arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2209.08256cs.CV
收藏

分割模型能否使用完全合成的数据进行训练?

Can segmentation models be trained with fully synthetically generated data?

King’s College London(伦敦国王学院)

Virginia Fernandez, Walter Hugo Lopez Pinaya, Pedro Borges, Petru-Daniel Tudosiu, Mark S Graham, Tom Vercauteren, M Jorge Cardoso

AI总结:

提出 brainSPADE 模型,结合基于扩散的标签生成器与语义图像生成器,按需生成含或不含病理的脑部标签及对应 MRI 图像,实验证明用其合成数据训练的分割模型性能与真实数据相当。

摘要:

为了实现良好的性能和泛化能力,医学图像分割模型应在具有充分变异性的相当大规模数据集上进行训练。由于伦理和治理限制,以及标注数据的相关成本,科学发展常常受到阻碍,模型只能在有限的数据上进行训练和测试。数据增强通常被用来人为增加数据分布中的变异性并提高模型的泛化能力。近期的研究探索了用于图像合成的深度生成模型,因为这种方法能够生成实际上无限量的多样化数据,从而解决泛化能力和数据访问问题。然而,许多提出的解决方案限制了用户对生成内容的控制。在这项工作中,我们提出了 brainSPADE,这是一种将基于扩散的合成标签生成器与语义图像生成器相结合的模型。我们的模型能够按需生成完全合成的脑部标签(包含或不包含感兴趣的病理),然后生成任意引导风格下的相应 MRI 图像。实验表明,brainSPADE 合成数据可用于训练分割模型,其性能与在真实数据上训练的模型相当。

英文摘要:

In order to achieve good performance and generalisability, medical image segmentation models should be trained on sizeable datasets with sufficient variability. Due to ethics and governance restrictions, and the costs associated with labelling data, scientific development is often stifled, with models trained and tested on limited data. Data augmentation is often used to artificially increase the variability in the data distribution and improve model generalisability. Recent works have explored deep generative models for image synthesis, as such an approach would enable the generation of an effectively infinite amount of varied data, addressing the generalisability and data access problems. However, many proposed solutions limit the user's control over what is generated. In this work, we propose brainSPADE, a model which combines a synthetic diffusion-based label generator with a semantic image generator. Our model can produce fully synthetic brain labels on-demand, with or without pathology of interest, and then generate a corresponding MRI image of an arbitrary guided style. Experiments show that brainSPADE synthetic data can be used to train segmentation models with performance comparable to that of models trained on real data.

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。

↑