arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2104.09701eess.IVcs.CV
收藏

通过更丰富的生成对抗网络在计算机断层扫描图像中进行自由形式的肿瘤合成

Free-form tumor synthesis in computed tomography images via richer generative adversarial network

Tianjin University(天津大学);CSIRO Data61(澳大利亚联邦科学与工业研究组织Data61);La Trobe University(乐卓博大学);Tianjin University of Traditional Chinese Medicine(天津中医药大学)

Qiangguo Jin, Hui Cui, Changming Sun, Zhaopeng Meng, Ran Su

摘要:

针对癌症的注释医学成像扫描的不足使得在精确肿瘤学中训练和验证数据饥渴的深度学习模型变得具有挑战性。我们提出了一种新的更丰富的生成对抗网络,用于计算机断层扫描(CT)图像中的自由形式3D肿瘤/病变合成。该网络由一个新的更丰富的卷积特征增强型扩张门控生成器(RicherDG)和一个混合损失函数组成。RicherDG具有扩大门控卷积层,以实现肿瘤绘画并扩大感知范围;它具有一个新的更丰富的卷积特征关联分支,以恢复多尺度卷积特征,特别是从肿瘤和周围健康组织之间的不确定边界中恢复。混合损失函数由不同范围的损失组成,旨在聚合互补信息以改善优化。我们对涵盖肝脏、肾脏肿瘤和肺结节的广泛公共CT图像数据集的合成结果进行了全面评估。定性和定量评价以及消融研究表明,与先进的肿瘤合成方法相比,合成结果有所改善。

英文摘要:

The insufficiency of annotated medical imaging scans for cancer makes it challenging to train and validate data-hungry deep learning models in precision oncology. We propose a new richer generative adversarial network for free-form 3D tumor/lesion synthesis in computed tomography (CT) images. The network is composed of a new richer convolutional feature enhanced dilated-gated generator (RicherDG) and a hybrid loss function. The RicherDG has dilated-gated convolution layers to enable tumor-painting and to enlarge perceptive fields; and it has a novel richer convolutional feature association branch to recover multi-scale convolutional features especially from uncertain boundaries between tumor and surrounding healthy tissues. The hybrid loss function, which consists of a diverse range of losses, is designed to aggregate complementary information to improve optimization. We perform a comprehensive evaluation of the synthesis results on a wide range of public CT image datasets covering the liver, kidney tumors, and lung nodules. The qualitative and quantitative evaluations and ablation study demonstrated improved synthesizing results over advanced tumor synthesis methods.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