Xie Lipeng, Xu Yongrui, Zheng Mingfeng, Chen Yundi, Sun Min, Archer Michael A, Wan Yuan, Mao Wenjun, Tong Yubing
medRxiv. 2024 May 5:2024.05.03.24306828. doi: 10.1101/2024.05.03.24306828.
To develop an anthropomorphic diagnosis system of pulmonary nodules (PN) based on Deep learning (DL) that is trained by weak annotation data and has comparable performance to full-annotation based diagnosis systems.
The proposed system uses deep learning (DL) models to classify PNs (benign vs. malignant) with weak annotations, which eliminates the need for time-consuming and labor-intensive manual annotations of PNs. Moreover, the PN classification networks, augmented with handcrafted shape features acquired through the ball-scale transform technique, demonstrate capability to differentiate PNs with diverse labels, including pure ground-glass opacities, part-solid nodules, and solid nodules.
The experiments were conducted on two lung CT datasets: (1) public LIDC-IDRI dataset with 1,018 subjects, (2) In-house dataset with 2740 subjects. Through 5-fold cross-validation on two datasets, the system achieved the following results: (1) an Area Under Curve (AUC) of 0.938 for PN localization and an AUC of 0.912 for PN differential diagnosis on the LIDC-IDRI dataset of 814 testing cases, (2) an AUC of 0.943 for PN localization and an AUC of 0.815 for PN differential diagnosis on the in-house dataset of 822 testing cases. These results demonstrate comparable performance to full annotation-based diagnosis systems.
Our system can efficiently localize and differentially diagnose PNs even in resource-limited environments with good robustness across different grade and morphology sub-groups in the presence of deviations due to the size, shape, and texture of the nodule, indicating its potential for future clinical translation.
An anthropomorphic diagnosis system of pulmonary nodules (PN) based on deep learning and weak annotation was found to achieve comparable performance to full-annotation dataset-based diagnosis systems, significantly reducing the time and the cost associated with the annotation.
A fully automatic system for the diagnosis of PN in CT scans using a suitable deep learning model and weak annotations was developed to achieve comparable performance (AUC = 0.938 for PN localization, AUC = 0.912 for PN differential diagnosis) with the full-annotation based deep learning models, reducing around 30%∼80% of annotation time for the experts.The integration of the hand-crafted feature acquired from human experts (natural intelligence) into the deep learning networks and the fusion of the classification results of multi-scale networks can efficiently improve the PN classification performance across different diameters and sub-groups of the nodule.
开发一种基于深度学习（DL）的肺结节（PN）拟人化诊断系统，该系统由弱标注数据训练，且性能与基于全标注的诊断系统相当。
所提出的系统使用深度学习（DL）模型对具有弱标注的PN进行分类（良性与恶性），这消除了对PN进行耗时且费力的手动标注的需求。此外，通过球尺度变换技术获取的手工形状特征增强的PN分类网络，展示了区分具有不同标签的PN的能力，包括纯磨玻璃影、部分实性结节和实性结节。
实验在两个肺部CT数据集上进行：（1）包含1018名受试者的公共LIDC-IDRI数据集，（2）包含2740名受试者的内部数据集。通过在两个数据集上进行5折交叉验证，该系统取得了以下结果：（1）在814个测试病例的LIDC-IDRI数据集上，PN定位的曲线下面积（AUC）为0.938，PN鉴别诊断的AUC为0.912；（2）在822个测试病例的内部数据集上，PN定位的AUC为0.943，PN鉴别诊断的AUC为0.815。这些结果表明其性能与基于全标注的诊断系统相当。
即使在资源有限的环境中，我们的系统也能有效地对PN进行定位和鉴别诊断，在存在由于结节大小、形状和纹理导致的偏差的情况下，对不同分级和形态亚组具有良好的稳健性，表明其未来临床转化的潜力。
发现一种基于深度学习和弱标注的肺结节（PN）拟人化诊断系统，其性能与基于全标注数据集的诊断系统相当，显著减少了与标注相关的时间和成本。
开发了一种使用合适的深度学习模型和弱标注对CT扫描中的PN进行诊断的全自动系统，以实现与基于全标注的深度学习模型相当的性能（PN定位的AUC = 0.938，PN鉴别诊断的AUC = 0.912），减少了专家约30%至80%的标注时间。将从人类专家（自然智能）获取的手工特征集成到深度学习网络中，以及多尺度网络分类结果的融合，可以有效地提高跨不同直径和结节亚组进行PN分类的性能。
