胸部计算机断层扫描检测到的肺结节的数据驱动风险分层和精确管理-动脉网

Data-driven risk stratification and precision management of pulmonary nodules detected on chest computed tomography

Nature 等信源发布 2024-09-17 17:54



可切换为仅中文







AbstractThe widespread implementation of low-dose computed tomography (LDCT) in lung cancer screening has led to the increasing detection of pulmonary nodules. However, precisely evaluating the malignancy risk of pulmonary nodules remains a formidable challenge. Here we propose a triage-driven Chinese Lung Nodules Reporting and Data System (C-Lung-RADS) utilizing a medical checkup cohort of 45,064 cases.

摘要低剂量计算机断层扫描（LDCT）在肺癌筛查中的广泛应用导致肺结节的检出率增加。然而，准确评估肺结节的恶性风险仍然是一项艰巨的挑战。。

The system was operated in a stepwise fashion, initially distinguishing low-, mid-, high- and extremely high-risk nodules based on their size and density. Subsequently, it progressively integrated imaging information, demographic characteristics and follow-up data to pinpoint suspicious malignant nodules and refine the risk scale.

该系统以逐步的方式运行，最初根据其大小和密度区分低，中，高和极高风险的结节。随后，它逐步整合成像信息，人口统计学特征和随访数据，以查明可疑的恶性结节并完善风险等级。

The multidimensional system achieved a state-of-the-art performance with an area under the curve (AUC) of 0.918 (95% confidence interval (CI) 0.918–0.919) on the internal testing dataset, outperforming the single-dimensional approach (AUC of 0.881, 95% CI 0.880–0.882). Moreover, C-Lung-RADS exhibited a superior sensitivity compared with Lung-RADS v2022 (87.1% versus 63.3%) in an independent cohort, which was screened using mobile computed tomography scanners to broaden screening accessibility in resource-constrained settings.

多维系统在内部测试数据集上实现了最先进的性能，曲线下面积（AUC）为0.918（95%置信区间（CI）为0.918-0.919），优于一维方法（AUC为0.881，95%CI为0.880-0.882）。此外，在一个独立的队列中，C-Lung-RADS与Lung-RADS v2022（87.1%比63.3%）相比表现出更高的敏感性，该队列使用移动计算机断层扫描扫描仪进行筛查，以扩大资源受限环境中的筛查可及性。

With its foundation in precise risk stratification and tailored management, this system has minimized unnecessary invasive procedures for low-risk cases and recommended prompt intervention for extremely high-risk nodules to avert diagnostic delays. This approach has the potential to enhance the decision-making paradigm and facilitate a more efficient diagnosis of lung cancer during routine checkups as well as screening scenarios..

该系统以精确的风险分层和量身定制的管理为基础，最大程度地减少了低风险病例不必要的侵入性手术，并建议对极高风险结节进行及时干预，以避免诊断延误。这种方法有可能增强决策范式，并有助于在常规检查和筛查情景中更有效地诊断肺癌。。

MainPulmonary nodule is one of the most frequently detected abnormalities in chest imaging and the critical aspect of diagnosis is to distinguish malignant nodules clinically relevant to lung cancer from benign nodules1,2. Despite numerous efforts, lung cancer persists as a predominant malignant tumor in terms of mortality rate with the highest economic burden globally, with a particularly significant impact in China3,4,5.

主肺结节是胸部影像学中最常见的异常之一，诊断的关键方面是区分临床上与肺癌相关的恶性结节和良性结节1,2。。

Individuals diagnosed with early-stage diseases are more likely to receive curative treatment and experience superior prognosis compared with patients diagnosed at advanced stage6. In China, there remains a gap in early-stage lung cancer detection rates compared with high-income countries (stage I: 17.3% in China versus 25.3% in the United States)7,8.

与晚期诊断的患者相比，被诊断患有早期疾病的患者更有可能接受治愈性治疗并获得更好的预后6。在中国，与高收入国家相比，早期肺癌检出率仍然存在差距（第一阶段：中国为17.3%，美国为25.3%）7,8。

This situation underscores the urgent need for widespread lung cancer screening in China to confirm the cases detected at an early stage.Low-dose computed tomography (LDCT) has been confirmed as an effective tool for lung cancer screening2,9. Pivotal studies such as the National Lung Screening Trial (NLST) and Nederlands–Leuvens Longkanker Screenings Onderzoek (NELSON) cohorts have demonstrated that LDCT significantly reduced lung cancer mortality10,11.

这种情况突出表明，迫切需要在中国进行广泛的肺癌筛查，以确认早期发现的病例。低剂量计算机断层扫描（LDCT）已被证实是肺癌筛查的有效工具2,9。关键研究，如国家肺部筛查试验（NLST）和荷兰-鲁汶-隆坎克筛查Onderzoek（NELSON）队列研究表明，LDCT可显着降低肺癌死亡率10,11。

In addition, a prospective multicenter cohort study in China has indicated that one-off LDCT screening reduced lung cancer mortality by 31% in high-risk populations12. With the widespread application of LDCT, the detection rate of pulmonary nodules has gradually improved13,14. However, at least 95% of pulmonary nodules screened are benign, necessitating precise management strategies to ensure appropriate intervention1.

此外，中国的一项前瞻性多中心队列研究表明，一次性LDCT筛查可将高危人群的肺癌死亡率降低31%12。随着LDCT的广泛应用，肺结节的检出率逐渐提高13,14。然而，至少95%的筛查肺结节是良性的，需要精确的管理策略来确保适当的干预1。

For instance, only 3.6% of detected lung nodules were diagnosed as malignant in the NLST, and the baseline false-positive rate (FPR) was a.

例如，在NLST中，只有3.6%的检测到的肺结节被诊断为恶性，基线假阳性率（FPR）为a。

(1)

$${\mathrm{IV}}_{i}=\left(\frac{{{\rm{\#}}P}_{i}}{{\rm{\#}}P}-\frac{{{\rm{\#}}N}_{i}}{{\rm{\#}}N}\right)\times {\mathrm{WOE}}_{i}=\left(\frac{{{\rm{\#}}P}_{i}}{{\rm{\#}}P}-\frac{{{\rm{\#}}N}_{i}}{{\rm{\#}}N}\right)\times \left(\mathrm{ln}\frac{{{\rm{\#}}P}_{i}}{{\rm{\#}}P}-\mathrm{ln}\frac{{{\rm{\#}}N}_{i}}{{\rm{\#}}N}\right),$$.

$${\mathrm{IV}}}{i}=\left（\frac{{{\rm{}}P}{i}}{\rm{}}P}-\frac{{\rm{}}N}{i}}{\rm{}}N}\ right）}次{\mathrm{WOE}}}}{i}=\left（\frac{{{\rm{}}P}}{i}}{\rm{}}P}-\frac{{\rm{}}N}}{\rm}}}N}次\left（\mathrm{ln}\frac{{{\rm{i}}{{\rm{}}P}-\mathrm{ln}\frac{{\rm{}}N}}\ui}}{\rm{\rm}}N}\右），$$。

(2)

$${\mathrm{IV}}=\sum _{i\in X}{\mathrm{I{V}}}_{i},$$

$${\mathrm{IV}=\sum{i\in X}{\mathrm{i{V}}}}\ui}$$

(3)

where $X$ is the group of categorical variables from 1 to 4, i is the current category, $\#$ denotes the number, $P$ refers to overall positives, ${P}_{i}$ refers to the positives in the ${i}{\rm{th}}$ category, $N$ refers to overall negatives and ${N}_{i}$ refers to the negatives in the ${i}{\rm{th}}$ category.

