Share：
Share this content in WeChat
X
[Chinese] [PDF] 9 1
• Clinical Article •
Predictive value of radiomics based on dynamic contrast-enhanced MRI for high-grade gliomas and solitary brain metastases
XU Chenxue  LU Haitao  XING Wei  ZHANG Yanwen  WANG Qiang 

DOI:10.12015/issn.1674-8034.2025.12.011.


[Abstract] Objective To construct a radiomics model by extracting high-throughput radiomics features from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) sequences, and further explore its value in preoperatively differentiating high-grade gliomas (HGG) from solitary brain metastases (SBM).Materials and Methods This retrospective study included 135 patients (HGG, n = 89; SBM, n = 46) treated at Changzhou First People's Hospital from May 2016 to December 2024, all with complete imaging data (DCE-MRI and conventional MRI) and histopathology. Two neuroradiologists (3 and 10 years of experience) manually delineated whole-tumor regions of interest, from which 963 radiomics features were extracted from each region of interest (ROI). Feature redundancy was reduced using the Mann-Whitney U test, Pearson correlation analysis, and hierarchical clustering, followed by least absolute shrinkage and selection operator (LASSO) for optimization. Subsequently, logistic regression (LR) was employed to establish three models: the DCE model, conventional MRI model, and fusion model. Model performance was evaluated by receiver operating characteristic analysis, and differences were compared with the DeLong test (P < 0.05 significant).Results In the test set, both the fusion model and the DCE model demonstrated high diagnostic performance [area under the curves (AUCs) are 0.934 (95% CI: 0.860 to 1.000) and 0.908 (95% CI: 0.812 to 1.000), respectively], each significantly outperforming the conventional MRI model (P = 0.001 and P = 0.011, respectively). Although the fusion model showed a numerically higher AUC than the DCE model, the difference was not statistically significant (P = 0.387).Conclusions The DCE model demonstrated high performance in preoperatively discriminating between high-grade gliomas and solitary brain metastases, achieving comparable diagnostic efficacy to the fused model.
[Keywords] radiomics；high-grade glioma；solitary brain metastasis；magnetic resonance imaging；dynamic contrast-enhanced

XU Chenxue1   LU Haitao1, 2*   XING Wei1, 2   ZHANG Yanwen1, 2   WANG Qiang3, 4  

1 Department of Medical Imaging, the Third Affiliated Hospital of Soochow University, Changzhou 213003, China

2 Department of Medical Imaging, the First People's Hospital of Changzhou, Changzhou 213003, China

3 Department of Neurosurgery, the Third Affiliated Hospital of Soochow University, Changzhou 213003, China

4 Department of Neurosurgery, the First People's Hospital of Changzhou, Changzhou 213003, China

Corresponding author: LU H T, E-mail: vluhaitao@163.com

Conflicts of interest   None.

Received  2025-09-21
Accepted  2025-12-02
DOI: 10.12015/issn.1674-8034.2025.12.011
DOI:10.12015/issn.1674-8034.2025.12.011.

