Share：
Share this content in WeChat
X
[Chinese][PDF]3703
• Technical Article •
MRS combined with subtraction technique in the prediction of high grade glioma radiomics grading
SONG Jing  ZONG Huiqian  ZHANG Ya  LIU Qing  WEI Haoye  YANG Cun  XIE Lizhi 

Cite this article as: SONG J, ZONG H Q, ZHANG Y, et al. MRS combined with subtraction technique in the prediction of high grade glioma radiomics grading[J]. Chin J Magn Reson Imaging, 2023, 14(6): 59-65. DOI:10.12015/issn.1674-8034.2023.06.009.


[Abstract] Objective Magnetic resonance spectroscopy (MRS) and subtraction techniques were introduced in the conventional imaging omics analysis to identify high-grade glioma grade.Materials and Methods The imaging data of 60 patients with pathologically confirmed high-grade glioma (25 patients with grade Ⅲ and 35 patients with grade Ⅳ) were retrospectively collected, and radiomic features such as shape and texture features were extracted based on contrast-enhanced T1-weighted imaging (CE-T1WI) images and subtracted images, and feature screening was performed using t-test and least absolute shrinkage and selection operator cross-validation, combined with the peak ratios of three metabolites of MRS, and a random forest algorithm was used to construct a high-grade glioma grading discrimination model and evaluate the model performance.Results The area under the curve (AUC) of the model constructed based on CE-T1WI images was 0.78 in the test set; the AUC of the model constructed based on subtracted images was 0.81 in the test set; the AUC of the model constructed based on metabolite peak ratios from MRS was 0.80 in the test set; and the AUC of the model constructed based on CE-T1WI images and three metabolite peak ratios from MRS was 0.95 in the test set.Conclusions Radiomics based on CE-T1WI images, subtraction images and MRS have good performance in identifying both grade Ⅲ and grade Ⅳ gliomas, with the subtraction model performing best in the single sequence model and the CE-T1WI combined with MRS model performing best in the combined sequence model. Multimodal radiomic analysis can provide a useful clinical aid for identifying grade Ⅲ and Ⅳ gliomas.
[Keywords] high-grade glioma；magnetic resonance spectroscopy；magnetic resonance imaging；radiomics；subtraction

SONG Jing1   ZONG Huiqian2*   ZHANG Ya2   LIU Qing2   WEI Haoye2   YANG Cun2   XIE Lizhi3  

1 Department of Medical Imaging, the Second Hospital of Hebei Medical University, Shijiazhuang 050000, China

2 Department of Medical Equipment, the Second Hospital of Hebei Medical University, Shijiazhuang 050000, China

3 General Electric Medical Systems Trading Development (Shanghai) Co., Shanghai 201203, China

Corresponding author: Zong HQ, E-mail: zonghuiqian@sina.com

Conflicts of interest   None.

ACKNOWLEDGMENTS Research Fund of Hebei Provincial Health and Health Commission (No. 20230518).
Received  2023-02-09
Accepted  2023-05-05
DOI: 10.12015/issn.1674-8034.2023.06.009
Cite this article as: SONG J, ZONG H Q, ZHANG Y, et al. MRS combined with subtraction technique in the prediction of high grade glioma radiomics grading[J]. Chin J Magn Reson Imaging, 2023, 14(6): 59-65. DOI:10.12015/issn.1674-8034.2023.06.009.