其中\（X \）是从1到4的一组分类变量，i是当前类别，\（\）表示数字，\（P \）表示总体正值\({P}_{i} \）指的是\（{i}{\rm{th}}）类别中的积极因素，\（N \）指的是整体消极因素和\({N}_{i} \）指的是\（{i}{\rm{th}}）范畴中的否定词。

A reasonable classification scheme entailed ${\rm{WOE}}_{i}$ increasing with i, indicating a higher malignancy proportion with escalating risk levels. Therefore, the initial four-category risk stratification rule was achieved in phase 1, with 1 representing low risk, 2 representing mid risk, 3 representing high risk and 4 representing extremely high risk, used for screening of non-low-risk nodules.Phase 2: malignancy evaluation by a deep learning modelDeep learning algorithms have indeed shown promising results in identifying malignancies, differentiating cancer subtypes and predicting tumor invasiveness73,74,75.

一个合理的分类方案需要“（{\ rm{WOE}}}ui}”）随着i的增加而增加，表明随着风险水平的升高，恶性肿瘤比例更高。因此，最初的四类风险分层规则是在第一阶段实现的，其中1代表低风险，2代表中风险，3代表高风险，4代表极高风险，用于筛查非低风险结节。阶段2：通过深度学习模型进行恶性肿瘤评估深度学习算法确实在识别恶性肿瘤，区分癌症亚型和预测肿瘤侵袭性方面显示出有希望的结果73,74,75。

In phase 2, a DCNN model was developed to generate image-level malignant probabilities of nodules.Construction of the DCNN modelDuring the training process of phase 2, a DCNN model was developed using 5,452 pulmonary nodules from as many participants. Nodule images were input to predict malignancy probability, aiming to differentiate malignant nodules from benign ones.The DCNN architecture included an input block, four continuous down-sampling blocks and an output block, referring to a prior publication76.

在第二阶段，开发了DCNN模型以产生结节的图像级恶性概率。DCNN模型的构建在第二阶段的训练过程中，使用来自尽可能多的参与者的5452个肺结节开发了DCNN模型。输入结节图像以预测恶性概率，旨在区分恶性结节和良性结节。DCNN体系结构包括一个输入块，四个连续下采样块和一个输出块，参考先前的出版物76。

Briefly, (1) the input block was a three-dimensional convolutional layer for converting images into semantic representations, (2) the four down-sampling blocks included four convolutional layers for generating feature maps, (3) a global average pooling (GAP) layer regularized the network to prevent overfitting and (4).

简而言之，（1）输入块是用于将图像转换为语义表示的三维卷积层，（2）四个下采样块包括用于生成特征图的四个卷积层，（3）全局平均池（GAP）层将网络正则化以防止过度拟合和（4）。

(4)

$${ {\mathcal L} }_{{\rm{CAM}}}=\frac{1}{H\times W}{\sum_{x,y}\Vert {\rm{nodule}}\_{\rm{mask}}^{\prime} (x,y)-{\rm{CA{M}}}_{i}^{\prime} (x,y)\Vert_{{l}_{1}}},$$

$${{\mathcal L}}}}}{\rm{CAM}}}=\frac{1}{H\times W}{\sum{x，y}\Vert{\rm{nodule}}\u{\ rm{mask}}^{\prime}（x，y）-{\rm{CA{M}}}}\ui}^{\prime}（x，y）\Vert一_{{l}_{1} }}$$

(5)

$${{\mathcal{L}}}_{\rm{CE}}=-\log \frac{{e}^{\left({z}_{i}\right)}}{{\sum }_{j}{e}^{\left({z}_{i}\right)}},$$

({z}_{i} \右）}}{{\和}}{j}{e}^{\左({z}_{i} \右）}}$$

(6)

$${z}_{i}=\frac{1}{H\times W}\sum _{x,y}{\rm{CAM}}_{i}\left(x,y\right),$$

（笑声）$${z}_{i} =\frac{1}{H \乘以W}\sum{x，y}{\rm{CAM}}\ui}\左（x，y \右）$$

(7)

where ${\rm{\alpha }}$ denotes the combined ratio; ${{\mathcal{L}}}_{\rm{CAM}}$ measures the Dice similarity coefficient between nodule_mask and ${\rm{CAM}}_{i}$, driving the network to learn more spatially discriminative feature representations and to focus on nodule regions; ${\rm{nodule}}\_{\rm{mask}}^{\prime}$ and ${\rm{CAM}}_{i}^{\prime}$ are the minimum–maximum normalization of ${\rm{nodule}}\_{\rm{mask}}$ and ${\rm{CAM}}$, respectively; ${\rm{CAM}}_{i}$ is defined as the CAM for class i; and ${\rm{CAM}}_{i}(x,y)$ indicates the importance of the activation at $(x,y)$, leading to an image belonging to class i.

其中\（{\ rm{\ alpha}}\）表示组合比率\（{{\mathcal{L}}}}{\rm{CAM}}}测量nodule\u mask和\（{\rm{CAM}}}}ui}\）之间的骰子相似系数，驱动网络学习更多的空间区分特征表示并关注结节区域\（{\ rm{nodule}}\{\ rm{mask}}^{\ prime}\）和\（{\ rm{CAM}}}\{i}^{\ prime}\）分别是\（{\ rm{nodule}}\{\ rm{mask}}\）和\（{\ rm{CAM}}}\）的最小-最大归一化\（{\ rm{CAM}}{i}\）被定义为i类的CAM；和\（{\ rm{CAM}}\{i}（x，y）\）表示在\（（x，y）\）激活的重要性，导致图像属于i类。

H and W denote the height and width of the nodule mask, respectively; l1 is the L1 norm, a type of norm used in mathematics to measure the size of a vector; e is Euler’s number, a mathematical constant approximately equal to 2.71828; zi represents the activation value of class i in the CAM; and j is an index variable used in the softmax function to represent different classes or categories.During the training process, other parameters were carefully set.

H和W分别表示结节面罩的高度和宽度；l1是l1范数，数学中用于测量向量大小的一种范数；e是欧拉数，一个大约等于2.71828的数学常数；zi表示CAM中i类的激活值；j是softmax函数中用于表示不同类别或类别的索引变量。在训练过程中，仔细设置了其他参数。

The learning rate used to refine the network was reduced from a large initial value (1 × 10−3) to a small value (1 × 10−5). The Adam optimizer was set to betas of (0.9, 0.999) and epsilon of 1 × 10−8. Data augmentation techniques such as shifting, scaling, flipping, cropping, rotating and adding noise were employed to improve model robustness.Model calibrationThe output of malignancy probability was calibrated by Platt scaling and temperature scaling in the training stage.

用于改进网络的学习率从较大的初始值（1××10-3）降低到较小的值（1××10-5）。Adam优化器设置为betas为（0.9，0.999），epsilon为1×10-8。采用移位，缩放，翻转，裁剪，旋转和添加噪声等数据增强技术来提高模型的鲁棒性。模型校准在训练阶段通过Platt标度和温度标度校准恶性概率的输出。

By adjusting the scaling and intercept parameters of the logistic regression model, the calibrated probabilities could better reflect the true likelihood of each class. Temperature scaling involved adjust.

通过调整逻辑回归模型的标度和截距参数，校准的概率可以更好地反映每个类别的真实可能性。温度标度涉及调整。

(8)

$${{\rm{VDT}}}=\frac{\mathrm{ln}2}{\rm{SGR}}=\frac{\mathrm{ln}2\times \Delta T}{\mathrm{ln}({V}_{2}/{V}_{1})},$$

$${{\rm{VDT}}=\frac{\mathrm{ln}2}{\rm{SGR}}=\frac{\mathrm{ln}2\时间\Delta T}{\mathrm{ln}({V}_{2}/{V}_{1} ）}$$

(9)

where ${V}_{1}$ and ${V}_{2}$ are the nodule volumes of a follow-up pair of images quantified from AI-based segmentation results78 and $\Delta T$ represents the time interval between scans. SGR was decomposed into its positive (SGR+) and negative (SGR−) components. This strategy enabled the model to independently evaluate the trends of growth and reduction in nodule volume changes.

