AUTHOR=Bodkin Stephan G. , Smith Andrew C. , Bergman Bryan C. , Huo Donglai , Weber Kenneth A. , Zarini Simona , Kahn Darcy , Garfield Amanda , Macias Emily , Harris-Love Michael O. 

TITLE=Utilization of Mid-Thigh Magnetic Resonance Imaging to Predict Lean Body Mass and Knee Extensor Strength in Obese Adults

JOURNAL=Frontiers in Rehabilitation Sciences

VOLUME=3

YEAR=2022

URL=https://www.frontiersin.org/journals/rehabilitation-sciences/articles/10.3389/fresc.2022.808538

DOI=10.3389/fresc.2022.808538

ISSN=2673-6861

ABSTRACT=<sec><title>Purpose</title><p>To train and test a machine learning model to automatically measure mid-thigh muscle cross-sectional area (CSA) to provide rapid estimation of appendicular lean mass (ALM) and predict knee extensor torque of obese adults.</p></sec><sec><title>Methods</title><p>Obese adults [body mass index (BMI) = 30–40 kg/m<sup>2</sup>, age = 30–50 years] were enrolled for this study. Participants received full-body dual-energy X-ray absorptiometry (DXA), mid-thigh MRI, and completed knee extensor and flexor torque assessments <italic>via</italic> isokinetic dynamometer. Manual segmentation of mid-thigh CSA was completed for all MRI scans. A convolutional neural network (CNN) was created based on the manual segmentation to develop automated quantification of mid-thigh CSA. Relationships were established between the automated CNN values to the manual CSA segmentation, ALM <italic>via</italic> DXA, knee extensor, and flexor torque.</p></sec><sec><title>Results</title><p>A total of 47 obese patients were enrolled in this study. Agreement between the CNN-automated measures and manual segmentation of mid-thigh CSA was high (&gt;0.90). Automated measures of mid-thigh CSA were strongly related to the leg lean mass (<italic>r</italic> = 0.86, <italic>p</italic> &lt; 0.001) and ALM (<italic>r</italic> = 0.87, <italic>p</italic> &lt; 0.001). Additionally, mid-thigh CSA was strongly related to knee extensor strength (<italic>r</italic> = 0.76, <italic>p</italic> &lt; 0.001) and moderately related to knee flexor strength (<italic>r</italic> = 0.48, <italic>p</italic> = 0.002).</p></sec><sec><title>Conclusion</title><p>CNN-measured mid-thigh CSA was accurate compared to the manual segmented values from the mid-thigh. These values were strongly predictive of clinical measures of ALM and knee extensor torque. Mid-thigh MRI may be utilized to accurately estimate clinical measures of lean mass and function in obese adults.</p></sec>