[1]
ZHAO Y R, WANG L M, CHEN Y D, et al. Pathological features and recurrence patterns of multiple high-grade gliomas[J]. Journal of Modern Oncology, 2023, 31(5): 843-847. DOI: 10.3969/j.issn.1672-4992.2023.05.011.
[2]
DREYFUSS J M, JOHNSON M D, PARK P J. Meta-analysis of glioblastoma multiforme versus anaplastic astrocytoma identifies robust gene markers[J/OL]. Mol Cancer, 2009, 8: 71 [2022-11-01]. https://pubmed.ncbi.nlm.nih.gov/19732454/. DOI: 10.1186/1476-4598-8-71.
[3]
ZIDAN S, TANTAWY H I, MAKIA M A. High grade gliomas: The role of dynamic contrast-enhanced susceptibility-weighted perfusion MRI and proton MR spectroscopic imaging in differentiating grade Ⅲ from grade Ⅳ[J]. Egypt J Radiol Nucl Med, 2016, 47(4): 1565-1573. DOI: 10.1016/j.ejrnm.2016.10.002.
[4]
SINGH G, MANJILA S, SAKLA N, et al. Radiomics and radiogenomics in gliomas: a contemporary update[J]. Br J Cancer, 2021, 125(5): 641-657. DOI: 10.1038/s41416-021-01387-w.
[5]
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.
[6]
ZHUO Z, QU L, ZHANG P, et al. Prediction of H3K27M-mutant brainstem glioma by amide proton transfer-weighted imaging and its derived radiomics[J]. Eur J Nucl Med Mol Imaging, 2021, 48(13): 4426-4436. DOI: 10.1007/s00259-021-05455-4.
[7]
LIU Z, JIANG Z, MENG L, et al. Handcrafted and Deep Learning-Based Radiomic Models Can Distinguish GBM from Brain Metastasis[J/OL]. J Oncol, 2021, 2021: 5518717 [2022-11-01]. https://pubmed.ncbi.nlm.nih.gov/34188680/. DOI: 10.1155/2021/5518717.
[8]
TAKAHASHI S, TAKAHASHI W, TANAKA S, et al. Radiomics Analysis for Glioma Malignancy Evaluation Using Diffusion Kurtosis and Tensor Imaging[J]. Int J Radiat Oncol Biol Phys, 2019, 105(4): 784-791. DOI: 10.1016/j.ijrobp.2019.07.011.
[9]
TIAN Q, YAN L F, ZHANG X, et al. Radiomics strategy for glioma grading using texture features from multiparametric MRI[J]. J Magn Reson Imaging, 2018, 48(6): 1518-1528. DOI: 10.1002/jmri.26010.
[10]
NAKAMOTO T, TAKAHASHI W, HAGA A, et al. Prediction of malignant glioma grades using contrast-enhanced T1-weighted and T2-weighted magnetic resonance images based on a radiomic analysis[J/OL]. Sci Rep, 2019, 19, 9(1): 19411 [2022-11-01]. https://pubmed.ncbi.nlm.nih.gov/31857632/. DOI: 10.1038/s41598-019-55922-0.
[11]
KOCHER M, RUGE M I, GALLDIKS N, et al. Applications of radiomics and machine learning for radiotherapy of malignant brain tumors[J]. Strahlenther Onkol, 2020, 196(10): 856-867. DOI: 10.1007/s00066-020-01626-8.
[12]
CARRETE L R, YOUNG J S, CHA S. Advanced Imaging Techniques for Newly Diagnosed and Recurrent Gliomas[J/OL]. Front Neurosci, 2022, 16: 787755 [2022-11-01]. https://pubmed.ncbi.nlm.nih.gov/35281485/. DOI: 10.3389/fnins.2022.787755.
[13]
VAMVAKAS A, WILLIAMS S C, THEODOROU K, et al. Imaging biomarker analysis of advanced multiparametric MRI for glioma grading[J/OL]. Phys Med, 2019, 60: 188-198 [2022-11-02]. https://pubmed.ncbi.nlm.nih.gov/30910431/. DOI: 10.1016/j.ejmp.2019.03.014.
[14]
ELLINGSON B M, BENDSZUS M, SORENSEN A G, et al. Emerging techniques and technologies in brain tumor imaging[J]. Neuro Oncol, 2014, 16(Suppl 7): 12-23. DOI: 10.1093/neuonc/nou221.
[15]
GORYAWALA M, ROY B, GUPTA R K, et al. T1-weighted and T2-weighted Subtraction MR Images for Glioma Visualization and Grading[J]. J Neuroimaging, 2021, 31(1): 124-131. DOI: 10.1111/jon.12800.
[16]
PRASANNA P, PATEL J, PARTOVI S, et al. Radiomic features from the peritumoral brain parenchyma on treatment-naïve multi-parametric MR imaging predict long versus short-term survival in glioblastoma multiforme: Preliminary findings[J]. Eur Radiol, 2017, 27(10): 4188-4197. DOI: 10.1007/s00330-016-4637-3.
[17]
KIM S M, KIM Y, JEONG K, et al. Logistic LASSO regression for the diagnosis of breast cancer using clinical demographic data and the BI-RADS lexicon for ultrasonography[J]. Ultrasonography, 2018, 37(1): 36-42. DOI: 10.14366/usg.16045.
[18]
VASQUEZ M M, HU C, ROE D J, et al. Least absolute shrinkage and selection operator type methods for the identification of serum biomarkers of overweight and obesity: simulation and application[J/OL]. BMC Med Res Methodol, 2016, 16: 154 [2022-11-01]. https://pubmed.ncbi.nlm.nih.gov/27842498/. DOI: 10.1186/s12874-016-0254-8.
[19]
OSTROM Q T, CIOFFI G, WAITE K, et al. CBTRUS Statistical Report: Primary Brain and Other Central Nervous System Tumors Diagnosed in the United States in 2014-2018[J]. Neuro Oncol, 2021, 23(12Suppl 2): Ⅲ1-Ⅲ105. DOI: 10.1093/neuonc/noab200.
