Yuan Han, Hong Chuan, Tran Nguyen Tuan Anh, Xu Xinxing, Liu Nan
Centre for Quantitative Medicine, Duke-NUS Medical School Singapore.
Department of Biostatistics and Bioinformatics Duke University Durham North Carolina USA.
Health Care Sci. 2024 Dec 15;3(6):456-474. doi: 10.1002/hcs2.119. eCollection 2024 Dec.
Pneumothorax is a medical emergency caused by the abnormal accumulation of air in the pleural space-the potential space between the lungs and chest wall. On 2D chest radiographs, pneumothorax occurs within the thoracic cavity and outside of the mediastinum, and we refer to this area as "lung + space." While deep learning (DL) has increasingly been utilized to segment pneumothorax lesions in chest radiographs, many existing DL models employ an end-to-end approach. These models directly map chest radiographs to clinician-annotated lesion areas, often neglecting the vital domain knowledge that pneumothorax is inherently location-sensitive.
We propose a novel approach that incorporates the lung + space as a constraint during DL model training for pneumothorax segmentation on 2D chest radiographs. To circumvent the need for additional annotations and to prevent potential label leakage on the target task, our method utilizes external datasets and an auxiliary task of lung segmentation. This approach generates a specific constraint of lung + space for each chest radiograph. Furthermore, we have incorporated a discriminator to eliminate unreliable constraints caused by the domain shift between the auxiliary and target datasets.
Our results demonstrated considerable improvements, with average performance gains of 4.6%, 3.6%, and 3.3% regarding intersection over union, dice similarity coefficient, and Hausdorff distance. These results were consistent across six baseline models built on three architectures (U-Net, LinkNet, or PSPNet) and two backbones (VGG-11 or MobileOne-S0). We further conducted an ablation study to evaluate the contribution of each component in the proposed method and undertook several robustness studies on hyper-parameter selection to validate the stability of our method.
The integration of domain knowledge in DL models for medical applications has often been underemphasized. Our research underscores the significance of incorporating medical domain knowledge about the location-specific nature of pneumothorax to enhance DL-based lesion segmentation and further bolster clinicians' trust in DL tools. Beyond pneumothorax, our approach is promising for other thoracic conditions that possess location-relevant characteristics.
气胸是一种医疗急症，由空气在胸膜腔（肺与胸壁之间的潜在腔隙）异常积聚所致。在二维胸部X光片上，气胸出现在胸腔内且在纵隔之外，我们将此区域称为“肺+间隙”。虽然深度学习（DL）越来越多地用于在胸部X光片中分割气胸病变，但许多现有的深度学习模型采用端到端方法。这些模型直接将胸部X光片映射到临床医生标注的病变区域，常常忽略了气胸本质上对位置敏感这一重要领域知识。
我们提出了一种新颖的方法，在二维胸部X光片上进行气胸分割的深度学习模型训练期间，将“肺+间隙”作为一种约束纳入其中。为了避免额外标注的需求并防止目标任务上的潜在标签泄露，我们的方法利用外部数据集和肺分割的辅助任务。这种方法为每张胸部X光片生成“肺+间隙”的特定约束。此外，我们引入了一个判别器来消除由辅助数据集和目标数据集之间的域转移引起的不可靠约束。
我们的结果显示出显著改进，在交并比、骰子相似系数和豪斯多夫距离方面平均性能提升分别为4.6%、3.6%和3.3%。这些结果在基于三种架构（U-Net、LinkNet或PSPNet）和两种主干网络（VGG-11或MobileOne-S0）构建的六个基线模型中是一致的。我们进一步进行了消融研究以评估所提方法中每个组件的贡献，并对超参数选择进行了多项稳健性研究以验证我们方法的稳定性。
在医疗应用的深度学习模型中，领域知识的整合常常未得到充分重视。我们的研究强调了纳入关于气胸位置特异性的医学领域知识对于增强基于深度学习的病变分割以及进一步提升临床医生对深度学习工具信任的重要性。除气胸外，我们的方法对于其他具有位置相关特征的胸部疾病也很有前景。
