文献类型：期刊文献

英文题名：Automatic segmentation of prostate MRI based on 3D pyramid pooling Unet

作者：Li, Yuchun[1];Lin, Cong[1,4];Zhang, Yu[2];Feng, Siling[1,5];Huang, Mengxing[1,5];Bai, Zhiming[3]

机构：[1]Hainan Univ, Sch informat & Commun Engn, State Key Lab Marine Resource Utilizat South China, Haikou, Peoples R China;[2]Hainan Univ, Coll Comp Sci & Technol, Haikou, Peoples R China;[3]Cent South Univ, Haikou Municipal Peoples Hosp, Affiliated Hosp, Xiangya Med Coll, Haikou, Peoples R China;[4]Guangdong Ocean Univ, Coll Elect & Informat Engn, Zhanjiang, Peoples R China;[5]Hainan Univ, Sch informat & Commun Engn, State Key Lab Marine Resource Utilizat South China, Haikou 570288, Peoples R China

年份：2023

卷号：50

期号：2

起止页码：906

外文期刊名：MEDICAL PHYSICS

收录：SCI-EXPANDED(收录号：WOS:000905856800001)、、EI(收录号：20224713145318)、Scopus(收录号：2-s2.0-85142171034)、WOS

基金：Hainan Provincial Natural Science Foundation of China, Grant/Award Number: 621MS019; Key R&D Project of Hainan, Grant/Award Number: ZDYF2021SHFZ243; Major Science and Technology Project of Haikou, rant/Award Number: 2020-009; Innovative Research Project of Postgraduates, Grant/Award Number: Qhyb2021-09;Innovative Research Project of Postgraduates, Grant/Award Number: Qhyb2021-10; National Natural Science Foundation of China, Grant/Award Number:82260362; National Key R&D Program of China, Grant/Award Number: 2021ZD0111000

语种：英文

外文关键词：3D Unet; deep supervision; MRI; prostate segmentation; pyramid pool

外文摘要：PurposeAutomatic segmentation of prostate magnetic resonance (MR) images is crucial for the diagnosis, evaluation, and prognosis of prostate diseases (including prostate cancer). In recent years, the mainstream segmentation method for the prostate has been converted to convolutional neural networks. However, owing to the complexity of the tissue structure in MR images and the limitations of existing methods in spatial context modeling, the segmentation performance should be improved further. MethodsIn this study, we proposed a novel 3D pyramid pool Unet that benefits from the pyramid pooling structure embedded in the skip connection (SC) and the deep supervision (DS) in the up-sampling of the 3D Unet. The parallel SC of the conventional 3D Unet network causes low-resolution information to be sent to the feature map repeatedly, resulting in blurred image features. To overcome the shortcomings of the conventional 3D Unet, we merge each decoder layer with the feature map of the same scale as the encoder and the smaller scale feature map of the pyramid pooling encoder. This SC combines the low-level details and high-level semantics at two different levels of feature maps. In addition, pyramid pooling performs multifaceted feature extraction on each image behind the convolutional layer, and DS learns hierarchical representations from comprehensive aggregated feature maps, which can improve the accuracy of the task. ResultsExperiments on 3D prostate MR images of 78 patients demonstrated that our results were highly correlated with expert manual segmentation. The average relative volume difference and Dice similarity coefficient of the prostate volume area were 2.32% and 91.03%, respectively. ConclusionQuantitative experiments demonstrate that, compared with other methods, the results of our method are highly consistent with the expert manual segmentation.