[20]
AVANZO M, WEI L, STANCANELLO J, et al. Machine and deep learning methods for radiomics[J/OL]. Med Phys, 2020, 47(5): e185-e202 [2022-11-02]. https://pubmed.ncbi.nlm.nih.gov/32418336/. DOI: 10.1002/mp.13678.
[21]
HASHIDO T, SAITO S, ISHIDA T. Radiomics-Based Machine Learning Classification for Glioma Grading Using Diffusion- and Perfusion-Weighted Magnetic Resonance Imaging[J]. J Comput Assist Tomogr, 2021, 45(4): 606-613. DOI: 10.1097/RCT.0000000000001180.
[22]
PARMAR C, GROSSMANN P, BUSSINK J, et al. Machine learning methods for quantitative radiomic biomarkers[J/OL]. Sci Rep, 2015, 5: 13087 [2022-11-01]. https://pubmed.ncbi.nlm.nih.gov/26278466/. DOI: 10.1038/srep13087.
[23]
SENGUPTA A, RAMANIHARAN A K, GUPTA R K, et al. Glioma grading using a machine-learning framework based on optimized features obtained from T1 perfusion MRI and volumes of tumor components[J]. J Magn Reson Imaging, 2019, 50(4): 1295-1306. DOI: 10.1002/jmri.26704.
[24]
GAO M, HUANG S, PAN X, et al. Machine Learning-Based Radiomics Predicting Tumor Grades and Expression of Multiple Pathologic Biomarkers in Gliomas[J/OL]. Front Oncol, 2020, 10: 1676 [2022-11-02]. https://pubmed.ncbi.nlm.nih.gov/33014836/. DOI: 10.3389/fonc.2020.01676.
[25]
HOSMER D W, LEMESHOW S. Applied Logistic Regression[M]. 2nd ed. New York: John Willey & Sons Inc, 2000, 91-142.
[26]
KHOULI R H EL, MACURA K J, BARKER P B, et al. Relationship of temporal resolution to diagnostic performance for dynamic contrast enhanced MRI of the breast[J]. J Magn Reson Imaging, 2009, 30(5): 999-1004. DOI: 10.1002/jmri.21947.
[27]
SU C, JIANG J, ZHANG S, et al. Radiomics based on multicontrast MRI can precisely differentiate among glioma subtypes and predict tumour-proliferative behaviour[J]. Eur Radiol, 2019, 29(4): 1986-1996. DOI: 10.1007/s00330-018-5704-8.
[28]
ZHANG Z, XIAO J, WU S, et al. Deep Convolutional Radiomic Features on Diffusion Tensor Images for Classification of Glioma Grades[J]. J Digit Imaging, 2020, 33(4): 826-837. DOI: 10.1007/s10278-020-00322-4.
[29]
DING J, ZHAO R, QIU Q, et al. Developing and validating a deep learning and radiomic model for glioma grading using multiplanar reconstructed magnetic resonance contrast-enhanced T1-weighted imaging: a robust, multi-institutional study[J]. Quant Imaging Med Surg, 2022, 12(2): 1517-1528. DOI: 10.21037/qims-21-722.
[30]
ELLINGSON B M, KIM H J, WOODWORTH D C, et al. Recurrent glioblastoma treated with bevacizumab: contrast-enhanced T1-weighted subtraction maps improve tumor delineation and aid prediction of survival in a multicenter clinical trial[J]. Radiology, 2014, 271(1): 200-210. DOI: 10.1148/radiol.13131305.
[31]
HOLLI K, LÄÄPERI A L, HARRISON L, et al. Characterization of breast cancer types by texture analysis of magnetic resonance images[J]. Acad Radiol, 2010, 17(2): 135-141. DOI: 10.1016/j.acra.2009.08.012.
[32]
RAFIQUE Z, AWAN M W, IQBAL S, et al. Diagnostic accuracy of magnetic resonance spectroscopy in predicting the grade of glioma keeping histopathology as the gold standard[J/OL]. Cureus, 2022, 14(2): e22056 [2022-11-13]. https://pubmed.ncbi.nlm.nih.gov/35340513/. DOI: 10.7759/cureus.22056.
[33]
WANG Q, ZHANG H, ZHANG J, et al. The diagnostic performance of magnetic resonance spectroscopy in differentiating high-from low-grade gliomas: a systematic review and meta-analysis[J]. Eur Radiol, 2016, 26(8): 2670-2684. DOI: 10.1007/s00330-015-4046-z.
[34]
ZHOU K Y, FANG X J, PENG J C, et al. The value of 1H-MRS and 3D-ASL in grading diagnosis of glioma[J]. J Clin Radiol, 2022, 41(7): 1217-1221. DOI: 10.13437/j.cnki.jcr.2022.07.037.
[35]
LI X, XIE J C, WANG J, et al. The application value of magnetic resonance MRS, DWI and SWI sequences in the grading diagnosis of glioma[J]. Chinese Journal of General Practice, 2022, 20(9): 1541-1544. DOI: 10.16766/j.cnki.issn.1674-4152.002644.
[36]
LIN K, CIDAN W, QI Y, et al. Glioma grading prediction using multiparametric magnetic resonance imaging-based radiomics combined with proton magnetic resonance spectroscopy and diffusion tensor imaging[J]. Med Phys, 2022, 49(7): 4419-4429. DOI: 10.1002/mp.15648.

PREV MRI features of sacroiliac joint bone marrow fat-saturated T2WI high signal in osteitis condensans ilii
NEXT The visualization of trigeminal nerve and its adjacent vessels using ultra-high field 7 T MRI
  


LINK


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