Share：
Share this content in WeChat
X
[Chinese] [PDF] 5 1
• Clinical Articles •
Differentiating pulmonary adenocarcinoma from inflammatory pulmonary masses using a multi-sequence MRI radiomics model
LI Huan  CUI Can  LIU Rong  GU Yan  ZHAO Ruiyi  ZENG Liang 

Cite this article as: LI H, CUI C, LIU R, et al. Differentiating pulmonary adenocarcinoma from inflammatory pulmonary masses using a multi-sequence MRI radiomics model[J]. Chin J Magn Reson Imaging, 2026, 17(3): 30-38. DOI:10.12015/issn.1674-8034.2026.03.005.


[Abstract] Objective To investigate the value of a multi-sequence MRI-based radiomics model in differentiating pulmonary adenocarcinoma from inflammatory pulmonary masses.Materials and Methods A retrospective analysis was conducted on 136 patients with pathologically confirmed lung adenocarcinoma and inflammatory pulmonary masses, whose clinical and MRI imaging data were collected for evaluation. All patients were randomly divided into a training set (n = 96) and a test set (n = 40) in a 7∶3 ratio. Through univariate and multivariate logistic regression analyses, indicators with discriminatory significance (P < 0.05) in the clinical and MRI features of patients with lung adenocarcinoma and inflammatory lung occupying lesions were screened. Tumor regions of interest (ROI) were delineated on four sequence images: T1-weighted imaging (T1WI), T2-weighted imaging (T2WI), apparent diffusion coefficient (ADC) maps, and dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). Radiomics features were then extracted using PyRadiomics. Highly correlated redundant features were removed based on the Spearman correlation coefficient. Subsequently, the remaining features were screened using ten-fold cross-validation and the least absolute shrinkage and selection operator (LASSO). Logistic regression (LR) was then applied to construct single-sequence and combined multi-sequence radiomics models. Model performance was evaluated using the receiver operating characteristic (ROC) curve and the area under the curve (AUC), based on which the optimal radiomics model was selected. Finally, clinical, clinical-imaging, and clinical-imaging-radiomics combined model were developed by integrating clinical features, MRI features, and the optimal radiomics model, and a visual nomogram was constructed. Statistical comparison of AUCs was performed using the DeLong test, and the clinical utility of the models was assessed via decision curve analysis (DCA).Results Age, smoking history, straightening sign, ADC value, and time-intensity curve (TIC) type were identified as independent predictors for pulmonary adenocarcinoma (P < 0.05). Among the radiomics models, the multi-sequence MRI (T1WI+T2WI+ADC+DCE-MRI) model achieved the best diagnostic performance, with AUCs of 0.888 and 0.738 in the training and test sets, respectively. Further integration of clinical, MRI, and radiomics features yielded a combined model, analysis of which showed that the combined model had higher predictive performance (P < 0.05), with AUCs reaching 0.924 and 0.853 in the training and test sets, respectively.Conclusions The combined model based on clinical, MRI features, and multi-sequence MRI radiomics shows good diagnostic efficacy in differentiating pulmonary adenocarcinoma from inflammatory pulmonary masses.
[Keywords] lung adenocarcinoma；pulmonary inflammatory mass；magnetic resonance imaging；radiomics；nomogram

LI Huan   CUI Can   LIU Rong   GU Yan   ZHAO Ruiyi   ZENG Liang*  

Department of Radiology, Affiliated Hospital of Nanjing University of Chinese Medicine, Nanjing 210029, China

Corresponding author: ZENG L, E-mail: ZengL8@126.com

Conflicts of interest   None.

Received  2025-11-26
Accepted  2026-03-03
DOI: 10.12015/issn.1674-8034.2026.03.005
Cite this article as: LI H, CUI C, LIU R, et al. Differentiating pulmonary adenocarcinoma from inflammatory pulmonary masses using a multi-sequence MRI radiomics model[J]. Chin J Magn Reson Imaging, 2026, 17(3): 30-38. DOI:10.12015/issn.1674-8034.2026.03.005.

