Share：
Share this content in WeChat
X
[Chinese][PDF]8633
• Clinical Articles •
Prediction of IDH mutations in glioma based on MRI multiparametric image fusion and DenseNet network
HU Zhenyuan  WEI Wei  HU Wenzhong  MA Menghang  LI Yan  WU Xusha  YIN Hong  XI Yibin 

Cite this article as: HU Z Y, WEI W, HU W Z, et al. Prediction of IDH mutations in glioma based on MRI multiparametric image fusion and DenseNet network[J]. Chin J Magn Reson Imaging, 2023, 14(7): 10-17. DOI:10.12015/issn.1674-8034.2023.07.003.


[Abstract] Objective Developing a high-accuracy prediction model based on artificial intelligence deep learning DenseNet network and multimodal fusion technology to predict the preoperative isocitrate dehydrogenase (IDH) gene mutation status in glioma patients.Materials and Methods Retrospective analysis of the preoperative multisequence MRI scan images of 256 (155 IDH wild type and 101 IDH mutant type) patients consecutively admitted to xijing hospital, air force military medical university, from January 2012 to September 2016, and the region of interest was outlined on T1-weighted imaging(T1WI), T2-weighted imaging (T2WI), and contrast-enhanced T1WI sequences; deep learning convolutional neural networks were used to extract and fuse the MRI multimodal features. The model performance differences between the multimodal fusion model and two simple stitching methods of multimodal features were quantitatively compared.Results The multimodal fusion had superior prediction performance than other single-modal simple splicing, achieving good discriminative performance with the training and testing set receiver operating characteristic curve area under the curve of 0.903 [95% confidence interval (CI), 0.845-0.961] and 0.904 (95% CI, 0.842-0.966), respectively; accuracy of 91.3% and 88.7%, respectively. The sensitivity reached 86.4% and 90.5% respectively; the specificity reached 94.5% and 87.5% respectively, and the model consistency was verified using the calibration curve, and the model calibration graph is close to the diagonal line, reflecting that the model has a good prediction effect. The DeLong test results showed a statistical difference (P＜0.05) in the model performance between the two methods of multimodal fusion and ablation, with the former being superior to the latter.Conclusions MRI multimodal fusion model based on deep learning DenseNet network can achieve non-invasive and low-cost prediction of IDH gene status of glioma before surgery by integrating multimodal MRI image information of tumor.
[Keywords] glioma；deep learning；intelligent medicine；magnetic resonance imaging；multimodal fusion；IDH

HU Zhenyuan1   WEI Wei1   HU Wenzhong2   MA Menghang1   LI Yan3   WU Xusha3   YIN Hong2, 3   XI Yibin3*  

1 School of Electronic Information, Xi'an Polytechnic University, Xi'an 710600, China

2 Department of Radiology, Xijing Hospital of the Fourth Military Medical University, Xi'an 710032, China

3 Medical Imaging Center, Xi'an People's Hospital (Xi'an Fourth Hospital), Xi'an 710004, China

Corresponding author: Xi YB, E-mail: xyb1113@qq.com

Conflicts of interest   None.

ACKNOWLEDGMENTS Shanxi Provincial Natural Science Basic Research Program (No. 2023-JC-YB-682, 2023-JC-ZD-58); Xi'an Science and Technology Plan for Scientific and Technical Staff of Universities and Institutes to Serve Enterprises (No. 22GXFW0036).
Received  2022-12-30
Accepted  2023-06-26
DOI: 10.12015/issn.1674-8034.2023.07.003
Cite this article as: HU Z Y, WEI W, HU W Z, et al. Prediction of IDH mutations in glioma based on MRI multiparametric image fusion and DenseNet network[J]. Chin J Magn Reson Imaging, 2023, 14(7): 10-17. DOI:10.12015/issn.1674-8034.2023.07.003.

