当前位置: X-MOL 学术Med. Biol. Eng. Comput. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Superpixel-based deep convolutional neural networks and active contour model for automatic prostate segmentation on 3D MRI scans.
Medical & Biological Engineering & Computing ( IF 3.2 ) Pub Date : 2020-06-21 , DOI: 10.1007/s11517-020-02199-5
Giovanni L F da Silva 1 , Petterson S Diniz 1 , Jonnison L Ferreira 1 , João V F França 1 , Aristófanes C Silva 1 , Anselmo C de Paiva 1 , Elton A A de Cavalcanti 1
Affiliation  

Automatic and reliable prostate segmentation is an essential prerequisite for assisting the diagnosis and treatment, such as guiding biopsy procedure and radiation therapy. Nonetheless, automatic segmentation is challenging due to the lack of clear prostate boundaries owing to the similar appearance of prostate and surrounding tissues and the wide variation in size and shape among different patients ascribed to pathological changes or different resolutions of images. In this regard, the state-of-the-art includes methods based on a probabilistic atlas, active contour models, and deep learning techniques. However, these techniques have limitations that need to be addressed, such as MRI scans with the same spatial resolution, initialization of the prostate region with well-defined contours and a set of hyperparameters of deep learning techniques determined manually, respectively. Therefore, this paper proposes an automatic and novel coarse-to-fine segmentation method for prostate 3D MRI scans. The coarse segmentation step combines local texture and spatial information using the Intrinsic Manifold Simple Linear Iterative Clustering algorithm and probabilistic atlas in a deep convolutional neural networks model jointly with the particle swarm optimization algorithm to classify prostate and non-prostate tissues. Then, the fine segmentation uses the 3D Chan-Vese active contour model to obtain the final prostate surface. The proposed method has been evaluated on the Prostate 3T and PROMISE12 databases presenting a dice similarity coefficient of 84.86%, relative volume difference of 14.53%, sensitivity of 90.73%, specificity of 99.46%, and accuracy of 99.11%. Experimental results demonstrate the high performance potential of the proposed method compared to those previously published.



中文翻译:

基于超像素的深度卷积神经网络和主动轮廓模型,可在3D MRI扫描上自动进行前列腺分割。

自动可靠的前列腺分割术是辅助诊断和治疗(例如指导活检程序和放射治疗)的必要先决条件。然而,由于前列腺和周围组织的相似外观,以及由于病理改变或图像分辨率不同而导致的不同患者之间的大小和形状的广泛差异,由于缺乏清晰的前列腺边界,自动分割仍然具有挑战性。在这方面,最新技术包括基于概率图集的方法,主动轮廓模型和深度学习技术。但是,这些技术具有需要解决的局限性,例如具有相同空间分辨率的MRI扫描,用轮廓分明的轮廓和一组手动确定的深度学习技术的超参数分别对前列腺区域进行初始化。因此,本文提出了一种用于前列腺3D MRI扫描的自动新颖的从粗到细分割方法。粗糙分割步骤使用本征流形简单线性迭代聚类算法和概率图集结合深度卷积神经网络模型中的局部纹理和空间信息,结合粒子群优化算法对前列腺和非前列腺组织进行分类。然后,精细分割使用3D Chan-Vese活动轮廓模型获得最终的前列腺表面。该方法已在Prostate 3T和PROMISE12数据库中进行了评估,其骰子相似系数为84.86%,相对体积差异为14.53%,灵敏度为90.73%,特异性为99.46%,准确度为99.11%。实验结果表明,与先前发布的方法相比,该方法具有更高的性能潜力。

更新日期:2020-06-22
down
wechat
bug