[1]
BRAY F, LAVERSANNE M, SUNG H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J]. CA Cancer J Clin, 2024, 74(3): 229-263. DOI: 10.3322/caac.21834.
[2]
SIEGEL R L, MILLER K D, WAGLE N S, et al. Cancer statistics, 2023[J]. CA Cancer J Clin, 2023, 73(1): 17-48. DOI: 10.3322/caac.21763.
[3]
SIEGEL R L, GIAQUINTO A N, JEMAL A. Cancer statistics, 2024[J]. CA Cancer J Clin, 2024, 74(1): 12-49. DOI: 10.3322/caac.21820.
[4]
WANG C D, SHAO J, SONG L J, et al. Persistent increase and improved survival of stage I lung cancer based on a large-scale real-world sample of 26, 226 cases[J]. Chin Med J, 2023, 136(16): 1937-1948. DOI: 10.1097/cm9.0000000000002729.
[5]
GOTO E, TAKAMOCHI K, KISHIKAWA S, et al. Stepwise progression of invasive mucinous adenocarcinoma based on radiological and biological characteristics[J]. Lung Cancer, 2023, 184: 107348. DOI: 10.1016/j.lungcan.2023.107348.
[6]
KIFJAK D, MURA R, POCHEPNIA S, et al. Beyond the usual - Atypical imaging presentation in lung cancer and implications for TNM-staging[J]. Curr Opin Oncol, 2026, 38(1): 31-38. DOI: 10.1097/CCO.0000000000001203.
[7]
LAMBIN P, LEIJENAAR R T H, DEIST T M, et al. Radiomics: the bridge between medical imaging and personalized medicine[J]. Nat Rev Clin Oncol, 2017, 14(12): 749-762. DOI: 10.1038/nrclinonc.2017.141.
[8]
SHUR J D, DORAN S J, KUMAR S, et al. Radiomics in oncology: a practical guide[J]. Radiographics, 2021, 41(6): 1717-1732. DOI: 10.1148/rg.2021210037.
[9]
WARKENTIN M T, AL-SAWAIHEY H, LAM S, et al. Radiomics analysis to predict pulmonary nodule malignancy using machine learning approaches[J]. Thorax, 2024, 79(4): 307-315. DOI: 10.1136/thorax-2023-220226.
[10]
LIU Z, YANG L, LIANG J P, et al. Radiomic features add incremental benefit to conventional radiological feature-based differential diagnosis of lung nodules[J]. Eur Radiol, 2025, 35(6): 2968-2978. DOI: 10.1007/s00330-024-11221-5.
[11]
WU S H, WANG X Y, SHAN W L, et al. Computed tomography-derived radiomics models for distinguishing difficult-to-diagnose inflammatory and malignant pulmonary nodules[J/OL]. Biomed Eng Comput Biol, 2025, 16: 11795972251371467 [2025-11-25]. https://pubmed.ncbi.nlm.nih.gov/40936589/. DOI: 10.1177/11795972251371467.
[12]
ZHANG M Q, LIU J N, SUN Y J. Advances of MRI in pulmonary diseases[J]. Chin J Med Imaging Technol, 2025, 41(5): 839-842. DOI: 10.13929/j.issn.1003-3289.2025.05.031.
[13]
AZOUR L, OHNO Y, BIEDERER J, et al. Lung MRI: indications, capabilities, and techniques-AJR expert panel narrative review[J/OL]. AJR Am J Roentgenol, 2025, 225(4): e2532637 [2025-11-25]. https://pubmed.ncbi.nlm.nih.gov/40397559/. DOI: 10.2214/AJR.25.32637.
[14]
ZHOU J X, WEN Y, DING R L, et al. Radiomics signature based on robust features derived from diffusion data for differentiation between benign and malignant solitary pulmonary lesions[J/OL]. Cancer Imaging, 2024, 24(1): 14 [2025-11-25]. https://pubmed.ncbi.nlm.nih.gov/38246984/. DOI: 10.1186/s40644-024-00660-4.
[15]
LI J, XIA Y, XU M L, et al. Application value of diffusion-weighted imaging reconstructed based on deep learning in benign and malignant differentiation of pulmonary lesions[J]. Chin J Magn Reson Imaging, 2024, 15(10): 15-21. DOI: 10.12015/issn.1674-8034.2024.10.004.
