ISSN 1004-4140
CN 11-3017/P
| Citation: |
XUE R H, WU T T, CHAI J, et al. Radiomics-based Nomogram for Predicting Invasiveness of Subsolid Pulmonary Nodules[J]. CT Theory and Applications, 2024, 33(3): 343-350. DOI: 10.15953/j.ctta.2023.213. (in Chinese).
|
Objective: The study aimed to develop and evaluate a clinical diagnostic model that combined computer tomography (CT) radiomic features with a nomogram for predicting invasiveness in subsolid pulmonary nodules. Methods: This retrospective study analyzed both clinical and imaging data from patients at our institution who were diagnosed with pathologically confirmed subsolid pulmonary nodules at our institution based on thin-slice CT images. Radiomic features were extracted from these CT images, and LASSO regression with K-fold cross-validation was used to select the most informative features. Three predictive models were constructed via multivariate logistic regression: the first incorporated clinical parameters and conventional imaging features; the second relied solely on radiomic characteristics; and the third was a hybrid clinical-radiomics model. The logistic regression results were visually represented using a nomogram. Receiver operating characteristic curves were utilized to compare the classification predictive performance of the three models for distinguishing ground-glass opacity lung adenocarcinoma IA and non-IA cases. Decision curve analysis (DCA) was employed to assess the clinical utility of these models across different cohorts. Results: A total of 204 subsolid pulmonary nodules from 192 patients were included. They were divided into invasive (n=114) and non-invasive groups (n=90) based on pathological typing. These nodules were divided into a training set (n=143, IA:non-IA 77∶66) and a test set (n=61, IA:non-IA 38∶23). A total of 1316 features were initially extracted from each subsolid nodule. Subsequently, two independent clinical predictors (mean CT value and maximum diameter) and three radiomic features were selected through feature selection and logistic regression for model building. The combined clinical-radiomics model demonstrated superior discriminative capability (AUC=0.920, 95%CI: 0.818~0.931) in distinguishing IA from non-IA within the training set compared to the radiomics model and the clinical model independently (AUC=0.907, 95%CI: 0.792~0.914; AUC=0.822, 95%CI: 0.764~0.895). In the test set, the inclusion of clinical data improved the diagnostic efficacy of the radiomics model. DCA demonstrated that the combined model generally provided greater clinical benefits in most scenarios. Conclusion: The developed clinical-radiomics joint model showed promising performance in predicting the subsolid pulmonary nodule invasiveness.
| [1] |
HANSELL D M, BANKIER A A, MACMAHON H, et al. Fleischner society: Glossary of terms for thoracic imaging[J]. Radiology, 2008, 246(3): 697−722. DOI: 10.1148/radiol.2462070712.
|
| [2] |
SIEGEL R L, MILLER K D, FUCHS H E, et al. Cancer statistics, 2022[J]. CA: A Cancer Journal for Clinicians, 2022, 72(1): 7−33. DOI: 10.3322/caac.21708.
|
| [3] |
NICHOLSON A G, TSAO M S, BEASLEY M B, et al. The 2021 WHO classification of lung tumors: Impact of advances since 2015[J]. Journal of Thoracic Oncology, 2022, 17(3): 362−387. DOI: 10.1016/j.jtho.2021.11.003.
|
| [4] |
WEICHERT W, WARTH A. Early lung cancer with lepidic pattern: Adenocarcinoma in situ, minimally invasive adenocarcinoma, and lepidic predominant adenocarcinoma[J]. Current Opinion in Pulmonary Medicine, 2014, 20(4): 309−316. DOI: 10.1097/MCP.0000000000000065.
|
| [5] |
LIU S, WANG R, ZHANG Y, et al. Precise diagnosis of intraoperative frozen section is an effective method to guide resection strategy for peripheral small-sized lung adenocarcinoma[J]. Journal of Clinical Oncology, 2016, 34(4): 307−313. DOI: 10.1200/JCO.2015.63.4907.
|
| [6] |
LAMBIN P, RIOS-VELAZQUEZ E, LEIJENAAR R, et al. Radiomics: Extracting more information from medical images using advanced feature analysis[J]. European Journal of Cancer, 2012, 48(4): 441−446. DOI: 10.1016/j.ejca.2011.11.036.
|
| [7] |
WANG T, SHE Y, YANG Y, et al. Radiomics for survival risk stratification of clinical and pathologic stage IA pure-solid non-small cell lung cancer[J]. Radiology, 2022, 302(2): 425−434. DOI: 10.1148/radiol.2021210109.
|
| [8] |
YANG L, YANG J, ZHOU X, et al. Development of a radiomics nomogram based on the 2D and 3D CT features to predict the survival of non-small cell lung cancer patients[J]. European Radiology, 2019, 29(5): 2196−2206. DOI: 10.1007/s00330-018-5770-y.
|
| [9] |
SUN Y, MA Z, ZHAO W, et al. Computed tomography radiomics in growth prediction of pulmonary ground-glass nodules[J]. European Journal of Radiology, 2023, 159: 110684. DOI: 10.1016/j.ejrad.2022.110684.
|
| [10] |
WU F Z, WU Y J, TANG E K. An integrated nomogram combined semantic-radiomic features to predict invasive pulmonary adenocarcinomas in subjects with persistent subsolid nodules[J]. Quantitative Imaging in Medicine and Surgery, 2023, 13(2): 654−668. DOI: 10.21037/qims-22-308.
|
| [11] |
SUN Y, LI C, JIN L, et al. Radiomics for lung adenocarcinoma manifesting as pure ground-glass nodules: Invasive prediction[J]. European Radiology, 2020, 30(7): 3650−3659. DOI: 10.1007/s00330-020-06776-y.
|
| [12] |
张晓菊. 《肺结节诊治中国专家共识(2018版)》解读[J]. 中华实用诊断与治疗杂志, 2019, 33(1): 1−3. DOI: 10.13507/j.issn.1674-3474.2019.01.001.
|
| [13] |
YUE X, LIU S, LIU S, et al. HRCT morphological characteristics distinguishing minimally invasive pulmonary adenocarcinoma from invasive pulmonary adenocarcinoma appearing as subsolid nodules with a diameter of ≤3 cm[J]. Clinical Radiology, 2018, 73(4): 411.e7-411.e15. DOI: 10.1016/j.crad.2017.11.014.
|
| [14] |
LI X, ZHANG W, YU Y, et al. CT features and quantitative analysis of subsolid nodule lung adenocarcinoma for pathological classification prediction[J]. BMC Cancer, 2020, 20(1): 60. DOI: 10.1186/s12885-020-6556-6.
|
| [15] |
ZHENG H, ZHANG H, WANG S, et al. Invasive prediction of ground glass nodule based on clinical characteristics and radiomics feature[J]. Frontiers in Genetics, 2022, 12: 783391. DOI: 10.3389/fgene.2021.783391.
|
| [16] |
PENG X, YANG S, ZHOU L, et al. Repeatability and reproducibility of computed tomography radiomics for pulmonary nodules: A multicenter phantom study[J]. Investigative Radiology, 2022, 57(4): 242−253. DOI: 10.1097/RLI.0000000000000834.
|
| [17] |
XU Y, LU L, SUN S H, et al. Effect of CT image acquisition parameters on diagnostic performance of radiomics in predicting malignancy of pulmonary nodules of different sizes[J]. European Radiology, 2022, 32(3): 1517−1527. DOI: 10.1007/s00330-021-08274-1.
|
Supported by: Beijing Renhe Information Technology Co. Ltd 
DownLoad: