[论文解读] Debiasing pipeline improves deep learning model generalization for X-ray based lung nodule detection

Michael J. Horry, Subrata Chakraborty+6 more|arXiv (Cornell University)|Jan 24, 2022

Radiomics and Machine Learning in Medical Imaging被引用 7

一句话总结

本文提出了一种针对胸部X光片的去偏倚预处理流程，可提升深度学习模型在肺结节检测中的泛化能力。通过应用直方图均衡化、肺野分割、紧密裁剪和肋骨抑制，该方法提升了模型在未见数据上的外部泛化性能，在独立测试集上达到89%的准确率，从而实现了低成本、保护隐私的农村人群筛查。

ABSTRACT

Lung cancer is the leading cause of cancer death worldwide and a good prognosis depends on early diagnosis. Unfortunately, screening programs for the early diagnosis of lung cancer are uncommon. This is in-part due to the at-risk groups being located in rural areas far from medical facilities. Reaching these populations would require a scaled approach that combines mobility, low cost, speed, accuracy, and privacy. We can resolve these issues by combining the chest X-ray imaging mode with a federated deep-learning approach, provided that the federated model is trained on homogenous data to ensure that no single data source can adversely bias the model at any point in time. In this study we show that an image pre-processing pipeline that homogenizes and debiases chest X-ray images can improve both internal classification and external generalization, paving the way for a low-cost and accessible deep learning-based clinical system for lung cancer screening. An evolutionary pruning mechanism is used to train a nodule detection deep learning model on the most informative images from a publicly available lung nodule X-ray dataset. Histogram equalization is used to remove systematic differences in image brightness and contrast. Model training is performed using all combinations of lung field segmentation, close cropping, and rib suppression operators. We show that this pre-processing pipeline results in deep learning models that successfully generalize an independent lung nodule dataset using ablation studies to assess the contribution of each operator in this pipeline. In stripping chest X-ray images of known confounding variables by lung field segmentation, along with suppression of signal noise from the bone structure we can train a highly accurate deep learning lung nodule detection algorithm with outstanding generalization accuracy of 89% to nodule samples in unseen data.

研究动机与目标

• 为解决深度学习模型在胸部X光片肺结节检测中泛化能力差的挑战。
• 减少因成像差异（如亮度、对比度、肋骨伪影）在不同来源数据中引入的数据偏差。
• 实现在农村及资源匮乏人群中的可扩展、低成本、保护隐私的肺癌筛查。
• 通过系统化的图像预处理提升模型在未见数据集上的鲁棒性。
• 验证各个预处理组件在提升模型泛化能力方面的贡献。

提出的方法

• 应用直方图均衡化以标准化不同数据集间图像的亮度和对比度。
• 使用肺野分割技术提取感兴趣区域并去除无关解剖结构。
• 实施紧密裁剪以聚焦肺野区域并减少背景噪声。
• 应用肋骨抑制以最小化骨骼结构对结节检测的干扰。
• 将上述预处理步骤整合为一个流程，并使用所有操作组合训练深度学习模型。
• 采用进化剪枝机制选择最具信息量的图像用于模型训练。

实验结果

研究问题

• RQ1 是否系统化的预处理流程能通过去偏倚胸部X光片显著提升深度学习模型在未见数据上的泛化能力？
• RQ2 预处理流程中的哪些独立组件（如肋骨抑制、直方图均衡化）对泛化能力提升贡献最大？
• RQ3 在单一数据集上训练的去偏倚模型能否有效泛化到独立的外部测试集？
• RQ4 图像均一化与选择性图像剪枝的结合如何影响检测准确率和鲁棒性？
• RQ5 该流程在多大程度上可支持联邦学习，以实现在农村医疗环境中的隐私保护部署？

主要发现

• 去偏倚流程显著提升了模型泛化能力，在独立未见的肺结节数据集上达到89%的准确率。
• 消融实验确认，每个组件（尤其是肋骨抑制和直方图均衡化）均对性能提升有显著贡献。
• 肺野分割和紧密裁剪降低了模型对无关解剖变异和背景噪声的敏感性。
• 进化剪枝机制有效筛选出信息量丰富的图像，提升了训练效率和模型性能。
• 多种预处理技术的组合优于在原始或轻度处理图像上训练的基线模型。
• 该流程通过确保跨多样化数据源的数据同质性并减少偏差，支持在联邦学习环境中的部署。