[16]
YAN Q Q, YI Y Q, SHEN J, et al. Preliminary study of 3T-MRI native T1-mapping radiomics in differential diagnosis of non-calcified solid pulmonary nodules/masses[J/OL]. Cancer Cell Int, 2021, 21(1): 539 [2025-11-25]. https://pubmed.ncbi.nlm.nih.gov/34663307/. DOI: 10.1186/s12935-021-02195-1.
[17]
TANG X, XU X P, HAN Z P, et al. Elaboration of a multimodal MRI-based radiomics signature for the preoperative prediction of the histological subtype in patients with non-small-cell lung cancer[J/OL]. BioMedical Eng OnLine, 2020, 19(1): 5 [2025-11-25]. https://pubmed.ncbi.nlm.nih.gov/31964407/. DOI: 10.1186/s12938-019-0744-0.
[18]
SONG C R, CHENG P, CHENG J L, et al. Differential diagnosis of nasopharyngeal carcinoma and nasopharyngeal lymphoma based on DCE-MRI and RESOLVE-DWI[J]. Eur Radiol, 2020, 30(1): 110-118. DOI: 10.1007/s00330-019-06343-0.
[19]
JIANG J Q, FU Y G, HU X Y, et al. The value of diffusion-weighted imaging based on monoexponential and biexponential models for the diagnosis of benign and malignant lung nodules and masses[J/OL]. Br J Radiol, 2020, 93(1110): 20190400 [2025-11-25]. https://pubmed.ncbi.nlm.nih.gov/32163295/. DOI: 10.1259/bjr.20190400.
[20]
DU Y H, ZHANG S, LIANG T, et al. Dynamic contrast-enhanced MRI perfusion parameters are imaging biomarkers for angiogenesis in lung cancer[J]. Acta Radiol, 2023, 64(2): 572-580. DOI: 10.1177/02841851221088581.
[21]
XIE K X, CUI C, LI X Q, et al. MRI-based clinical-imaging-radiomics nomogram model for discriminating between benign and malignant solid pulmonary nodules or masses[J]. Acad Radiol, 2024, 31(10): 4231-4241. DOI: 10.1016/j.acra.2024.03.042.
[22]
LI J R, SINA A A I, ANTAW F, et al. Digital decoding of single extracellular vesicle phenotype differentiates early malignant and benign lung lesions[J/OL]. Adv Sci, 2023, 10: 2204207 [2025-11-25]. https://pubmed.ncbi.nlm.nih.gov/36394090/. DOI: 10.1002/advs.202204207.
[23]
BRONCANO J, STEINBRECHER K, MARQUIS K M, et al. Diffusion-weighted imaging of the chest: a primer for radiologists[J/OL]. Radiographics, 2023, 43(7): e220138 [2025-11-25]. https://pubmed.ncbi.nlm.nih.gov/37347699/. DOI: 10.1148/rg.220138.
[24]
DONG Z Y, LI Y X, SHI X Y, et al. Research progress of diffusion-weighted magnetic resonance imaging technology in the diagnosis and treatment of lung cancer[J]. Chin J Magn Reson Imaging, 2025, 16(8): 215-220. DOI: 10.12015/issn.1674-8034.2025.08.032.
[25]
GAO Y Q, LU J, XU H, et al. Differentiating pulmonary inflammatory nodules from lung cancer based on whole-focus dynamic enhanced MRI intensity histogram[J]. Chin J Magn Reson Imaging, 2023, 14(7): 42-48. DOI: 10.12015/issn.1674-8034.2023.07.008.
[26]
LI F, WANG L, PAN J. Value of dynamic contrast-enhanced MRI in the differential diagnosis of breast mass-like adenosis and breast cancer[J]. J Clin Radiol, 2022, 41(7): 1281-1285. DOI: 10.13437/j.cnki.jcr.2022.07.014.
[27]
HUANG Q, XU N, YIN J, et al. Radiomics reveals the biological basis for non-small cell lung cancer prognostic stratification by reflecting tumor immune microenvironment heterogeneity[J/OL]. Front Immunol, 2025, 16: 1708692 [2025-11-25]. https://pubmed.ncbi.nlm.nih.gov/41293176/. DOI: 10.3389/fimmu.2025.1708692.