在哪里\({V}_{1} \）和\({V}_{2} \）是从基于AI的分割结果量化的后续图像对的结节体积78，\（\ Delta T \）表示扫描之间的时间间隔。SGR被分解为其正（SGR+）和负（SGR-）成分。该策略使模型能够独立评估结节体积变化的增长和减少趋势。

Multidimensional features were fused as input for the GBR model to output malignant probability and corresponding risk level. The GBR model was trained on 15,290 CT examinations (including follow-up scans) from 5,452 participants and represented as follows:$$G\left(x\right)={\rm{Sigmoid}}(\;{g}_{\rm{AI}}\left(I\;\right)+{g}_{\rm{C}}\left({x}_{\rm{C}}\right)+{g}_{\rm{F}}\left({x}_{\rm{F}}\right)),$$.

多维特征被融合作为GBR模型的输入，以输出恶性概率和相应的风险水平。GBR模型接受了来自5452名参与者的15290次CT检查（包括随访扫描）的训练，表现如下：$$G \ left（x \ right）={\ rm{乙状结肠}}（\；{g}_{\rm{AI}}\左（I \；\右）+{g}_{\rm{C}}\左({x}_{\rm{C}}\右）+{g}_{\rm{F}}\左({x}_。

(10)

where $G\left(x\right)$ is the output malignancy probability of the GBR model, ${g}_{\rm{AI}}$ is the DCNN and the input i is the CT nodal patch to generate the logits of malignancy probability. ${x}_{\rm{C}}$ and ${g}_{\rm{C}}$ represent the clinical features and coefficients, while ${x}_{\rm{F}}$ and ${g}_{\rm{F}}$ represent the nodule follow-up features and coefficients.

其中\（G左（x右）\）是GBR模型的输出恶性概率\({g}_{\ rm{AI}}\）是DCNN，输入i是CT节点补丁，以生成恶性概率的对数\（笑声）({x}_{\rm{C}}\）和\({g}_{\ rm{C}}\）代表临床特征和系数，而\({x}_{\rm{F}}\）和\({g}_{\rm{F}}\）代表结节的随访特征和系数。

Three items in the formula served as weak prediction models, trained sequentially to compensate the weakness of their predecessor and assembled together to become the ultimate trained model. The least absolute shrinkage and selection operator (LASSO) algorithm was used to select the most important features and generate optimal coefficients.

公式中的三个项目用作弱预测模型，依次训练以弥补其前任的弱点，并组装在一起成为最终的训练模型。最小绝对收缩和选择算子（LASSO）算法用于选择最重要的特征并生成最佳系数。

Logistic loss was used as the classification loss function. Finally, the GBR model was represented as follows:$$\begin{array}{l}{\rm{Sigmoid}}\left(1\times {\rm{AI}}\; {\rm{prediction}}+0.11\times {\rm{sex}}\left({\rm{female}}\right)+3.00\times 10^{-4}\times {\rm{age}}\right.\\\qquad\quad+\,0.07\times {\rm{smoking}}+0.19\times {\rm{history}}\; {\rm{of}}\; {\rm{cancer}}+0.09\\\qquad\quad\left.\times\, {\rm{family}}\; {\rm{history}}\; {\rm{of}}\; {\rm{cancer}}+185\times {\rm{SGR}}_{+}\right).\end{array}$$.

Logistic损失被用作分类损失函数。最后，GBR模型表示如下：$$\开始{数组}{l}{\ rm{乙状结肠}}\左（1倍{\ rm{AI}}；{\ rm{预测}}+0.11倍{\ rm{性别}\左（{\ rm{女性}}\右）+3.00倍10 ^{-4}\倍{\ rm{年龄}\右\\\qquad \ quad+\，0.07倍{\ rm{吸烟}}+0.19倍{\ rm{历史}}；{\rm{of}}\；{\rm{cancer}}+0.09 \ \ qquad \ quad \左。\时代\，{\rm{家庭}}\；{\rm{历史}}\；{\rm{of}}\；。\结束{数组}$$。

(11)

First, continuous variables were normalized to 0–1. Fivefold cross-validation was applied to avoid grouping bias and model overfitting.Second, a GBR algorithm started creating a single leaf based on the imaging model (${\rm{Sigmoid}}({g}_{\rm{AI}}\left(I\right))$), with the error between the output and label $y$ serving as the learning target for the second model.Third, when training the second model, LASSO regression was used to select the most important clinical features (${x}_{\rm{C}}$) and generate the optimal coefficients (${g}_{\rm{C}}$).

首先，将连续变量标准化为0-1。应用五倍交叉验证以避免分组偏差和模型过度拟合。其次，GBR算法开始基于成像模型创建单个叶片（\（{\ rm{Sigmoid}}）({g}_{\rm{AI}}\ left（I \ right））\），输出和标签\（y \）之间的错误作为第二个模型的学习目标。第三，在训练第二个模型时，使用套索回归来选择最重要的临床特征(\({x}_{\rm{C}}\）并生成最佳系数(\({g}_{\rm{C}}\）。

To eliminate the strong correlation between sex and smoking, clinical features were fitted in two steps. First, sex was balanced through subsampling to fit the regression model for other clinical features. This step was repeated 50 times, using a bagging strategy to create an ensemble of multiple models.

为了消除性别与吸烟之间的强相关性，临床特征分两步拟合。首先，通过二次抽样来平衡性别，以拟合其他临床特征的回归模型。这一步骤重复了50次，使用装袋策略创建了多个模型的集合。

Second, a gradient-boosting strategy was employed for univariate regression on the sex dimension. This updated the model ${\mathrm{Sigmoid}}({g}_{\mathrm{AI}}\left(I\right)+g_{{\rm{C}}}\left({x}_{{\rm{C}}}\right))$, and the error between the output and label $y$ served as the learning target for the third model.Fourth, LASSO regression was used in training the third model to select critical follow-up features (${x}_{\rm{F}}$) and optimize coefficients (${g}_{\rm{F}}$).Finally, a more robust multidimensional model was generated (${\rm{Sigmoid}}(\;{g}_{\rm{AI}}\left(I\;\right)+{g}_{\rm{C}}\left({x}_{\rm{C}}\right)+{g}_{\rm{F}}\left({x}_{\rm{F}}\right))$) to distinguish malignant nodules effectively compared with single- and dual-dimensional models.

其次，采用梯度提升策略对性别维度进行单变量回归。这更新了模型\（{\mathrm{Sigmoid}}({g}_{\mathrm{AI}}\左（I \右）+g\u{\ rm{C}}\左({x}_{{\ rm{C}}\ right））\），输出和标签\（y \）之间的错误作为第三个模型的学习目标。第四，套索回归用于训练第三个模型，以选择关键的后续特征(\({x}_{\rm{F}}\）和优化系数(\({g}_{\rm{F}}\）。最后，生成了一个更健壮的多维模型（\（{\ rm{Sigmoid}}（\；{g}_{\rm{AI}}\左（I \；\右）+{g}_{\rm{C}}\左({x}_{\rm{C}}\右）+{g}_{\rm{F}}\左({x}_与一维和二维模型相比，可以有效地区分恶性结节。

The output provided the malignancy probability for each nodule. Nodules with a probability of malignancy below 0.5 were predicted to be benign and retai.

输出提供了每个结节的恶性概率。恶性概率低于0.5的结节被预测为良性和复发性结节。

Data availability

数据可用性

The clinical data for this study were collected with the approval of the ethics committee and are subject to restrictions for this research. No publicly available datasets were used in this study. De-identified tabular data are strictly for noncommercial academic research and necessitate a formal agreement on data usage.

这项研究的临床数据是在伦理委员会的批准下收集的，并且受到这项研究的限制。本研究未使用公开可用的数据集。不确定的表格数据严格用于非商业性学术研究，需要就数据使用达成正式协议。

Al requests complying with legal and ethical requirements for data use will be granted. Data requests may be made to the corresponding author (Weimin Li, weimi003@scu.edu.cn). Requests will be processed within 2 months..

所有符合数据使用法律和道德要求的请求都将获得批准。可以向通讯作者（李伟民，weimi003@scu.edu.cn)。请求将在2个月内处理。。

Code availability

代码可用性

The code is available on Github (https://github.com/simonsf/C-Lung-RADS).

