Performance of the SDTNet on DS2.

Accurate segmentation of lung lesions in CT-scan images is essential to diagnose lung cancer. The challenges in lung nodule diagnosis arise due to their small size and diverse nature. We designed a transformer-based model EDTNet (Encoder Decoder Transformer Network) for PNS (Pulmonary Nodule Segmentation). Traditional CNN-based encoders and decoders are hindered by their inability to capture long-range spatial dependencies, leading to suboptimal performance in complex object segmentation tasks. To address the limitation, we leverage an enhanced spatial attention-based Vision Transformer (ViT) as an encoder and decoder in the EDTNet. The EDTNet integrates two successive transformer blocks, a patch-expanding layer, down-sampling layers, and up-sampling layers to improve segmentation capabilities. In addition, ESLA (Enhanced spatial aware local attention) and EGLA (Enhanced global aware local attention) blocks are added to provide attention to the spatial features. Furthermore, skip connections are introduced to facilitate symmetrical interaction between the corresponding encoder and decoder layer, enabling the retrieval of intricate details in the output. The EDTNet performance is compared with several models on DS1 and DS2, including Unet, ResUNet++, U-NET 3+, DeepLabV3+, SegNet, Trans-Unet, and Swin-UNet, demonstrates superior quantitative and visual results. On DS1, the EDTNet achieved 96.27%, 95.81%, 96.15% precision, IoU (Intersection over Union), and DSC (Sorensen–Dice coefficient). Moreover, the model has demonstrated sensitivity, IoU and SDC of 98.84%, 96.06% and 97.85% on DS2.

CT扫描影像中的肺部病变精准分割是肺癌诊断的关键环节。肺结节诊断面临的挑战源于其体积微小且形态异质性强。本研究设计了一款基于Transformer的EDTNet（Encoder Decoder Transformer Network，编码器-解码器Transformer网络）模型，用于肺结节分割（Pulmonary Nodule Segmentation，PNS）任务。传统基于卷积神经网络（Convolutional Neural Network, CNN）的编码器与解码器无法捕获长程空间依赖关系，在复杂目标分割任务中性能欠佳。为解决这一局限，本研究在EDTNet中采用增强型空间注意力视觉Transformer（Vision Transformer, ViT）作为编码器与解码器模块。EDTNet整合了连续两个Transformer模块、一个图像块扩展层、下采样层与上采样层，以提升分割性能。此外，本研究还加入了ESLA（Enhanced spatial aware local attention，增强型空间感知局部注意力）模块与EGLA（Enhanced global aware local attention，增强型全局感知局部注意力）模块，以对空间特征施加注意力机制。进一步地，模型引入跳跃连接以实现对应编码器与解码器层之间的对称交互，从而能够还原输出结果中的精细细节。本研究在DS1与DS2两个数据集上，将EDTNet与Unet、ResUNet++、U-NET 3+、DeepLabV3+、SegNet、Trans-Unet以及Swin-UNet等多款模型进行对比，结果显示EDTNet在量化指标与可视化效果上均表现更优。在DS1数据集上，EDTNet的精确率、交并比（Intersection over Union, IoU）与索伦森-迪克森系数（Sorensen–Dice coefficient, DSC）分别达到96.27%、95.81%与96.15%。此外，该模型在DS2数据集上的灵敏度、交并比与SDC分别达到98.84%、96.06%与97.85%。