Karimi Ali, Seraj Javad, Mirzadeh Sarcheshmeh Fatemeh, Fazli Kasra, Seraj Amirali, Eslami Parisa, Khanmohamadi Mohamadreza, Sajjadian Moosavi Helia, Ghattan Kashani Hadi, Sajjadian Moosavi Abdoulreza, Shariat Panahi Masoud
School of Electrical and Computer Engineering, College of Engineering, University of Tehran, Tehran, Iran.
School of Mechanical Engineering, College of Engineering, University of Tehran, Tehran, Iran.
Sci Rep. 2025 Jan 11;15(1):1670. doi: 10.1038/s41598-025-85632-9.
This paper introduces a novel method for spleen segmentation in ultrasound images, using a two-phase training approach. In the first phase, the SegFormerB0 network is trained to provide an initial segmentation. In the second phase, the network is further refined using the Pix2Pix structure, which enhances attention to details and corrects any erroneous or additional segments in the output. This hybrid method effectively combines the strengths of both SegFormer and Pix2Pix to produce highly accurate segmentation results. We have assembled the Spleenex dataset, consisting of 450 ultrasound images of the spleen, which is the first dataset of its kind in this field. Our method has been validated on this dataset, and the experimental results show that it outperforms existing state-of-the-art models. Specifically, our approach achieved a mean Intersection over Union (mIoU) of 94.17% and a mean Dice (mDice) score of 96.82%, surpassing models such as Splenomegaly Segmentation Network (SSNet), U-Net, and Variational autoencoder based methods. The proposed method also achieved a Mean Percentage Length Error (MPLE) of 3.64%, further demonstrating its accuracy. Furthermore, the proposed method has demonstrated strong performance even in the presence of noise in ultrasound images, highlighting its practical applicability in clinical environments.
本文介绍了一种用于超声图像中脾脏分割的新方法，该方法采用两阶段训练方法。在第一阶段，训练SegFormerB0网络以提供初始分割。在第二阶段，使用Pix2Pix结构对网络进行进一步优化，该结构增强了对细节的关注并纠正了输出中任何错误或额外的分割。这种混合方法有效地结合了SegFormer和Pix2Pix的优势，以产生高度准确的分割结果。我们已经组装了Spleenex数据集，该数据集由450张脾脏超声图像组成，这是该领域首个此类数据集。我们的方法已在该数据集上得到验证，实验结果表明它优于现有的最先进模型。具体而言，我们的方法实现了94.17%的平均交并比（mIoU）和96.82%的平均骰子系数（mDice）得分，超过了诸如脾肿大分割网络（SSNet）、U-Net和基于变分自编码器的方法等模型。所提出的方法还实现了3.64%的平均百分比长度误差（MPLE），进一步证明了其准确性。此外，即使在超声图像存在噪声的情况下，所提出的方法也表现出强大的性能，突出了其在临床环境中的实际适用性。