[1]
WANG B, WANG Z J, JIA Y H, et al. Water exchange detected by shutter speed dynamic contrast enhanced-MRI help distinguish solitary brain metastasis from glioblastoma[J/OL]. Eur J Radiol, 2022, 156: 110526 [2025-09-20]. https://pubmed.ncbi.nlm.nih.gov/36219917/. DOI: 10.1016/j.ejrad.2022.110526.
[2]
LIU X X, ZHANG Q R, LI J R, et al. Coordinatized lesion location analysis empowering ROI-based radiomics diagnosis on brain gliomas[J]. Eur Radiol, 2023, 33(12): 8776-8787. DOI: 10.1007/s00330-023-09871-y.
[3]
ZHANG S R, ZHOU Y S, ZHANG R D, et al. Value of multi-parameter MRI imaging model in differentiating high-grade glioma from primary central nervous system lymphoma[J]. Chin J Magn Reson Imaging, 2023, 14(10): 53-57, 64. DOI: 10.12015/issn.1674-8034.2023.10.010.
[4]
ZHANG Y Z, ZHANG H B, ZHANG H W, et al. Glioblastoma and solitary brain metastasis: differentiation by integrating demographic-MRI and deep-learning radiomics signatures[J]. J Magn Reson Imaging, 2024, 60(3): 909-920. DOI: 10.1002/jmri.29123.
[5]
QIAN Z H, LI Y M, WANG Y Z, et al. Differentiation of glioblastoma from solitary brain metastases using radiomic machine-learning classifiers[J/OL]. Cancer Lett, 2019, 451: 128-135 [2025-09-20]. https://pubmed.ncbi.nlm.nih.gov/30878526/. DOI: 10.1016/j.canlet.2019.02.054.
[6]
FORDHAM A J, HACHERL C C, PATEL N, et al. Differentiating glioblastomas from solitary brain metastases: an update on the current literature of advanced imaging modalities[J/OL]. Cancers (Basel), 2021, 13(12): 2960 [2025-09-20]. https://pubmed.ncbi.nlm.nih.gov/34199151/. DOI: 10.3390/cancers13122960.
[7]
BAI J, HE M Y, GAO E Y, et al. High-performance presurgical differentiation of glioblastoma and metastasis by means of multiparametric neurite orientation dispersion and density imaging (NODDI) radiomics[J]. Eur Radiol, 2024, 34(10): 6616-6628. DOI: 10.1007/s00330-024-10686-8.
[8]
WANG S C, XIAO F, SUN W B, et al. Radiomics analysis based on magnetic resonance imaging for preoperative overall survival prediction in isocitrate dehydrogenase wild-type glioblastoma[J/OL]. Front Neurosci, 2022, 15: 791776 [2025-09-20]. https://pubmed.ncbi.nlm.nih.gov/35153659/. DOI: 10.3389/fnins.2021.791776.
[9]
BAE S, AN C, AHN S S, et al. Robust performance of deep learning for distinguishing glioblastoma from single brain metastasis using radiomic features: model development and validation[J/OL]. Sci Rep, 2020, 10(1): 12110 [2025-09-20]. https://pubmed.ncbi.nlm.nih.gov/32694637/. DOI: 10.1038/s41598-020-68980-6.
[10]
WANG M, MIAO Y W. Advances in multi-sequence MRI for differentiating high-grade glioma and brain metastasis[J]. Chin J Magn Reson Imag, 2025, 16(2): 159-164. DOI: 10.12015/issn.1674-8034.2025.02.026.
[11]
ARTZI M, BRESSLER I, BASHAT D BEN. Differentiation between glioblastoma, brain metastasis and subtypes using radiomics analysis[J]. J Magn Reson Imaging, 2019, 50(2): 519-528. DOI: 10.1002/jmri.26643.
[12]
TAN R, SUI C X, WANG C, et al. MRI-based intratumoral and peritumoral radiomics for preoperative prediction of glioma grade: a multicenter study[J/OL]. Front Oncol, 2024, 14: 1401977 [2025-09-20]. https://pubmed.ncbi.nlm.nih.gov/38803534/. DOI: 10.3389/fonc.2024.1401977.
[13]