该代码可在Github上获得(https://github.com/simonsf/C-Lung-RADS)。

ReferencesMazzone, P. J. & Lam, L. Evaluating the patient with a pulmonary nodule: a review. JAMA 327, 264–273 (2022).Article

参考文献Mazzone，P.J。＆Lam，L。评估肺结节患者：综述。JAMA 327264-273（2022）。文章

PubMed

Google Scholar

谷歌学者

Adams, S. J. et al. Lung cancer screening. Lancet 401, 390–408 (2023).Article

Adams，S.J.等人。肺癌筛查。柳叶刀401390-408（2023）。文章

PubMed

Google Scholar

谷歌学者

Bray, F. et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J. Clin. 74, 229–263 (2024).Article

。CA Cancer J.Clin。74229-263（2024）。文章

PubMed

Google Scholar

谷歌学者

Chen, S. et al. Estimates and projections of the global economic cost of 29 cancers in 204 countries and territories from 2020 to 2050. JAMA Oncol. 9, 465–472 (2023).Article

Chen，S.等人对2020年至2050年204个国家和地区29种癌症的全球经济成本进行了估计和预测。JAMA Oncol。9465-472（2023）。文章

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Han, B. et al. Cancer incidence and mortality in China, 2022. J. Natl Cancer Cent. 4, 47–53 (2024).Article

Han，B.等人，《中国癌症发病率和死亡率》，2022年。J、自然癌症分。4，47-53（2024）。文章

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Detterbeck, F. C. et al. The International Association for the Study of Lung Cancer lung cancer staging project: proposals for revision of the classification of residual tumor after resection for the forthcoming (ninth) edition of the TNM Classification of Lung Cancer. J. Thorac. Oncol.

Detterbeck，F.C.等人，《国际肺癌研究协会肺癌分期项目：为即将出版的第九版TNM肺癌分类修订切除后残留肿瘤分类的建议》。J、胸部。。

19, 1052–1072 (2024).Article .

191052-1072（2024）。文章。

PubMed

Google Scholar

谷歌学者

Zeng, H. et al. Changing cancer survival in China during 2003–15: a pooled analysis of 17 population-based cancer registries. Lancet Glob. Health 6, 555–567 (2018).Article

Zeng，H.等人，《2003-15年中国癌症生存率的变化：对17个基于人群的癌症登记处的汇总分析》。柳叶刀球。健康6555-567（2018）。文章

Google Scholar

谷歌学者

Zeng, H. et al. Disparities in stage at diagnosis for five common cancers in China: a multicentre, hospital-based, observational study. Lancet Public Health 6, 877–887 (2021).Article

Zeng，H.等人。中国五种常见癌症诊断阶段的差异：一项多中心，医院为基础的观察性研究。柳叶刀公共卫生6877-887（2021）。文章

Google Scholar

谷歌学者

Oudkerk, M., Liu, S., Heuvelmans, M. A., Walter, J. E. & Field, J. K. Lung cancer LDCT screening and mortality reduction—evidence, pitfalls and future perspectives. Nat. Rev. Clin. Oncol. 18, 135–151 (2021).Article

Oudkerk，M.，Liu，S.，Heuvelmans，M.A.，Walter，J.E。和Field，J.K。肺癌LDCT筛查和降低死亡率的证据，陷阱和未来前景。国家修订临床。Oncol公司。18135-151（2021）。文章

PubMed

Google Scholar

谷歌学者

Aberle, D. R. et al. Reduced lung-cancer mortality with low-dose computed tomographic screening. N. Engl. J. Med. 365, 395–409 (2011).Article

Aberle，D.R.等人通过低剂量计算机断层扫描筛查降低肺癌死亡率。N、英语。J、医学365395-409（2011）。文章

PubMed

Google Scholar

谷歌学者

de Koning, H. J. et al. Reduced lung-cancer mortality with volume CT screening in a randomized trial. N. Engl. J. Med. 382, 503–513 (2020).Article

de Koning，H.J.等人在一项随机试验中通过容积CT筛查降低肺癌死亡率。N、英语。J、医学382503-513（2020）。文章

PubMed

Google Scholar

谷歌学者

Li, N. et al. One-off low-dose CT for lung cancer screening in China: a multicentre, population-based, prospective cohort study. Lancet Respir. Med. 10, 378–391 (2022).Article

Li，N。等。中国肺癌筛查的一次性低剂量CT：一项多中心，基于人群的前瞻性队列研究。柳叶刀呼吸。医学杂志10378-391（2022）。文章

CAS

中科院

PubMed

Google Scholar

谷歌学者

Gould, M. K. et al. Recent trends in the identification of incidental pulmonary nodules. Am. J. Respir. Crit. Care Med. 192, 1208–1214 (2015).Article

。Am.J.Respir。暴击。护理医学1921208-1214（2015）。文章

PubMed

Google Scholar

谷歌学者

Hendrix, W. et al. Trends in the incidence of pulmonary nodules in chest computed tomography: 10-year results from two Dutch hospitals. Eur. Radio. 33, 8279–8288 (2023).Article

Hendrix，W.等人，《胸部计算机断层扫描中肺结节发病率的趋势：荷兰两家医院的10年结果》。欧洲电台。338279–8288（2023）。文章

Google Scholar

谷歌学者

Pinsky, P. F. et al. Performance of Lung-RADS in the National Lung Screening Trial: a retrospective assessment. Ann. Intern. Med. 162, 485–491 (2015).Article

Pinsky，P.F.等人，《国家肺部筛查试验中肺部RADS的表现：回顾性评估》。安，实习生。医学162485-491（2015）。文章

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Gould, M. K. et al. Evaluation of individuals with pulmonary nodules: when is it lung cancer? Diagnosis and management of lung cancer, 3rd ed: American College of Chest Physicians evidence-based clinical practice guidelines. Chest 143, e93S–e120S (2013).Article

Gould，M.K.等人。肺结节患者的评估：何时是肺癌？肺癌的诊断和管理，第三版：美国胸科医师学会循证临床实践指南。胸围143，e93S–e120S（2013）。文章

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Baldwin, D. R. & Callister, M. E. The British Thoracic Society guidelines on the investigation and management of pulmonary nodules. Thorax 70, 794–798 (2015).Article

。胸部70794-798（2015）。文章

PubMed

Google Scholar

谷歌学者

MacMahon, H. et al. Guidelines for management of incidental pulmonary nodules detected on CT images: from the Fleischner Society 2017. Radiology 284, 228–243 (2017).Article

MacMahon，H.等人，《CT图像上偶然发现的肺结节管理指南：来自Fleischner Society 2017》。放射学284228-243（2017）。文章

PubMed

Google Scholar

谷歌学者

Kastner, J. et al. Lung-RADS version 1.0 versus Lung-RADS version 1.1: comparison of categories using nodules from the National Lung Screening Trial. Radiology 300, 199–206 (2021).Article

Kastner，J.等人，《Lung RADS 1.0版与Lung RADS 1.1版：使用国家肺部筛查试验中的结节进行类别比较》。放射学300199-206（2021）。文章

PubMed

Google Scholar

谷歌学者

Bai, C. et al. Evaluation of pulmonary nodules: clinical practice consensus guidelines for Asia. Chest 150, 877–893 (2016).Article

Bai，C.等人，《肺结节的评估：亚洲临床实践共识指南》。胸部150877-893（2016）。文章

PubMed

Google Scholar

谷歌学者

Azour, L., Ko, J. P., Naidich, D. P. & Moore, W. H. Shades of gray: subsolid nodule considerations and management. Chest 159, 2072–2089 (2021).Article

Azour，L.，Ko，J.P.，Naidich，D.P。和Moore，W.H。灰色阴影：地下结核的考虑和管理。胸部1592072-2089（2021）。文章

PubMed

Google Scholar

谷歌学者

American College of Radiology. Lung-RADS 2022. https://www.acr.org/Clinical-Resources/Reporting-and-Data-Systems/Lung-Rads (2022).Swensen, S. J., Silverstein, M. D., Ilstrup, D. M., Schleck, C. D. & Edell, E. S. The probability of malignancy in solitary pulmonary nodules. Application to small radiologically indeterminate nodules.

