Automation of Quantifying Axonal Loss in Patients with Peripheral Neuropathies through Deep Learning Derived Muscle Fat Fraction

Abstract

Background Axonal loss denervates muscle, leading to an increase of fat accumulation in the muscle. Therefore, fat fraction (FF) in whole limb muscle using MRI has emerged as a monitoring biomarker for axonal loss in patients with peripheral neuropathies. In this study, we are testing whether deep learning‐based model can automate quantification of the FF in individual muscles. While individual muscle is smaller with irregular shape, manually segmented muscle MRI images have been accumulated in this lab; and make the deep learning feasible. Purpose To automate segmentation on muscle MRI images through deep learning for quantifying individual muscle FF in patients with peripheral neuropathies. Study Type Retrospective. Subjects 24 patients and 19 healthy controls. Field Strength/Sequences 3T; Interleaved 3D GRE. Assessment A 3D U‐Net model was implemented in segmenting muscle MRI images. This was enabled by leveraging a large set of manually segmented muscle MRI images. B1+ and B1− maps were used to correct image inhomogeneity. Accuracy of the automation was evaluated using Pixel Accuracy (PA), Dice Coefficient (DC) in binary masks; and Bland‐Altman and Pearson correlation by comparing FF values between manual and automated methods. Statistical Tests PA and DC were reported with their median value and standard deviation. Two methods were compared using the ± 95% confidence intervals (CI) of Bland‐Altman analysis and the Pearson’s coefficient (r2). Results DC values were from 0.83 ± 0.17 to 0.98 ± 0.02 in thigh and from 0.63 ± 0.18 to 0.96 ± 0.02 in calf muscles. For FF values, the overall ± 95% CI and r2 were [0.49, –0.56] and 0.989 in thigh and [0.84, –0.71] and 0.971 in the calf. Data Conclusion Automated results well agreed with the manual results in quantifying FF for individual muscles. This method mitigates the formidable time consumption and intense labor in manual segmentations; and enables the use of individual muscle FF as outcome measures in upcoming longitudinal studies. Level of Evidence 3 Technical Efficacy Stage 1

Automatic quadriceps and patellae segmentation of MRI with cascaded U2‐Net and SASSNet deep learning model

“…Moreover, imbalanced medical image data and high variability of target object shapes and locations often lead to unexpected segmentation results 57 . The anatomic symmetry and low distribution errors were ignored in previous models by a priori cropping 1,3,4,17–22,25,27 . The second refinement stage alleviates the memory bottleneck by focusing only on the cropped high‐resolution volume, provided by the low‐resolution first stage.…”

“…After an extensive review of the literature, 15 manuscripts 1–4,17–26 focusing on segmentation of the lower leg musculature were found (Table 1). Half are not comparable, as only the central part of the muscle was segmented 1,17,20,21,23,26 . Using just the central muscle belly inflates that accuracy as automatic segmentation is most error‐prone closest to its origins and insertions, where the muscle is rapidly changing size and shape (Figures 5 and 7).…”

“…A similar, but different issue is that of low‐distribution errors when the object of interest is represented in only a small percentage of the images. Both of these issues have typically been handled by manually cropping the image prior to the autosegmentation 1,3,4,17–22,25,27 . Lastly, as the resolution of MR images have rapidly improved, GPU bottlenecks have become an increasing problem 37 …”

“…As manual segmentation of individual muscles is time consuming,numerous semi-and fully automatic musclesegmentation are rapidly becoming available. [1][2][3][4] However, an interesting and more challenging problem has remained unanswered. These segmentation routines have rarely been developed or validated for the pediatric model.…”

“…Both of these issues have typically been handled by manually cropping the image prior to the autosegmentation. 1,3,4,[17][18][19][20][21][22]25,27 Lastly, as the resolution of MR images have rapidly improved, GPU bottlenecks have become an increasing problem. 37 To address the aforementioned problems, we set out to develop and validate a simple, yet effective cascaded U 2 -Net and SASSNet model for individual quadriceps muscle (in their entirety) segmentation on MR images, specifically targeting a pediatric cohort.…”