Wang Shuhuan, Li Zhiqing, Liao Lingmin, Zhang Chunquan, Zhao Jiali, Sang Liang, Qian Wei, Pan GuangYao, Huang Long, Ma He
College of Medicine and Biological Information Engineering, Northeastern University, Shenyang, Liaoning 110169, People's Republic of China.
Department of Ultrasound, The Second Affiliated Hospital of Nanchang University, No. 1 Minde Road, Nanchang, Jiangxi, People's Republic of China.
Phys Med Biol. 2023 Jul 31;68(16). doi: 10.1088/1361-6560/ace6f1.
Deep learning has demonstrated its versatility in the medical field, particularly in medical image segmentation, image classification, and other forms of automated diagnostics. The clinical diagnosis of thyroid nodules requires radiologists to locate nodules, diagnose conditions based on nodule boundaries, textures and their experience. This task is labor-intensive and tiring; therefore, an automated system for accurate thyroid nodule segmentation is essential. In this study, a model named DPAM-PSPNet was proposed, which automatically segments nodules in thyroid ultrasound images and enables to segment malignant nodules precisely.In this paper, accurate segmentation of nodule edges is achieved by introducing the dual path attention mechanism (DPAM) in PSPNet. In one channel, it captures global information with a lightweight cross-channel interaction mechanism. In other channel, it focus on nodal margins and surrounding information through the residual bridge network. We also updated the integrated loss function to accommodate the DPAM-PSPNet.The DPAM-PSPNet was tested against the classical segmentation model. Ablation experiments were designed for the two-path attention mechanism and the new loss function, and generalization experiments were designed on the public dataset. Our experimental results demonstrate that DPAM-PSPNet outperforms other existing methods in various evaluation metrics. In the model comparison experiments, it achieved performance with an mIOU of 0.8675, mPA of 0.9357, mPrecision of 0.9202, and Dice coefficient of 0.9213.The DPAM-PSPNet model can segment thyroid nodules in ultrasound images with little training data and generate accurate boundary regions for these nodules.
深度学习已在医学领域展现出其多功能性，尤其是在医学图像分割、图像分类及其他形式的自动诊断方面。甲状腺结节的临床诊断需要放射科医生定位结节，并根据结节边界、纹理及其经验来诊断病情。这项任务既耗费人力又令人疲惫；因此，一个用于准确分割甲状腺结节的自动化系统至关重要。在本研究中，提出了一种名为DPAM - PSPNet的模型，它能自动分割甲状腺超声图像中的结节，并能够精确分割恶性结节。本文通过在PSPNet中引入双路径注意力机制（DPAM）实现了结节边缘的精确分割。在一个通道中，它通过轻量级跨通道交互机制捕获全局信息。在另一个通道中，它通过残差桥接网络关注结节边缘和周围信息。我们还更新了综合损失函数以适应DPAM - PSPNet。将DPAM - PSPNet与经典分割模型进行了测试。针对双路径注意力机制和新的损失函数设计了消融实验，并在公共数据集上设计了泛化实验。我们的实验结果表明，在各种评估指标中，DPAM - PSPNet优于其他现有方法。在模型比较实验中，它的mIOU为0.8675、mPA为0.9357、mPrecision为0.9202、Dice系数为0.9213。DPAM - PSPNet模型能够在训练数据较少的情况下分割超声图像中的甲状腺结节，并为这些结节生成准确的边界区域。