美国放射学院。肺RADS 2022。https://www.acr.org/Clinical-Resources/Reporting-and-Data-Systems/Lung-Rads（2022年）。Swensen，S.J.，Silverstein，M.D.，Ilstrup，D.M.，Schleck，C.D。＆Edell，E.S。孤立性肺结节恶性肿瘤的可能性。应用于放射学不确定的小结节。

Arch. Intern. Med. 157, 849–855 (1997).Article .

。实习生。医学157849-855（1997）。文章。

CAS

中科院

PubMed

Google Scholar

谷歌学者

McWilliams, A. et al. Probability of cancer in pulmonary nodules detected on first screening CT. N. Engl. J. Med. 369, 910–919 (2013).Article

McWilliams，A。等人。首次筛查CT时发现肺结节癌变的可能性。N。Engl。J、医学369910-919（2013）。文章

CAS

中科院

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Song, F. et al. Comparison of different classification systems for pulmonary nodules: a multicenter retrospective study in China. Cancer Imag. 24, 15 (2024).Article

Song，F。等。肺结节不同分类系统的比较：中国的多中心回顾性研究。癌症图像。24、15（2024年）。文章

Google Scholar

谷歌学者

Chen, K. et al. Development and validation of machine learning-based model for the prediction of malignancy in multiple pulmonary nodules: analysis from multicentric cohorts. Clin. Cancer Res. 27, 2255–2265 (2021).Article

Chen，K.等人。基于机器学习的多发性肺结节恶性肿瘤预测模型的开发和验证：来自多中心队列的分析。临床。癌症研究272255-2265（2021）。文章

PubMed

Google Scholar

谷歌学者

Esteva, A. et al. Dermatologist-level classification of skin cancer with deep neural networks. Nature 542, 115–118 (2017).Article

Esteva，A。等人。皮肤科医生用深度神经网络对皮肤癌进行分类。自然542115-118（2017）。文章

CAS

中科院

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Zhang, K. et al. Clinically applicable AI system for accurate diagnosis, quantitative measurements, and prognosis of COVID-19 pneumonia using computed tomography. Cell 181, 1423–1433 (2020).Article

。细胞1811423-1433（2020）。文章

CAS

中科院

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Shao, J. et al. A multimodal integration pipeline for accurate diagnosis, pathogen identification, and prognosis prediction of pulmonary infections. Innovation 5, 100648 (2024).CAS

Shao，J。等人。用于肺部感染的准确诊断，病原体鉴定和预后预测的多模式整合管道。创新5100648（2024）。中科院

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Wang, C. et al. Development and validation of an abnormality-derived deep-learning diagnostic system for major respiratory diseases. NPJ Digit. Med. 5, 124 (2022).Article

Wang，C.等人。主要呼吸系统疾病异常衍生深度学习诊断系统的开发和验证。NPJ数字。医学5124（2022）。文章

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Chen, R. J. et al. Pan-cancer integrative histology-genomic analysis via multimodal deep learning. Cancer Cell 40, 865–878 (2022).Article

Chen，R.J.等人。通过多模式深度学习进行泛癌综合组织学基因组分析。癌细胞40865-878（2022）。文章

CAS

中科院

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Xu, X., Wang, C., Guo, J., Yang, L. & Yi, Z. DeepLN: a framework for automatic lung nodule detection using multi-resolution CT screening images. Knowl. Based Syst. 189, 105128 (2019).Article

Xu，X.，Wang，C.，Guo，J.，Yang，L。＆Yi，Z。DeepLN：使用多分辨率CT筛查图像进行自动肺结节检测的框架。诺尔。基于系统。189105128（2019）。文章

Google Scholar

谷歌学者

Kann, B. H., Hosny, A. & Aerts, H. Artificial intelligence for clinical oncology. Cancer Cell 39, 916–927 (2021).Article

Kann，B.H.，Hosny，A。＆Aerts，H。临床肿瘤学人工智能。癌细胞39916-927（2021）。文章

CAS

中科院

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Shao, J. et al. Novel tools for early diagnosis and precision treatment based on artificial intelligence. Chin. Med J. Pulm. Crit. Care Med. 1, 148–160 (2023).Article

邵，J。等。基于人工智能的早期诊断和精确治疗的新工具。下巴。医学J.Pulm。暴击。。文章

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Ardila, D. et al. End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography. Nat. Med. 25, 954–961 (2019).Article

Ardila，D.等人。低剂量胸部计算机断层扫描三维深度学习的端到端肺癌筛查。《自然医学》25954-961（2019）。文章

CAS

中科院

PubMed

Google Scholar

谷歌学者

Baldwin, D. R. et al. External validation of a convolutional neural network artificial intelligence tool to predict malignancy in pulmonary nodules. Thorax 75, 306–312 (2020).Article

Baldwin，D.R.等人。用于预测肺结节恶性程度的卷积神经网络人工智能工具的外部验证。胸部75306-312（2020）。文章

PubMed

Google Scholar

谷歌学者

Venkadesh, K. V. et al. Deep learning for malignancy risk estimation of pulmonary nodules detected at low-dose screening CT. Radiology 300, 438–447 (2021).Article

Venkadesh，K.V.等人，《低剂量筛查CT检测到的肺结节恶性风险评估的深度学习》。放射学300438-447（2021）。文章

PubMed

Google Scholar

谷歌学者

Massion, P. P. et al. Assessing the accuracy of a deep learning method to risk stratify indeterminate pulmonary nodules. Am. J. Respir. Crit. Care Med. 202, 241–249 (2020).Article

Massion，P.P.等人评估深度学习方法对不确定肺结节进行风险分层的准确性。Am.J.Respir。暴击。护理医学202241-249（2020）。文章

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Shi, F. et al. Semi-supervised deep transfer learning for benign-malignant diagnosis of pulmonary nodules in chest CT images. IEEE Trans. Med. Imag. 41, 771–781 (2022).Article

Shi，F。等人。半监督深度转移学习用于胸部CT图像中肺结节的良恶性诊断。IEEE Trans。医学图像。41771-781（2022）。文章

Google Scholar

谷歌学者

Chen, R. J. et al. Algorithmic fairness in artificial intelligence for medicine and healthcare. Nat. Biomed. Eng. 7, 719–742 (2023).Article

Chen，R.J.等人。用于医学和医疗保健的人工智能中的算法公平性。自然生物医学。工程7719-742（2023）。文章

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Chen, L. et al. An artificial-intelligence lung imaging analysis system (ALIAS) for population-based nodule computing in CT scans. Comput. Med. Imag. Graph. 89, 101899 (2021).Article

Chen，L.等人。用于CT扫描中基于人群的结节计算的人工智能肺成像分析系统（ALIAS）。计算机。医学图像。图表。89101899（2021）。文章

Google Scholar

谷歌学者

Ohno, Y. et al. Differentiation of benign from malignant pulmonary nodules by using a convolutional neural network to determine volume change at chest CT. Radiology 296, 432–443 (2020).Article

。文章

PubMed

Google Scholar

谷歌学者

Kakinuma, R. et al. Natural history of pulmonary subsolid nodules: a prospective multicenter study. J. Thorac. Oncol. 11, 1012–1028 (2016).Article

Kakinuma，R。等。肺亚固体结节的自然史：一项前瞻性多中心研究。J、胸部。Oncol公司。111012-1028（2016）。文章

PubMed

Google Scholar

谷歌学者

Li, D. et al. Ten-year follow-up results of pure ground-glass opacity-featured lung adenocarcinomas after surgery. Ann. Thorac. Surg. 116, 230–237 (2023).Article

Li，D.等人。手术后纯磨玻璃样混浊特征性肺腺癌的十年随访结果。安。胸部。外科杂志116230-237（2023）。文章

PubMed

Google Scholar

谷歌学者

Chen, H. et al. The 2023 American Association for Thoracic Surgery (AATS) expert consensus document: management of subsolid lung nodules. J. Thorac. Cardiovasc. Surg. (2024).Azour, L. et al. Subsolid nodules: significance and current understanding. Clin. Chest Med. 45, 263–277 (2024).Article .