ZHOU J J, ZHANG Y, CHANG K T, et al. Diagnosis of benign and malignant breast lesions on DCE-MRI by using radiomics and deep learning with consideration of peritumor tissue[J]. J Magn Reson Imaging, 2020, 51(3): 798-809. DOI: 10.1002/jmri.26981.
[14]
JIANG X Y, SHAO Z Q, CHAI Y T, et al. Non-contrast CT-based radiomic signature of pericoronary adipose tissue for screening non-calcified plaque[J/OL]. Phys Med Biol, 2022, 67(10) [2025-09-20]. https://pubmed.ncbi.nlm.nih.gov/35453134/. DOI: 10.1088/1361-6560/ac69a7.
[15]
TAHA B, BOLEY D, SUN J, et al. State of radiomics in glioblastoma[J]. Neurosurgery, 2021, 89(2): 177-184. DOI: 10.1093/neuros/nyab124.
[16]
HAN Y Q, ZHANG L L, NIU S Z, et al. Differentiation between glioblastoma multiforme and metastasis from the lungs and other sites using combined clinical/routine MRI radiomics[J/OL]. Front Cell Dev Biol, 2021, 9: 710461 [2025-09-20]. https://pubmed.ncbi.nlm.nih.gov/34513840/. DOI: 10.3389/fcell.2021.710461.
[17]
JIN X, LI W Z, GUO Y H, et al. Predicting progression-free survival using dynamic contrast-enhanced imaging-based radiomics in advanced nasopharyngeal carcinoma patients treated with nimotuzumab[J]. Eur Radiol, 2025, 35(8): 5135-5145. DOI: 10.1007/s00330-025-11433-3.
[18]
YANG X Q, WANG X F, ZUO Z C, et al. Radiomics-based analysis of dynamic contrast-enhanced magnetic resonance image: a prediction nomogram for lymphovascular invasion in breast cancer[J/OL]. Magn Reson Imag, 2024, 112: 89-99 [2025-09-20]. https://pubmed.ncbi.nlm.nih.gov/38971267/. DOI: 10.1016/j.mri.2024.07.001.
[19]
WANG J, HU Y, ZHOU X J, et al. A radiomics model based on DCE-MRI and DWI may improve the prediction of estimating IDH1 mutation and angiogenesis in gliomas[J/OL]. Eur J Radiol, 2022, 147: 110141 [2025-09-20]. https://pubmed.ncbi.nlm.nih.gov/34995947/. DOI: 10.1016/j.ejrad.2021.110141.
[20]
ZHONG S, REN J X, YU Z P, et al. Predicting glioblastoma molecular subtypes and prognosis with a multimodal model integrating convolutional neural network, radiomics, and semantics[J]. J Neurosurg, 2022, 139(2): 305-314. DOI: 10.3171/2022.10.JNS22801.
[21]
ZHANG H B, ZHANG H W, ZHANG Y Z, et al. Deep learning radiomics for the assessment of telomerase reverse transcriptase promoter mutation status in patients with glioblastoma using multiparametric MRI[J]. J Magn Reson Imaging, 2023, 58(5): 1441-1451. DOI: 10.1002/jmri.28671.
[22]
TUSTISON N J, AVANTS B B, COOK P A, et al. N4ITK: improved N3 bias correction[J]. IEEE Trans Med Imaging, 2010, 29(6): 1310-1320. DOI: 10.1109/TMI.2010.2046908.
[23]
YU Y F, TAN Y J, XIE C M, et al. Development and validation of a preoperative magnetic resonance imaging radiomics-based signature to predict axillary lymph node metastasis and disease-free survival in patients with early-stage breast cancer[J/OL]. JAMA Netw Open, 2020, 3(12): e2028086 [2025-09-20]. https://pubmed.ncbi.nlm.nih.gov/33289845/. DOI: 10.1001/jamanetworkopen.2020.28086.
[24]
PENG H, HUO J H, LI B, et al. Predicting isocitrate dehydrogenase (IDH) mutation status in gliomas using multiparameter MRI radiomics features[J]. J Magn Reson Imaging, 2021, 53(5): 1399-1407. DOI: 10.1002/jmri.27434.
[25]
XU Z C, ZHANG Y, LIU Q, et al. Differential study of high-grade glioma and metastatic tumor based on MRI and semantic features[J]. Chin J Magn Reson Imaging, 2024, 15(8): 103-109. DOI: 10.12015/issn.1674-8034.2024.08.016.
