Dong Kaiqi, Hu Peijun, Zhu Yan, Tian Yu, Li Xiang, Zhou Tianshu, Bai Xueli, Liang Tingbo, Li Jingsong
Engineering Research Center of EMR and Intelligent Expert System, Ministry of Education, College of Biomedical Engineering and Instrument Science, Zhejiang University, Hangzhou, China.
Research Center for Data Hub and Security, Zhejiang Laboratory, Hangzhou, China.
Med Phys. 2024 Dec;51(12):8999-9016. doi: 10.1002/mp.17385. Epub 2024 Sep 22.
Accurate pancreas and pancreatic tumor segmentation from abdominal scans is crucial for diagnosing and treating pancreatic diseases. Automated and reliable segmentation algorithms are highly desirable in both clinical practice and research.
Segmenting the pancreas and tumors is challenging due to their low contrast, irregular morphologies, and variable anatomical locations. Additionally, the substantial difference in size between the pancreas and small tumors makes this task difficult. This paper proposes an attention-enhanced multiscale feature fusion network (AMFF-Net) to address these issues via 3D attention and multiscale context fusion methods.
First, to prevent missed segmentation of tumors, we design the residual depthwise attention modules (RDAMs) to extract global features by expanding receptive fields of shallow layers in the encoder. Second, hybrid transformer modules (HTMs) are proposed to model deep semantic features and suppress irrelevant regions while highlighting critical anatomical characteristics. Additionally, the multiscale feature fusion module (MFFM) fuses adjacent top and bottom scale semantic features to address the size imbalance issue.
The proposed AMFF-Net was evaluated on the public MSD dataset, achieving 82.12% DSC for pancreas and 57.00% for tumors. It also demonstrated effective segmentation performance on the NIH and private datasets, outperforming previous State-Of-The-Art (SOTA) methods. Ablation studies verify the effectiveness of RDAMs, HTMs, and MFFM.
We propose an effective deep learning network for pancreas and tumor segmentation from abdominal CT scans. The proposed modules can better leverage global dependencies and semantic information and achieve significantly higher accuracy than the previous SOTA methods.
从腹部扫描中准确分割胰腺和胰腺肿瘤对于胰腺疾病的诊断和治疗至关重要。在临床实践和研究中，自动化且可靠的分割算法都非常必要。
由于胰腺和肿瘤对比度低、形态不规则以及解剖位置多变，对其进行分割具有挑战性。此外，胰腺和小肿瘤之间的尺寸差异巨大，使得这项任务变得困难。本文提出了一种注意力增强多尺度特征融合网络（AMFF-Net），通过三维注意力和多尺度上下文融合方法来解决这些问题。
首先，为防止肿瘤分割遗漏，我们设计了残差深度注意力模块（RDAM），通过扩展编码器中浅层的感受野来提取全局特征。其次，提出了混合变压器模块（HTM）来对深度语义特征进行建模，并抑制无关区域，同时突出关键的解剖特征。此外，多尺度特征融合模块（MFFM）融合相邻的上下尺度语义特征，以解决尺寸不平衡问题。
所提出的AMFF-Net在公开的MSD数据集上进行了评估，胰腺分割的DSC达到82.12%，肿瘤分割的DSC达到57.00%。它在NIH和私有数据集上也展示了有效的分割性能，优于先前的最先进（SOTA）方法。消融研究验证了RDAM、HTM和MFFM的有效性。
我们提出了一种用于从腹部CT扫描中分割胰腺和肿瘤的有效深度学习网络。所提出的模块能够更好地利用全局依赖性和语义信息，并且比先前的SOTA方法实现了显著更高的准确率。
