Yan Chaoyang, Lu Jing-Jing, Chen Kang, Wang Lei, Lu Haoda, Yu Li, Sun Mengyan, Xu Jun
Institute for AI in Medicine, School of Automation, Nanjing University of Information Science and Technology, Nanjing, China.
Department of Radiology, Beijing United Family Hospital, Beijing, China.
Magn Reson Med. 2022 Jan;87(1):431-445. doi: 10.1002/mrm.28939. Epub 2021 Aug 2.
MRI of organs and musculoskeletal structures in the female pelvis presents a unique display of pelvic anatomy. Automated segmentation of pelvic structures plays an important role in personalized diagnosis and treatment on pelvic structures disease. Pelvic organ systems are very complicated, and it is a challenging task for 3D segmentation of massive pelvic structures on MRI.
A new Scale- and Slice-aware Net ( aNet) is presented for 3D dense segmentation of 54 organs and musculoskeletal structures in female pelvic MR images. A Scale-aware module is designed to capture the spatial and semantic information of different-scale structures. A Slice-aware module is introduced to model similar spatial relationships of consecutive slices in 3D data. Moreover, aNet leverages a weight-adaptive loss optimization strategy to reinforce the supervision with more discriminative capability on hard samples and categories.
Experiments have been performed on a pelvic MRI cohort of 27 MR images from 27 patient cases. Across the cohort and 54 categories of organs and musculoskeletal structures manually delineated, aNet was shown to outperform the UNet framework and other state-of-the-art fully convolutional networks in terms of sensitivity, Dice similarity coefficient and relative volume difference.
The experimental results on the pelvic 3D MR dataset show that the proposed aNet achieves excellent segmentation results compared to other state-of-the-art models. To our knowledge, aNet is the first model to achieve 3D dense segmentation for 54 musculoskeletal structures on pelvic MRI, which will be leveraged to the clinical application under the support of more cases in the future.
女性盆腔器官和肌肉骨骼结构的磁共振成像（MRI）呈现出独特的盆腔解剖结构显示。盆腔结构的自动分割在盆腔结构疾病的个性化诊断和治疗中起着重要作用。盆腔器官系统非常复杂，对MRI上大量盆腔结构进行三维分割是一项具有挑战性的任务。
提出了一种新的尺度和切片感知网络（aNet），用于对女性盆腔MR图像中的54个器官和肌肉骨骼结构进行三维密集分割。设计了一个尺度感知模块来捕捉不同尺度结构的空间和语义信息。引入了一个切片感知模块来对三维数据中连续切片的相似空间关系进行建模。此外，aNet利用权重自适应损失优化策略，加强对硬样本和类别具有更强判别能力的监督。
对来自27例患者的27幅MR图像的盆腔MRI队列进行了实验。在整个队列以及手动勾勒的54类器官和肌肉骨骼结构中，aNet在灵敏度、骰子相似系数和相对体积差异方面优于UNet框架和其他先进的全卷积网络。
盆腔三维MR数据集的实验结果表明，与其他先进模型相比，所提出的aNet取得了优异的分割结果。据我们所知，aNet是第一个在盆腔MRI上对54个肌肉骨骼结构实现三维密集分割的模型，未来在更多病例的支持下，它将被应用于临床。
