Department of Automatic Control and Systems Engineering, University of Sheffield, Sheffield, United Kingdom.
Division of Clinical Medicine, University of Sheffield, Sheffield, United Kingdom.
PLoS One. 2024 Oct 4;19(10):e0308664. doi: 10.1371/journal.pone.0308664. eCollection 2024.
This study aims at improving the lower-limb muscle segmentation accuracy of deep learning approaches based on Magnetic Resonance Imaging (MRI) scans, crucial for the diagnostic and therapeutic processes in musculoskeletal diseases. In general, segmentation methods such as U-Net deep learning neural networks can achieve good Dice Similarity Coefficient (DSC) values, e.g. around 0.83 to 0.91 on various cohorts. Some generic post-processing strategies have been studied to incorporate connectivity constraints into the resulting masks for the purpose of further improving the segmentation accuracy. In this paper, a novel mean shape (MS) based post-processing method is proposed, utilizing Statistical Shape Modelling (SSM) to fine-tune the segmentation output, taking into consideration the muscle anatomical shape. The methodology was compared to existing post-processing techniques and a commercial semi-automatic tool on MRI scans from two cohorts of post-menopausal women (10 Training, 8 Testing, voxel size 1.0x1.0x1.0 mm3). The MS based method obtained a mean DSC of 0.83 across the different analysed muscles and the best performance for the Hausdorff Distance (HD, 20.6 mm) and the Average Symmetric Surface Distance (ASSD, 2.1 mm). These findings highlight the feasibility and potential of using anatomical mean shape in post-processing of human lower-limb muscle segmentation task and indicate that the proposed method can be popularized to other biological organ segmentation mission.
本研究旨在提高基于磁共振成像（MRI）扫描的深度学习方法在肌肉骨骼疾病的诊断和治疗过程中对下肢肌肉分割的准确性。通常，基于 U-Net 等深度学习神经网络的分割方法可以实现较好的 Dice 相似系数（DSC）值，例如在各种队列中约为 0.83 到 0.91。已经研究了一些通用的后处理策略，以便将连通性约束纳入到所得掩模中，以进一步提高分割准确性。在本文中，提出了一种新的基于均值形状（MS）的后处理方法，利用统计形状建模（SSM）来微调分割输出，考虑到肌肉解剖形状。该方法与现有的后处理技术和商业半自动工具在两个绝经后妇女队列的 MRI 扫描上进行了比较（10 个训练集，8 个测试集，体素大小为 1.0x1.0x1.0mm3）。基于 MS 的方法在不同分析的肌肉中获得了 0.83 的平均 DSC，在 Hausdorff 距离（HD，20.6mm）和平均对称表面距离（ASSD，2.1mm）方面取得了最佳性能。这些发现突出了在人体下肢肌肉分割任务的后处理中使用解剖平均形状的可行性和潜力，并表明所提出的方法可以推广到其他生物器官分割任务中。