Chen，H.等人，《2023年美国胸外科协会（AATS）专家共识文件：皮下肺结节的管理》。J、胸部。心血管。。Azour，L.等人，《地下结核：意义和当前理解》。临床。胸部医学45263-277（2024）。文章。

PubMed

Google Scholar

谷歌学者

Travis, W. D. et al. The IASLC Lung Cancer Staging Project: proposals for coding T categories for subsolid nodules and assessment of tumor size in part-solid tumors in the forthcoming eighth edition of the TNM Classification of Lung Cancer. J. Thorac. Oncol. 11, 1204–1223 (2016).Article .

Travis，W.D.等人，《IASLC肺癌分期项目：在即将出版的《TNM肺癌分类》第八版中，为亚固体结节编码T类别并评估部分实体瘤的肿瘤大小的建议》。J、胸部。Oncol公司。111204-1223（2016）。文章。

PubMed

Google Scholar

谷歌学者

Lipkova, J. et al. Artificial intelligence for multimodal data integration in oncology. Cancer Cell 40, 1095–1110 (2022).Article

Lipkova，J。等人。用于肿瘤学多模式数据集成的人工智能。癌细胞401095-1110（2022）。文章

CAS

中科院

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Shao, J., Ma, J., Zhang, Q., Li, W. & Wang, C. Predicting gene mutation status via artificial intelligence technologies based on multimodal integration (MMI) to advance precision oncology. Semin. Cancer Biol. 91, 1–15 (2023).Article

Shao，J.，Ma，J.，Zhang，Q.，Li，W。＆Wang，C。通过基于多模式整合（MMI）的人工智能技术预测基因突变状态，以推进精准肿瘤学。塞米。癌症生物学。91，1-15（2023）。文章

CAS

中科院

PubMed

Google Scholar

谷歌学者

Boehm, K. M., Khosravi, P., Vanguri, R., Gao, J. & Shah, S. P. Harnessing multimodal data integration to advance precision oncology. Nat. Rev. Cancer 22, 114–126 (2022).Article

Boehm，K.M.，Khosravi，P.，Vanguri，R.，Gao，J。＆Shah，S.P。利用多模式数据集成来推进精确肿瘤学。《国家癌症评论》22114-126（2022）。文章

CAS

中科院

PubMed

Google Scholar

谷歌学者

Zhou, H. Y. et al. A transformer-based representation-learning model with unified processing of multimodal input for clinical diagnostics. Nat. Biomed. Eng. 7, 743–755 (2023).Article

Zhou，H.Y.等人。基于变压器的表征学习模型，用于临床诊断的多模态输入的统一处理。自然生物医学。工程7743-755（2023）。文章

PubMed

Google Scholar

谷歌学者

Prosper, A. E., Kammer, M. N., Maldonado, F., Aberle, D. R. & Hsu, W. Expanding role of advanced image analysis in CT-detected indeterminate pulmonary nodules and early lung cancer characterization. Radiology 309, e222904 (2023).Article

Prosper，A.E.，Kammer，M.N.，Maldonado，F.，Aberle，D.R。＆Hsu，W。扩展高级图像分析在CT检测到的不确定肺结节和早期肺癌表征中的作用。放射学309，e222904（2023）。文章

PubMed

Google Scholar

谷歌学者

Zhou, Y. et al. The application of artificial intelligence and radiomics in lung cancer. Precis. Clin. Med. 3, 214–227 (2020).Article

Zhou，Y。等。人工智能和放射组学在肺癌中的应用。精确。临床。医学杂志3214-227（2020）。文章

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Wang, F. et al. Risk-stratified approach for never- and ever-smokers in lung cancer screening: a prospective cohort study in China. Am. J. Respir. Crit. Care Med. 207, 77–88 (2023).Article

。Am.J.Respir。暴击。护理医学207,77-88（2023）。文章

PubMed

Google Scholar

谷歌学者

Krist, A. H. et al. Screening for lung cancer: US Preventive Services Task Force recommendation statement. JAMA 325, 962–970 (2021).Article

Krist，A.H.等人，《肺癌筛查：美国预防服务工作组建议声明》。JAMA 325962–970（2021）。文章

PubMed

Google Scholar

谷歌学者

Park, S. et al. Volume doubling times of lung adenocarcinomas: correlation with predominant histologic subtypes and prognosis. Radiology 295, 703–712 (2020).Article

Park，S.等。肺腺癌的体积倍增时间：与主要组织学亚型和预后的相关性。放射学295703-712（2020）。文章

PubMed

Google Scholar

谷歌学者

Venkadesh, K. V. et al. Prior CT improves deep learning for malignancy risk estimation of screening-detected pulmonary nodules. Radiology 308, e223308 (2023).Article

Venkadesh，K.V.等人先前的CT改进了深度学习，用于筛查检测到的肺结节的恶性风险评估。放射学308，e223308（2023）。文章

PubMed

Google Scholar

谷歌学者

Cao, W. et al. Uptake of lung cancer screening with low-dose computed tomography in China: a multi-centre population-based study. eClinicalMedicine 52, 101594 (2022).Article

Cao，W。等。中国低剂量计算机断层扫描对肺癌筛查的摄取：一项基于多中心人群的研究。eClinicalMedicine 52101594（2022）。文章

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Shao, J. et al. Deep learning empowers lung cancer screening based on mobile low-dose computed tomography in resource-constrained sites. Front. Biosci. 27, 212 (2022).Article

Shao，J。等人。深度学习能够在资源受限的地点基于移动低剂量计算机断层扫描进行肺癌筛查。正面。生物科学。27212（2022）。文章

Google Scholar

谷歌学者

Dhoot, R. et al. Implementing a mobile diagnostic unit to increase access to imaging and laboratory services in western Kenya. BMJ Glob. Health 3, 000947 (2018).Article

Dhoot，R.等人。在肯尼亚西部实施移动诊断单元，以增加成像和实验室服务的使用。BMJ全球。健康3000947（2018）。文章

Google Scholar

谷歌学者

Bartlett, E. C. et al. Baseline results of the west London lung cancer screening pilot study—impact of mobile scanners and dual risk model utilisation. Lung Cancer 148, 12–19 (2020).Article

Bartlett，E.C.等人。西伦敦肺癌筛查试点研究的基线结果移动扫描仪和双重风险模型利用的影响。肺癌148,12-19（2020）。文章

PubMed

Google Scholar

谷歌学者

Chiarantano, R. S. et al. Implementation of an integrated lung cancer prevention and screening program using a mobile computed tomography (CT) unit in Brazil. Cancer Control 29, 10732748221121385 (2022).Article

Chiarantano，R.S.等人。在巴西使用移动计算机断层扫描（CT）单元实施综合肺癌预防和筛查计划。癌症控制2910732748221121385（2022）。文章

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Wang, C. et al. Persistent increase and improved survival of stage I lung cancer based on a large-scale real-world sample of 26,226 cases. Chin. Med J. (Engl.) 136, 1937–1948 (2023).Article

Wang，C.等人。基于26226例大规模现实样本，I期肺癌的持续增加和生存率提高。下巴。医学杂志（英语）1361937-1948（2023）。文章

PubMed

Google Scholar

谷歌学者

Farjah, F. et al. Association of the intensity of diagnostic evaluation with outcomes in incidentally detected lung nodules. JAMA Intern. Med. 181, 480–489 (2021).Article

Farjah，F。等人。诊断评估强度与偶然发现的肺结节结果的关联。JAMA实习生。。文章

PubMed

Google Scholar

谷歌学者

Yankelevitz, D. F., Yip, R. & Henschke, C. I. Impact of duration of diagnostic workup on prognosis for early lung cancer. J. Thorac. Oncol. 18, 527–537 (2023).Article

Yankelevitz，D.F.，Yip，R。＆Henschke，C.I。诊断检查持续时间对早期肺癌预后的影响。J、胸部。Oncol公司。18527-537（2023）。文章

