用自编码器从少量肺部CT中自动学特征，提升间质性肺病诊断能力
Unmasking Interstitial Lung Diseases: Leveraging Masked Autoencoders for Diagnosis
- 用5000+张肺CT训练掩码自编码器，无监督学习影像特征
- 在小样本标注数据下，诊断准确率显著优于传统方法
- 适合医学影像少样本场景，尤其对罕见病诊断有实用价值
掩码自编码器（MAE）是一种强大的无标签数据预训练方法，能够学习鲁棒且信息丰富的特征表示。这在弥漫性肺疾病研究中尤为有利，因为标注的影像数据集稀缺。为此，我们在一个包含超过5000张胸部计算机断层扫描（CT）的精选数据集上训练了MAE，该数据集结合了内部数据和具有相似影像模式的公开数据（如新冠和细菌性肺炎）。随后，将预训练的MAE微调用于弥漫性肺疾病诊断的下游分类任务。结果表明，即使缺乏大规模标注数据，MAE也能有效提取临床相关的特征并提升诊断性能。代码与模型已开源：https://github.com/eedack01/lung_masked_autoencoder。
原文摘要 · Abstract (English)
Masked autoencoders (MAEs) have emerged as a powerful approach for pre-training on unlabelled data, capable of learning robust and informative feature representations. This is particularly advantageous in diffused lung disease research, where annotated imaging datasets are scarce. To leverage this, we train an MAE on a curated collection of over 5,000 chest computed tomography (CT) scans, combining in-house data with publicly available scans from related conditions that exhibit similar radiological patterns, such as COVID-19 and bacterial pneumonia. The pretrained MAE is then fine-tuned on a downstream classification task for diffused lung disease diagnosis. Our findings demonstrate that MAEs can effectively extract clinically meaningful features and improve diagnostic performance, even in the absence of large-scale labelled datasets. The code and the models are available here: https://github.com/eedack01/lung_masked_autoencoder.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。