[1]
ZHANG D Z, JIANG H P, YE J Y, et al. A novel lncRNA, RPL34-AS1, promotes proliferation and angiogenesis in glioma by regulating VEGFA[J]. J Cancer, 2021, 12(20): 6189-6197. DOI: 10.7150/jca.59337.
[2]
WANG H X, XU T, HUANG Q L, et al. Immunotherapy for malignant glioma: current status and future directions[J]. Trends Pharmacol Sci, 2020, 41(2): 123-138. DOI: 10.1016/j.tips.2019.12.003.
[3]
OSTROM Q T, GITTLEMAN H, LIAO P, et al. CBTRUS Statistical Report: primary brain and other central nervous system tumors diagnosed in the United States in 2010-2014[J/OL]. Neuro Oncol, 2017, 19(suppl_5): v1-v88 [2022-12-29]. https://pubmed.ncbi.nlm.nih.gov/29117289/. DOI: 10.1093/neuonc/nox158.
[4]
RENI M, MAZZA E, ZANON S, et al. Central nervous system gliomas[J/OL]. Crit Rev Oncol Hematol, 2017, 113: 213-234 [2022-12-29]. https://pubmed.ncbi.nlm.nih.gov/28427510/. DOI: 10.1016/j.critrevonc.2017.03.021.
[5]
YASUKAWA K, SHIMIZU A, MOTOYAMA H, et al. Impact of remnant carcinoma in situ at the ductal stump on long-term outcomes in patients with distal cholangiocarcinoma[J]. World J Surg, 2021, 45(1): 291-301. DOI: 10.1007/s00268-020-05799-2.
[6]
XIE Y D, HAN Y H, ZHANG X F, et al. Application of new radiosensitizer based on nano-biotechnology in the treatment of glioma[J/OL]. Front Oncol, 2021, 11: 633827 [2022-12-29]. https://pubmed.ncbi.nlm.nih.gov/33869019/. DOI: 10.3389/fonc.2021.633827.
[7]
KUROKAWA R, KUROKAWA M, BABA A, et al. Major changes in 2021 World Health Organization classification of central nervous system tumors[J]. Radiographics, 2022, 42(5): 1474-1493. DOI: 10.1148/rg.210236.
[8]
ŚLEDZIŃSKA P, BEBYN M G, FURTAK J, et al. Prognostic and predictive biomarkers in gliomas[J/OL]. Int J Mol Sci, 2021, 22(19): 10373 [2022-12-29]. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8508830/. DOI: 10.3390/ijms221910373.
[9]
GRITSCH S, BATCHELOR T T, GONZALEZ CASTRO L N. Diagnostic, therapeutic, and prognostic implications of the 2021 World Health Organization classification of tumors of the central nervous system[J]. Cancer, 2022, 128(1): 47-58. DOI: 10.1002/cncr.33918.
[10]
ROSA M da C, YAMASHITA A S, RIGGINS G J. Evaluation of a DNA demethylating agent in combination with all-trans retinoic acid for IDH1-mutant gliomas[J]. Neuro-Oncology, 2022, 24(5): 711-723. DOI: 10.1093/neuonc/noab263.
[11]
TESILEANU C M S, VALLENTGOED W R, SANSON M, et al. Non-IDH1-R132H IDH1/2 mutations are associated with increased DNA methylation and improved survival in astrocytomas, compared to IDH1-R132H mutations[J]. Acta Neuropathol, 2021, 141(6): 945-957. DOI: 10.1007/s00401-021-02291-6.
[12]
LEI H, LIANG D, ZHUO Y, et al. Magnetic resonance imaging: progresses and perspective[J]. Sci Sin-Vitae, 2020, 50(11): 1285-1295. DOI: 10.1360/ssv-2020-0164.
[13]
FANG L L, WANG X, LIAN Z Y, et al. Supervoxel-based brain tumor segmentation with multimodal MRI images[J]. SIViP, 2022, 16(5): 1215-1223. DOI: 10.1007/s11760-021-02072-4.
[14]
BOLOGNA M, CORINO V, CALARESO G, et al. Baseline MRI-radiomics can predict overall survival in non-endemic EBV-related nasopharyngeal carcinoma patients[J/OL]. Cancers (Basel), 2020, 12(10): 2958 [2022-12-29]. https://pubmed.ncbi.nlm.nih.gov/33066161/. DOI: 10.3390/cancers12102958.
