A Knowledge-Based Modality-Independent Technique for Concurrent Thigh Muscle Segmentation: Applicable to CT and MR Images

Abstract

The mass of the lower extremity muscles is a clinically significant metric. Manual segmentation of these muscles is a timeconsuming task. Most of the segmentation methods for the thigh muscles are based on statistical models and atlases which need manually segmented datasets. The goal of this work is an automatic segmentation of the thigh muscles with only one initial segmented slice. A new automatic method is proposed for concurrent individual thigh muscles segmentation using a hybrid level set method and anatomical information of the muscles. In the proposed method, the muscle regions are extracted by the Fast and Robust Fuzzy C-Means Clustering (FRFCM) method, and then a contour is determined for each muscle which changes according to the muscle shape variation through its length. The anatomical information is used to control the contours variations and to refine the final boundaries. The method was validated by 22 CT datasets. The average dice similarity coefficient (DSC) of the method for individual muscle segmentation with one and two initial slices were 89.29 ± 2.59 (%) and 91.77 ± 1.87 (%), respectively. Also, the average symmetric surface distances (ASSDs) were 0.93 ± 0.29 mm and 0.64 ± 0.18 mm. Furthermore, applying to ten MRI datasets, the average DSC and ASSD for muscles were 90.9 ± 2.61 (%) and 0.71 ± 0.33 mm, respectively. The quantitative and intuitive results of the proposed method show the effectiveness of this method in segmentation of large and small muscles in CT and MR images. The consumed computation time is lower than the previous works, and this method does not need any training datasets.

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.…”

“…57 The anatomic symmetry and low distribution errors were ignored in previous models by a priori cropping. 1,3,4,[17][18][19][20][21][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. Ni et al 3 employed a similar two-stage cascaded 3D U-Net.…”

“…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). Further, most research questions based on muscle segmentation require the segmentation of the entire muscle.…”

“…When full volumes of individual muscles are quantified, the scope has been limited to small databases, 9,11,12,14,15 examining total limb or muscle group volume, 9,15 and typically to a single muscle. 14 Conventional auto-and semiautomatic segmentation methods (Table 1) for the leg musculature rely heavily on traditional algorithmic approaches, specifically image registration [17][18][19][20][21][22] and atlas-based methods. 4,[23][24][25] The ambiguous boundaries, caused by homogeneity texture appearance, is a key source of error across all these algorithms.…”

Image Segmentation with Adaptive Spatial Priors from Joint Registration

“…One kind of approaches is semiautomatic methods (e.g. [1,16,29,28,26]), that obtain priors by the manual delineation of lines, polygons, or manual segmentation of beginning slices, and so on. These semi-automatic methods have proven some efficiency but need manual intervention that can be time consuming and tedious.…”