit Giannesini¹, and David Bendahan^1
1Aix-Marseille Univ, CNRS, CRMBM, Marseille, France, ²Laboratoire Interuniversitaire de Biologie de la Motricité, Chambéry, France

 In sickle cell disease (SCD), hydroxyurea can increase foetal haemoglobin production and restore oxygen-carrying capacity of patients. As endurance training is also known to improve muscle aerobic capacity, one can wonder whether hydroxyurea and endurance training can have a synergetic effect. Townes mice (model of SCD) have been trained with and without hydroxyurea for 8 weeks. Muscle energetics and function were assessed in response to a rest-stimulation-recovery protocol before and after interventions. While training increased force production, mitochondrial function was improved when training was combined to hydroxyurea supplementation. In SCD, this combination would have a synergetic effect in muscle energetics.

Carly Anne Lockard^1,2, Melissa T Hooijmans³, Xingyu Zhou^2,4, and Bruce M Damon^1,2
1Stephens Family Clinical Research Institute, Carle Health, Urbana, IL, United States, ²Carle Clinical Imaging Research Program, Carle Health, Urbana, IL, United States, ³Department of Radiology and Nuclear Medicine, Amsterdam Movement Sciences, Amsterdam University Medical Center, Amsterdam, Netherlands, ⁴Biomedical Engineering, Vanderbilt University, Nashville, TN, United States

Diffusion-tensor MRI tractography has been applied to characterize muscle diffusion properties and architecture. Tract propagation is performed using various propagation algorithms, settings, and step sizes. Smoothing, such as by polynomial fitting, has been shown to reduce tract path errors due to noise. Optimal settings for tracking algorithm, parameters, and order of subsequent smoothing polynomials have not been established. This work examines the effect of tracking algorithm settings, step size, and polynomial order on muscle architecture and diffusion estimates and polynomial fit residuals behavior. Results emphasize the high sensitivity of curvature estimates to smoothing polynomial order.

Thaejaesh Sooriyakumaran¹, Joshua McGillivray¹, and Michael D Noseworthy^1
1School of Biomedical Engineering, McMaster University, Hamilton, ON, Canada

This investigation looks to use graph signal processing to generate a functional graph over the anatomical leg and assess the extracted functional information and its viability in modeling underlying physiological factors. The generated graph is constructed on the edge dimensions of node to node coherences and fractal dimension differences. The resultant graph structure is then analyzed to observe if extracted functional data shows alignment with underlying structures via a generalized linear mixed-effect mode