[15]
MES S W, VAN VELDEN F H P, PELTENBURG B, et al. Outcome prediction of head and neck squamous cell carcinoma by MRI radiomic signatures[J]. Eur Radiol, 2020, 30(11): 6311-6321. DOI: 10.1007/s00330-020-06962-y.
[16]
AGNES S A, ANITHA J, PANDIAN S I A, et al. Classification of mammogram images using multiscale all convolutional neural network (MA-CNN)[J/OL]. J Med Syst, 2019, 44(1): 30 [2022-12-29]. https://pubmed.ncbi.nlm.nih.gov/31838610/. DOI: 10.1007/s10916-019-1494-z.
[17]
WEI S F, WU W, JEON G, et al. Improving resolution of medical images with deep dense convolutional neural network[J/OL]. Concurrency Computat Pract Exper, 2020, 32(1) [2022-12-29]. https://onlinelibrary.wiley.com/doi/10.1002/cpe.5084. DOI: 10.1002/cpe.5084.
[18]
WU Y, MA Y J, LIU J, et al. Self-attention convolutional neural network for improved MR image reconstruction[J/OL]. Inf Sci (N Y), 2019, 490: 317-328 [2022-12-29]. https://pubmed.ncbi.nlm.nih.gov/32817993/. DOI: 10.1016/j.ins.2019.03.080.
[19]
GAO X, CHEN T, NIU R Q, et al. Recognition and mapping of landslide using a fully convolutional DenseNet and influencing factors[J/OL]. IEEE J Sel Top Appl Earth Obs Remote Sens, 2021, 14: 7881-7894 [2022-12-29]. https://ieeexplore.ieee.org/document/9502947. DOI: 10.1109/JSTARS.2021.3101203.
[20]
FANG L L, WANG X. Brain tumor segmentation based on the dual-path network of multi-modal MRI images[J/OL]. Pattern Recognit, 2022, 124: 108434 [2022-12-29]. https://www.sciencedirect.com/science/article/abs/pii/S0031320321006105?via%3Dihub. DOI: 10.1016/j.patcog.2021.108434.
[21]
ZHANG W F, YU J, ZHAO W H, et al. DMRFNet: Deep Multimodal Reasoning and Fusion for Visual Question Answering and explanation generation[J]. Inf Fusion, 2021, 72: 70-79. DOI: 10.1016/j.inffus.2021.02.006.
[22]
CHEN R J, LU M Y, WANG J W, et al. Pathomic fusion: an integrated framework for fusing histopathology and genomic features for cancer diagnosis and prognosis[J]. IEEE Trans Med Imaging, 2022, 41(4): 757-770. DOI: 10.1109/TMI.2020.3021387.
[23]
SHAKER, EL-SAPPAGH. Multimodal multitask deep learning model for Alzheimer's disease progression detection based on time series data[J/OL]. Neurocomputing, 2020, 412: 197-215 [2022-12-29]. https://www.sciencedirect.com/science/article/abs/pii/S0925231220309383?via%3Dihub. DOI: 10.1016/j.neucom.2020.05.087.
[24]
HUANG H X, ZHANG J J, ZHANG J, et al. Low-rank pairwise alignment bilinear network for few-shot fine-grained image classification[J/OL]. IEEE Trans Multimed, 2021, 23: 1666-1680 [2022-12-29]. https://ieeexplore.ieee.org/document/9115215. DOI: 10.1109/TMM.2020.3001510.
[25]
HUANG Y Q, LIANG C H, HE L, et al. Development and validation of a radiomics nomogram for preoperative prediction of lymph node metastasis in colorectal cancer[J]. J Clin Oncol, 2016, 34(18): 2157-2164. DOI: 10.1200/JCO.2015.65.9128.
[26]
ZHANG X H, YUAN K, WANG H M, et al. Nomogram to predict mortality of endovascular thrombectomy for ischemic stroke despite successful recanalization[J/OL]. J Am Heart Assoc, 2020, 9(3): e014899 [2022-12-29]. https://pubmed.ncbi.nlm.nih.gov/31973604/. DOI: 10.1161/JAHA.119.014899.
[27]