[28]
LEE S H, RIMNER A, GELB E, et al. Correlation between tumor metabolism and semiquantitative perfusion magnetic resonance imaging metrics in non-small cell lung cancer[J]. Int J Radiat Oncol, 2018, 102(4): 718-726. DOI: 10.1016/j.ijrobp.2018.02.031.
[29]
PETRALIA G, SUMMERS P E, AGOSTINI A, et al. Dynamic contrast-enhanced MRI in oncology: how we do it[J]. La Radiol Med, 2020, 125(12): 1288-1300. DOI: 10.1007/s11547-020-01220-z.
[30]
WENGERT G J, DABI Y, KERMARREC E, et al. O-RADS MRI classification of indeterminate adnexal lesions: time-intensity curve analysis is better than visual assessment[J]. Radiology, 2022, 303(3): 566-575. DOI: 10.1148/radiol.210342.
[31]
HERTEL A, STREUER A, DIEHL S, et al. Targeting tumoral heterogeneity in lung cancer: a novel, CT-texture-guided targeted biopsy approach with exome sequencing[J/OL]. NPJ Precis Oncol, 2025, 9(1): 342 [2025-11-25]. https://pubmed.ncbi.nlm.nih.gov/41203783/. DOI: 10.1038/s41698-025-01148-5.
[32]
TREBESCHI S, DRAGO S G, BIRKBAK N J, et al. Predicting response to cancer immunotherapy using noninvasive radiomic biomarkers[J]. Ann Oncol, 2019, 30(6): 998-1004. DOI: 10.1093/annonc/mdz108.
[33]
BEIG N, KHORRAMI M, ALILOU M, et al. Perinodular and intranodular radiomic features on lung CT images distinguish adenocarcinomas from granulomas[J]. Radiology, 2019, 290(3): 783-792. DOI: 10.1148/radiol.2018180910.
[34]
WANG X H, WAN Q, CHEN H J, et al. Classification of pulmonary lesion based on multiparametric MRI: utility of radiomics and comparison of machine learning methods[J]. Eur Radiol, 2020, 30(8): 4595-4605. DOI: 10.1007/s00330-020-06768-y.
[35]
FENG F, QIANG F L, SHEN A J, et al. Dynamic contrast-enhanced MRI versus 18F-FDG PET/CT: Which is better in differentiation between malignant and benign solitary pulmonary nodules [J]. Chin J Cancer Res, 2018, 30(1): 21-30. DOI: 10.21147/j.issn.1000-9604.2018.01.03.
[36]
LIU X C, KAN X F, WANG Y J, et al. The application value of MRI multi-parameters in the differential diagnosis of solid solitary pulmonary nodule[J]. J Clin Radiol, 2025, 44(1): 87-94. DOI: 10.13437/j.cnki.jcr.2025.01.030.
[37]
ZHANG X Y, TENG X Z, ZHANG J, et al. Enhancing pathological complete response prediction in breast cancer: the role of dynamic characterization of DCE-MRI and its association with tumor heterogeneity[J/OL]. Breast Cancer Res, 2024, 26(1): 77 [2025-11-25]. https://pubmed.ncbi.nlm.nih.gov/38745321/. DOI: 10.1186/s13058-024-01836-3.
[38]
MASSON-GREHAIGNE C, LAFON M, PALUSSIÈRE J, et al. Single- and multi-site radiomics may improve overall survival prediction for patients with metastatic lung adenocarcinoma[J]. Diagn Interv Imaging, 2024, 105(11): 439-452. DOI: 10.1016/j.diii.2024.07.005.

PREV An experimental study of intravoxel incoherent motion diffusion-weighted imaging for evaluating the therapeutic effects of bone marrow mesenchymal stem cell in a rat model of peripheral nerve injury
NEXT Intravoxel incoherent motion histogram parameters combined with clinical features to predict the response of neoadjuvant therapy in patients with rectal cancer
  


LINK


Tel & Fax: +8610-67113815    E-mail: editor@cjmri.cn