Lin Jian, Yu Xiaomin, Wang Zhengxian, Ma Chaoqiong
Sichuan Academy of Medical Science and Sichuan Provincial People's Hospital, Chengdu, China.
School of Electronic Engineering, Chengdu University of Information Technology, Chengdu, China.
PLoS One. 2024 Dec 27;19(12):e0309118. doi: 10.1371/journal.pone.0309118. eCollection 2024.
In unsupervised transfer learning for medical image segmentation, where existing algorithms face the challenge of error propagation due to inaccessible source domain data. In response to this scenario, source-free domain transfer algorithm with reduced style sensitivity (SFDT-RSS) is designed. SFDT-RSS initially pre-trains the source domain model by using the generalization strategy and subsequently adapts the pre-trained model to target domain without accessing source data. Then, SFDT-RSS conducts interpatch style transfer (ISS) strategy, based on self-training with Transformer architecture, to minimize the pre-trained model's style sensitivity, enhancing its generalization capability and reducing reliance on a single image style. Simultaneously, the global perception ability of the Transformer architecture enhances semantic representation to improve style generalization effectiveness. In the domain transfer phase, the proposed algorithm utilizes a model-agnostic adaptive confidence regulation (ACR) loss to adjust the source model. Experimental results on five publicly available datasets for unsupervised cross-domain organ segmentation demonstrate that compared to existing algorithms, SFDT-RSS achieves segmentation accuracy improvements of 2.83%, 2.64%, 3.21%, 3.01%, and 3.32% respectively.
在医学图像分割的无监督迁移学习中，现有算法面临因源域数据不可获取而导致的误差传播挑战。针对这种情况，设计了具有降低风格敏感性的无源域迁移算法（SFDT-RSS）。SFDT-RSS最初通过使用泛化策略对源域模型进行预训练，随后在不访问源数据的情况下将预训练模型适配到目标域。然后，SFDT-RSS基于Transformer架构的自训练进行补丁间风格迁移（ISS）策略，以最小化预训练模型的风格敏感性，增强其泛化能力并减少对单一图像风格的依赖。同时，Transformer架构的全局感知能力增强了语义表示，以提高风格泛化效果。在域迁移阶段，所提出的算法利用模型无关的自适应置信度调节（ACR）损失来调整源模型。在五个用于无监督跨域器官分割的公开可用数据集上的实验结果表明，与现有算法相比，SFDT-RSS分别实现了2.83%、2.64%、3.21%、3.01%和3.32%的分割精度提升。