FRIEDRICH M, SANKOWSKI R, BUNSE L, et al. Tryptophan metabolism drives dynamic immunosuppressive myeloid states in IDH-mutant gliomas[J]. Nat Cancer, 2021, 2(7): 723-740. DOI: 10.1038/s43018-021-00201-z.
[28]
GUO S C, WANG L H, CHEN Q J, et al. Multimodal MRI image decision fusion-based network for glioma classification[J/OL]. Front Oncol, 2022, 12: 819673 [2022-12-29]. https://pubmed.ncbi.nlm.nih.gov/35280828/. DOI: 10.3389/fonc.2022.819673.
[29]
HE M, HAN K F, ZHANG Y, et al. Hierarchical-order multimodal interaction fusion network for grading gliomas[J/OL]. Phys Med Biol, 2021, 66(21): 215016 [2022-12-29]. https://iopscience.iop.org/article/10.1088/1361-6560/ac30a1. DOI: 10.1088/1361-6560/ac30a1.
[30]
ZENG H L, XING Z, GAO F L, et al. A multimodal domain adaptive segmentation framework for IDH genotype prediction[J].Int J Comput Assist Radiol Surg, 2022, 17(10): 1923-1931. DOI: 10.1007/s11548-022-02700-5.
[31]
PASQUINI L, NAPOLITANO A, TAGLIENTE E, et al. Deep learning can differentiate IDH-mutant from IDH-wild GBM[J/OL]. J Pers Med, 2021, 11(4): 290 [2022-12-29]. https://pubmed.ncbi.nlm.nih.gov/33918828/. DOI: 10.3390/jpm11040290.
[32]
MUHAMMAD G, ALHUSSEIN M. Convergence of artificial intelligence and Internet of Things in smart healthcare: a case study of voice pathology detection[J/OL]. IEEE Access, 2021, 9: 89198-89209 [2022-12-29]. https://ieeexplore.ieee.org/document/9458291. DOI: 10.1109/ACCESS.2021.3090317.
[33]
MUKHERJEE S. Emerging frontiers in smart environment and healthcare-A vision[J].Inf Syst Front, 2020, 22(1): 23-27. DOI: 10.1007/s10796-019-09965-3.
[34]
DONG S, WANG P, ABBAS K. A survey on deep learning and its applications[J/OL]. Comput Sci Rev, 2021, 40: 100379 [2022-12-29]. https://www.sciencedirect.com/science/article/abs/pii/S1574013721000198?via%3Dihub. DOI: 10.1016/J.COSREV.2021.100379.
[35]
PARK H, PARK B, LEE S S. Radiomics and deep learning: hepatic applications[J/OL]. Korean J Radiol, 2020, 21: 387-401 [2022-12-29]. https://www.kjronline.org/DOIx.php?id=10.3348/kjr.2019.0752. DOI: 10.3348/kjr.2019.0752.
[36]
HUANG W W, ZHANG H, QUAN X W, et al. A two-level dynamic adaptive network for medical image fusion[J/OL]. IEEE Trans Instrum Meas, 2022, 71: 1-17 [2022-12-29]. https://ieeexplore.ieee.org/document/9762233. DOI: 10.1109/TIM.2022.3169546.
[37]
VALADA A, MOHAN R, BURGARD W. Self-supervised model adaptation for multimodal semantic segmentation[J]. Int J Comput Vis, 2020, 128(5): 1239-1285. DOI: 10.1007/s11263-019-01188-y.
[38]
LIU B S, LV Y B, GU Y, et al. Implementation of a lightweight semantic segmentation algorithm in road obstacle detection[J/OL]. Sensors (Basel), 2020, 20(24): 7089 [2022-12-29]. https://pubmed.ncbi.nlm.nih.gov/33322029/. DOI: 10.3390/s20247089.
[39]
HUANG G, LIU Z, VAN DER MAATEN L, et al. Densely connected convolutional networks[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). July 21-26, 2017, Honolulu, HI, USA. IEEE, 2017: 2261-2269. DOI: 10.1109/CVPR.2017.243.

PREV Regulation of amygdala by rtfMRI-NF technique in improving insomnia disorder
NEXT Application of Sy-MRI combined with DWI in predicting MGMT methylation in glioma
  


LINK


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