CAS

中科院

PubMed

Google Scholar

谷歌学者

Meyer, M. et al. Management of progressive pulmonary nodules found during and outside of CT lung cancer screening studies. J. Thorac. Oncol. 12, 1755–1765 (2017).Article

Meyer，M.等人。CT肺癌筛查研究期间和之外发现的进行性肺结节的管理。J、胸部。Oncol公司。121755-1765（2017）。文章

PubMed

Google Scholar

谷歌学者

Crosby, D. et al. Early detection of cancer. Science 375, eaay9040 (2022).Article

Crosby，D.等人，《癌症的早期发现》。科学375，eaay9040（2022）。文章

CAS

中科院

PubMed

Google Scholar

谷歌学者

Chabon, J. J. et al. Integrating genomic features for non-invasive early lung cancer detection. Nature 580, 245–251 (2020).Article

Chabon，J.J.等人。整合基因组特征用于非侵袭性早期肺癌检测。自然580245-251（2020）。文章

CAS

中科院

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

He, J. et al. Accurate classification of pulmonary nodules by a combined model of clinical, imaging, and cell-free DNA methylation biomarkers: a model development and external validation study. Lancet Digit. Health 5, 647–656 (2023).Article

He，J.等人。通过临床，影像学和无细胞DNA甲基化生物标志物的组合模型对肺结节进行准确分类：模型开发和外部验证研究。柳叶刀数字。健康5647-656（2023）。文章

Google Scholar

谷歌学者

Mazzone, P. J. et al. Clinical validation of a cell-free DNA fragmentome assay for augmentation of lung cancer early detection. Cancer Discov. (2024).Sidorenkov, G. et al. Multi-source data approach for personalized outcome prediction in lung cancer screening: update from the NELSON trial.

Mazzone，P.J。等人。无细胞DNA片段组检测增强肺癌早期检测的临床验证。癌症发现。（2024年）。Sidorenkov，G。等人。肺癌筛查中个性化结果预测的多源数据方法：来自NELSON试验的更新。

Eur. J. Epidemiol. 38, 445–454 (2023).Article .

欧洲流行病学杂志。38, 445-454 (2023).第[UNK]条。

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

King, G. & Zeng, L. Logistic regression in rare events data. Politic. Anal. 9, 137–163 (2001).Article

King，G。＆Zeng，L。罕见事件数据中的逻辑回归。政治。肛门。9137-163（2001）。文章

Google Scholar

谷歌学者

Zhang, R. et al. Deep learning for malignancy risk estimation of incidental sub-centimeter pulmonary nodules on CT images. Eur. Radio. 34, 4218–4229 (2024).Article

Zhang，R。等。CT图像上偶发亚厘米肺结节恶性风险评估的深度学习。欧洲电台。344218-4229（2024）。文章

Google Scholar

谷歌学者

Wang, C. et al. Deep learning for predicting subtype classification and survival of lung adenocarcinoma on computed tomography. Transl. Oncol. 14, 101141 (2021).Article

。翻译。Oncol公司。14101141（2021）。文章

CAS

中科院

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Pan, Z. et al. Predicting invasiveness of lung adenocarcinoma at chest CT with deep learning ternary classification models. Radiology 311, e232057 (2024).Article

Pan，Z.等。用深度学习三分类模型预测胸部CT肺腺癌的侵袭性。放射学311，e232057（2024）。文章

PubMed

Google Scholar

谷歌学者

Ouyang, X. et al. Dual-sampling attention network for diagnosis of COVID-19 from community acquired pneumonia. IEEE Trans. Med. Imag. 39, 2595–2605 (2020).Article

欧阳，X。等。社区获得性肺炎新型冠状病毒肺炎诊断的双重抽样注意网络。IEEE Trans。医学图像。392595-2605（2020）。文章

Google Scholar

谷歌学者

Mehrara, E., Forssell-Aronsson, E., Ahlman, H. & Bernhardt, P. Specific growth rate versus doubling time for quantitative characterization of tumor growth rate. Cancer Res. 67, 3970–3975 (2007).Article

Mehrara，E.，Forssell-Aronsson，E.，Ahlman，H。＆Bernhardt，P。特定生长速率与肿瘤生长速率定量表征的倍增时间。癌症研究673970-3975（2007）。文章

CAS

中科院

PubMed

Google Scholar

谷歌学者

Shi, F. et al. Deep learning empowered volume delineation of whole-body organs-at-risk for accelerated radiotherapy. Nat. Commun. 13, 6566 (2022).Article

Shi，F。等人。深度学习使加速放疗风险全身器官的体积描绘得以实现。。136566（2022年）。文章

CAS

中科院

PubMed

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Download referencesAcknowledgementsWe thank all those who participated in the construction of MCC and MSC. This research was supported by the National Natural Science Foundation of China (grant 92159302 to W.L., grant 82100119 to C.W., grant 82341083 to C.W., grant 62131015 to D.S. and grant U23A20295 to D.S.), National Key Research and Development Program of China (grant 2023YFF1204304 to F.S.), Science and Technology Project of Sichuan (grant 2022ZDZX0018 to W.L.), the Science and Technology Project of Chengdu (grant 2023-YF09-00007-SN to C.W.) and the 1.3.5 Project for Disciplines Excellence of West China Hospital, Sichuan University (grant ZYYC23027 to C.W.).Author informationAuthor notesThese authors contributed equally: Chengdi Wang, Jun Shao, Yichu He.Authors and AffiliationsDepartment of Pulmonary and Critical Care Medicine, Targeted Tracer Research and Development Laboratory, Frontiers Science Center for Disease-Related Molecular Network, State Key Laboratory of Respiratory Health and Multimorbidity, West China Hospital, West China School of Medicine, Sichuan University, Chengdu, ChinaChengdi Wang, Jun Shao, Xingting Liu, Liuqing Yang & Weimin LiFrontiers Medical Center, Tianfu Jincheng Laboratory, Chengdu, ChinaChengdi Wang & Weimin LiDepartment of Research and Development, United Imaging Intelligence, Shanghai, ChinaYichu He, Jiaojiao Wu, Ying Wei & Feng ShiSchool of Biomedical Engineering and State Key Laboratory of Advanced Medical Materials and Devices, ShanghaiTech University, Shanghai, ChinaXiang Sean Zhou, Yiqiang Zhan & Dinggang ShenShanghai Clinical Research and Trial Center, Shanghai, ChinaDinggang ShenAuthorsChengdi WangView author publicationsYou can also search for this author in.

下载参考文献致谢我们感谢所有参与MCC和MSC建设的人。这项研究得到了国家自然科学基金（授予W.L.92159302，授予C.W.82100119，授予C.W.82341083，授予D.S.62131015，授予D.S.U23A20295），国家重点研究发展计划（授予F.S.2023YFF1204304），四川省科技项目（授予W.L.2022ZDZX0018），成都市科技项目（授予C.W.2023-YF09-00007-SN）和四川大学华西医院1.3.5学科卓越项目（授予ZYZYX0018）的支持YC23027至C.W.）。作者信息作者注意到这些作者做出了同样的贡献：王成迪，邵军，何一楚。作者和附属机构四川大学华西医学院华西医院呼吸健康与多发病国家重点实验室疾病相关分子网络前沿科学中心肺部与重症监护医学系，靶向示踪剂研究与发展实验室，四川大学华西医院华西医院，成都，中国王成迪，邵军，刘兴庭，杨柳青和魏敏前沿医学中心，天府金城实验室Sean Zhou，Zhan Yiqiang Zhan＆Dinggang Shenshai中国上海临床研究与试验中心Dinggang Shenshai Authors Chengdi WangView作者出版物您也可以在中搜索该作者。

PubMed Google ScholarJun ShaoView author publicationsYou can also search for this author in

PubMed Google ScholarJun ShaoView作者出版物您也可以在

PubMed Google ScholarYichu HeView author publicationsYou can also search for this author in