[26]
KANG K M, CHOI S H, CHUL-KEE P, et al. Differentiation between glioblastoma and primary CNS lymphoma: application of DCE-MRI parameters based on arterial input function obtained from DSC-MRI[J]. Eur Radiol, 2021, 31(12): 9098-9109. DOI: 10.1007/s00330-021-08044-z.
[27]
JUNG B C, AREVALO-PEREZ J, LYO J K, et al. Comparison of glioblastomas and brain metastases using dynamic contrast-enhanced perfusion MRI[J]. J Neuroimaging, 2016, 26(2): 240-246. DOI: 10.1111/jon.12281.
[28]
ESSIG M, SHIROISHI M S, NGUYEN T B, et al. Perfusion MRI: the five most frequently asked technical questions[J]. AJR Am J Roentgenol, 2013, 200(1): 24-34. DOI: 10.2214/AJR.12.9543.
[29]
XU Z W, KE C, LIU J, et al. Diagnostic performance between MR amide proton transfer (APT) and diffusion kurtosis imaging (DKI) in glioma grading and IDH mutation status prediction at 3 T[J/OL]. Eur J Radiol, 2021, 134: 109466 [2025-09-20]. https://pubmed.ncbi.nlm.nih.gov/33307459/. DOI: 10.1016/j.ejrad.2020.109466.
[30]
SOHN B, PARK K, AHN S S, et al. Dynamic contrast-enhanced MRI radiomics model predicts epidermal growth factor receptor amplification in glioblastoma, IDH-wildtype[J]. J Neurooncol, 2023, 164(2): 341-351. DOI: 10.1007/s11060-023-04435-y.
[31]
GAO E Y, WANG P P, BAI J, et al. Radiomics analysis of diffusion kurtosis imaging: distinguishing between glioblastoma and single brain metastasis[J]. Acad Radiol, 2024, 31(3): 1036-1043. DOI: 10.1016/j.acra.2023.07.023.
[32]
MARAQAH H H, ABU-ASAB M S, LEE H S, et al. Astrocytoma and IDH1-wildtype glioblastoma (GBM) cells colonize tumor vessels and deploy vascular mimicry[J/OL]. Res Sq, 2023: rs.3.rs-2456733 [2025-09-20]. https://pubmed.ncbi.nlm.nih.gov/36712042/. DOI: 10.21203/rs.3.rs-2456733/v1.
[33]
MOHAMMADI S, GHADERI S, JOUZDANI A F, et al. Differentiation between high-grade glioma and brain metastasis using cerebral perfusion-related parameters (cerebral blood volume and cerebral blood flow): a systematic review and meta-analysis of perfusion-weighted MRI techniques[J]. J Magn Reson Imaging, 2025, 61(2): 758-768. DOI: 10.1002/jmri.29473.
[34]
HAO Z Y, GAO Y, WU Q, et al. Application value of MAP-MRI and DCE-MRI in differentiating glioblastoma from brain metastases[J]. Chin J Magn Reson Imaging, 2023, 14(2): 12-20. DOI: 10.12015/issn.1674-8034.2023.02.003.
[35]
GUO F Y, SUN S W, DENG X Q, et al. Predicting axillary lymph node metastasis in breast cancer using a multimodal radiomics and deep learning model[J/OL]. Front Immunol, 2024, 15: 1482020 [2025-09-20]. https://pubmed.ncbi.nlm.nih.gov/39735531/. DOI: 10.3389/fimmu.2024.1482020.
[36]
XU X D, LIANG F R, WEI R L, et al. Differentiation of high-grade gliomas from solitary brain metastases based on multi-parameter MRI feature fusion model[J]. Chin J Magn Reson Imaging, 2022, 13(11): 53-59, 65. DOI: 10.12015/issn.1674-8034.2022.11.010.

PREV Application of MUSE-DWI combined with amide proton transfer quantitative imaging in evaluating the consistency of meningiomas
NEXT Study on the application of a combined model based on quantitative magnetic resonance parameters and serological indicators for evaluating the activity of Graves<sup><sup>,</sup></sup>s ophthalmopathy
  


LINK


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