In recent years, numerous neural network architectures specifically designed for the instance segmentation of nuclei in microscopic images have been released. These models embed nuclei-specific priors to outperform generic architectures like U-Nets; however, they require large annotated datasets, which are often not available. Generative models (GANs, diffusion models) have been used to compensate for this by synthesizing training data. These two-stage approaches are computationally expensive, as first a generative model and then a segmentation model has to be trained. We propose CyclePose, a hybrid framework integrating synthetic data generation and segmentation training. CyclePose builds on a CycleGAN architecture, which allows unpaired translation between microscopy images and segmentation masks. We embed a segmentation model into CycleGAN and leverage a cycle consistency loss for self-supervision. Without annotated data, CyclePose outperforms other weakly or unsupervised methods on two public datasets. Code is available at https://github.com/jonasutz/CyclePose

翻译：近年来，已发布多种专门用于显微图像中细胞核实例分割的神经网络架构。这些模型嵌入了细胞核特异性先验知识，其性能超越了U-Net等通用架构；然而，它们需要大量标注数据集，而这些数据往往难以获得。生成模型（GAN、扩散模型）已被用于通过合成训练数据来弥补这一缺陷。这类两阶段方法计算成本高昂，因为需要先后训练生成模型和分割模型。我们提出CyclePose——一个集成合成数据生成与分割训练的混合框架。CyclePose基于CycleGAN架构构建，该架构支持显微图像与分割掩码之间的非配对转换。我们将分割模型嵌入CycleGAN，并利用循环一致性损失进行自监督。在无需标注数据的情况下，CyclePose在两个公开数据集上的表现优于其他弱监督或无监督方法。代码发布于https://github.com/jonasutz/CyclePose