PubMed Google ScholarYichu HeView作者出版物您也可以在

PubMed Google ScholarJiaojiao WuView author publicationsYou can also search for this author in

PubMed谷歌学者Jiaojiao WuView作者出版物您也可以在

PubMed Google ScholarXingting LiuView author publicationsYou can also search for this author in

PubMed Google ScholarXingting LiuView作者出版物您也可以在

PubMed Google ScholarLiuqing YangView author publicationsYou can also search for this author in

PubMed Google ScholarLiuqing YangView作者出版物您也可以在

PubMed Google ScholarYing WeiView author publicationsYou can also search for this author in

PubMed谷歌学术评论作者出版物您也可以在

PubMed Google ScholarXiang Sean ZhouView author publicationsYou can also search for this author in

PubMed谷歌学者Xiang Sean ZhouView作者出版物您也可以在

PubMed Google ScholarYiqiang ZhanView author publicationsYou can also search for this author in

PubMed Google ScholarYiqiang ZhanView作者出版物您也可以在

PubMed Google ScholarFeng ShiView author publicationsYou can also search for this author in

PubMed Google ScholarFeng ShiView作者出版物您也可以在

PubMed Google ScholarDinggang ShenView author publicationsYou can also search for this author in

PubMed Google ScholarDinggang ShenView作者出版物您也可以在

PubMed Google ScholarWeimin LiView author publicationsYou can also search for this author in

PubMed Google ScholarWeimin LiView作者出版物您也可以在

PubMed Google ScholarContributionsC.W., F.S., D.S. and W.L. conceived the idea and designed the experiments. C.W., J.S., Y.H. and J.W. implemented and performed the experiments. C.W., J.S., Y.H., J.W., X.L., L.Y., Y.W., X.S.Z. and Y.Z. analyzed the data and experimental results.

PubMed谷歌学术贡献中心。W、，F.S.，D.S.和W.L.构思了这个想法并设计了实验。C、 W.，J.S.，Y.H.和J.W.实施并执行了实验。C、 W.，J.S.，Y.H.，J.W.，X.L.，L.Y.，Y.W.，X.S.Z.和Y.Z.分析了数据和实验结果。

C.W., J.S., Y.H., J.W. and F.S. wrote the paper. All the authors reviewed, edited and approved the paper.Corresponding authorsCorrespondence to.

C、 W.，J.S.，Y.H.，J.W.和F.S.写了这篇论文。所有作者都审阅，编辑并批准了该论文。通讯作者通讯。

Chengdi Wang, Feng Shi, Dinggang Shen or Weimin Li.Ethics declarations

王成迪，冯石，沈定刚或李伟民。道德宣言

Competing interests

相互竞争的利益

F.S., Y.H., J.W. and Y.W. are employees of United Imaging Intelligence. The company had no involvement in the design, execution, surveillance, data analysis or interpretation of the study. The other authors have no competing interests.

F、 S.，Y.H.，J.W.和Y.W.是联合成像情报公司的员工。该公司没有参与研究的设计，执行，监督，数据分析或解释。其他作者没有相互竞争的利益。

Peer review

同行评审

Peer review information

同行评审信息

Nature Medicine thanks Florian Fintelmann, Colin Jacobs and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Primary Handling Editor: Lorenzo Righetto, in collaboration with the Nature Medicine team.

《自然医学》感谢Florian Fintelmann、Colin Jacobs和另一位匿名审稿人对这项工作的同行评审所做的贡献。主要处理编辑：Lorenzo Righetto与《自然医学》团队合作。

Additional informationPublisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.Extended dataExtended Data Fig. 1 Preprocessing and filtering of lung cancer screening datasets.Diagram describing the inclusion and exclusion in this study.Extended Data Fig.

Additional informationPublisher的注释Springer Nature在已发布地图和机构隶属关系中的管辖权主张方面保持中立。扩展数据扩展数据图1肺癌筛查数据集的预处理和过滤。描述本研究中包含和排除的图表。扩展数据图。

2 The ground truth for determining the risk of malignancy in pulmonary nodules.a, The flowchart of evaluation of the pulmonary nodules. b, The ground truth of the malignancy risk of pulmonary nodules annotated by clinicians. c, Purposes and ground truth of the two phases of C-Lung-RADS system. Phase 1 aimed to classify initial risk of nodules; Phase 2/2+ aimed to identify suspicious malignant nodules and refine risk level of nodules.Extended Data Fig.

。b、临床医生注释的肺结节恶性风险的基本事实。c、 c-Lung-RADS系统两个阶段的目的和基本事实。第一阶段旨在对结节的初始风险进行分类；阶段2/2+旨在识别可疑的恶性结节并提高结节的风险水平。扩展数据图。

3 Size distribution of solid components of mGGNs in the primary dataset and independent testing dataset.a, d, Size distribution of solid component in all mGGNs in the primary dataset and independent testing dataset, respectively (gray line, Gaussian fitting curve). b, e, Size of solid components of mGGNs with different ratings in the primary dataset and independent testing dataset, respectively.

3主数据集和独立测试数据集中mGGNs固体成分的尺寸分布。a，d，主数据集和独立测试数据集中所有mGGNs中固体成分的尺寸分布（灰线，高斯拟合曲线）。b、 e，分别在主数据集和独立测试数据集中具有不同评级的MGGN固体成分的大小。

The lines and plus signs in the box-and-whisker plots represent the median and mean values, respectively. The whiskers range from 25th percentile minus 1.5 times interquartile range (IQR) to 75th percentile plus 1.5 times IQR, and outliers below and above the whiskers are drawn as individual dots. The number of mGGNs with different ratings in the primary dataset and independent testing dataset could be referred to Supplementary Table 3.

盒子和胡须图中的线和加号分别代表中值和平均值。晶须的范围从第25百分位减去四分位间距（IQR）的1.5倍到第75百分位加上IQR的1.5倍，晶须下方和上方的异常值被绘制为单个点。在主要数据集和独立测试数据集中具有不同评级的MGGN的数量可以参考补充表3。

Statistical analyses were performed among four categories using Kruskal-Wallis H tests followed by Dunnett’s multiple comparison tests. Asterisks represent two-tailed adju.

使用Kruskal-Wallis H检验和Dunnett的多重比较检验对四个类别进行统计分析。星号代表双尾Adu。

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material.

开放获取本文是根据知识共享署名非商业性NoDerivatives 4.0国际许可证授权的，该许可证允许以任何媒介或格式进行任何非商业性使用，共享，分发和复制，只要您对原始作者和来源给予适当的信任，提供知识共享许可证的链接，并指出您是否修改了许可材料。

You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/..

要查看此许可证的副本，请访问http://creativecommons.org/licenses/by-nc-nd/4.0/..

Reprints and permissionsAbout this articleCite this articleWang, C., Shao, J., He, Y. et al. Data-driven risk stratification and precision management of pulmonary nodules detected on chest computed tomography.

转载和许可本文引用本文Wang，C.，Shao，J.，He，Y。等人。数据驱动的风险分层和胸部计算机断层扫描检测到的肺结节的精确管理。

Nat Med (2024). https://doi.org/10.1038/s41591-024-03211-3Download citationReceived: 31 July 2023Accepted: 22 July 2024Published: 17 September 2024DOI: https://doi.org/10.1038/s41591-024-03211-3Share this articleAnyone you share the following link with will be able to read this content:Get shareable linkSorry, a shareable link is not currently available for this article.Copy to clipboard.

Nat Med（2024）。https://doi.org/10.1038/s41591-024-03211-3Download引文接收日期：2023年7月31日接收日期：2024年7月22日发布日期：2024年9月17日OI：https://doi.org/10.1038/s41591-024-03211-3Share本文与您共享以下链接的任何人都可以阅读此内容：获取可共享链接对不起，本文目前没有可共享的链接。复制到剪贴板。

Provided by the Springer Nature SharedIt content-sharing initiative

由Springer Nature SharedIt内容共享计划提供

全球产业链接平台

重庆市渝北区金星科技大厦A区5楼512室

联系电话：023-67139735（重庆）

关于我们

产品服务