Lymphoma is cancer originated in the lymphatic system. Clinically, automatic and accurate lymphoma segmentation is critical yet challenging. Recently, UNet-like architectures are widely used for medical image segmentation. The pure UNet-like architectures can model the spatial correlation between the feature maps very well, whereas they discard the critical temporal correlation. Some prior works combine UNet with recurrent neural networks (RNNs) to utilize the spatial and temporal correlation simultaneously. However, it is inconvenient to incorporate some advanced techniques proposed for UNet to RNNs, which hampers their further improvements. In this paper, we propose a recurrent dense siamese decoder architecture, which simulates RNNs and can densely utilize the spatial temporal correlation between the decoder feature maps following a “UNet” approach. We combine it with a modified hyper dense encoder. Therefore, the proposed model is a UNet with a hyper dense encoder and a recurrent dense siamese decoder (HD-RDS-UNet). To stabilize the training process, we propose a weighted Dice loss with stable gradient and self-adaptive parameters. We perform patient-independent five-fold cross-validation on 3D volumes collected from whole-body PET/CT scans of patients with lymphomas. The experimental results show that the volume-wise average Dice score and sensitivity are 85.58% and 94.63%, respectively. The patient-wise average Dice score and sensitivity are 85.85% and 95.01%, respectively. The different configurations of HD-RDS-UNet consistently show superiority in the performance comparison. Besides, a trained HD-RDS-UNet can be easily pruned, resulting in significantly reduced inference time and memory usage, while keeping very good segmentation performance.

中文翻译：

---

HD-RDS-UNet：利用解码器特征图之间的时空相关性进行淋巴瘤分割


淋巴瘤是起源于淋巴系统的癌症。临床上，自动、准确的淋巴瘤分割至关重要但具有挑战性。最近，类似 UNet 的架构被广泛用于医学图像分割。纯粹的类似 UNet 的架构可以很好地模拟特征图之间的空间相关性，而它们丢弃了关键的时间相关性。一些先前的工作将 UNet 与循环神经网络 (RNN) 结合起来，同时利用空间和时间相关性。然而，将 UNet 提出的一些先进技术结合到 RNN 中并不方便，这阻碍了它们的进一步改进。在本文中，我们提出了一种循环密集连体解码器架构，它模拟 RNN，并且可以按照“UNet”方法密集地利用解码器特征图之间的时空相关性。我们将其与改进的超密集编码器结合起来。因此，所提出的模型是具有超密集编码器和循环密集暹罗解码器的UNet（HD-RDS-UNet）。为了稳定训练过程，我们提出了具有稳定梯度和自适应参数的加权Dice损失。我们对从淋巴瘤患者全身 PET/CT 扫描中收集的 3D 体积进行独立于患者的五倍交叉验证。实验结果表明，体积平均Dice得分和灵敏度分别为85.58%和94.63%。患者平均 Dice 评分和敏感性分别为 85.85% 和 95.01%。 HD-RDS-UNet的不同配置在性能比较中始终表现出优越性。此外，经过训练的 HD-RDS-UNet 可以轻松修剪，从而显着减少推理时间和内存使用，同时保持非常好的分割性能。