Share：
Share this content in WeChat
X
[Chinese][PDF]2776
• Clinical Article •
Value of combining radiomics and deep-learning with hematological inflammatory markers in predicting the prognosis of glioma
ZHAO Shan  YAN Zikang  YANG Junjun  ZHANG Wentao  PAN Shijiao  XU Shengsheng 

Cite this article as: ZHAO S, YAN Z K, YANG J J, et al. Value of combining radiomics and deep-learning with hematological inflammatory markers in predicting the prognosis of glioma[J]. Chin J Magn Reson Imaging, 2024, 15(1): 88-94, 100. DOI:10.12015/issn.1674-8034.2024.01.014.


[Abstract] Objective To explore the value of a nomogram constructed by integrating radiomics and deep learning-based score (RD-score) with hematological inflammatory markers in preoperatively predicting the prognosis of glioma.Materials and Methods A total of 166 clinically diagnosed glioma patients were retrospectively enrolled and randomly divided into a training set (133 cases) and a validation set (33 cases) in an 8:2 ratio. Clinical and hematological inflammatory marker of the patients were collected. Composite variables, including systemic immune inflammation index (SII), systemic immune response index (SIRI), derived neutrophil-to-lymphocyte ratio (dNLR), neutrophil-to-lymphocyte ratio (NLR), monocyte-to-lymphocyte ratio (MLR), and platelet-to-lymphocyte ratio (PLR) were constructed, and their optimal cut-off values were calculated. Delineating the volume of interest (VOI) for gliomas and extracting radiomics and deep learning features, utilizing least absolute shrinkage and selection operator (LASSO)-Cox for feature selection. Constructing radiomics-score (Rad-score), deep learning-score (DL-score), and RD-score, and comparing their receiver operating characteristic area under the curve (AUC) to assess predictive performance. Kaplan-Meier survival analysis was used to stratify glioma patients based on their RD-score. Integrating clinical data, hematological inflammatory marker, and RD-score, employing multivariable Cox regression to build RD-score model, clinical hematology model, and joint model to predict overall survival (OS). Calculating AUC to evaluate the efficiency of each model in predicting glioma 1, 3, and 5-year survival rates. Drawing joint model nomogram and assessing their performance using C-index, calibration curves, and decision curve analysis (DCA).Results After feature selection, 10 radiomics features and 8 deep learning features were selected. The predictive performance of RD-score surpassed that of Rad-score and DL-score (DeLong test, P<0.05). The constructed RD-score divided gliomas into high-risk group (RD-score≥1.09) and low-risk group (RD-score<1.09). The results of the multivariable Cox regression showed that age, tumor grade, postoperative chemotherapy, SIRI, and RD-score were independent prognostic factors for glioma. The joint model, incorporating these factors, exhibited higher AUC in the training and validation sets compared to the RD-score model and clinical hematology model (DeLong test, P<0.05). The visual nomogram of the joint model predicted OS with C-indices of 0.844 and 0.849 in the training and validation sets, respectively. Calibration curves indicated good consistency between observed and predicted values, and DCA demonstrated a high net benefit for the nomogram.Conclusions The nomogram constructed by combining radiomics and deep learning-based RD-score with clinical- hematological inflammatory marker can effectively predict the prognosis of glioma patients preoperatively.
[Keywords] glioma；magnetic resonance imaging；radiomics；deep learning；machine learning；prognosis；hematological inflammatory markers

ZHAO Shan1   YAN Zikang2   YANG Junjun3   ZHANG Wentao1   PAN Shijiao1   XU Shengsheng1*  

1 Department of Radiology, the First Affiliated Hospital of Chongqing Medical University, Chongqing 400016, China

2 Department of Bioinformatics, the Basic Medical School of Chongqing Medical University, Chongqing 400016, China

3 Key Laboratory of Optoelectronic Technology and Systems of the Ministry of Education, Chongqing University, Chongqing 400044, China

Corresponding author: XU S S, E-mail: xuss@cqmu.edu.cn

Conflicts of interest   None.

Received  2023-08-01
Accepted  2023-12-05
DOI: 10.12015/issn.1674-8034.2024.01.014
Cite this article as: ZHAO S, YAN Z K, YANG J J, et al. Value of combining radiomics and deep-learning with hematological inflammatory markers in predicting the prognosis of glioma[J]. Chin J Magn Reson Imaging, 2024, 15(1): 88-94, 100. DOI:10.12015/issn.1674-8034.2024.01.014.

[1]
OSTROM Q T, BAUCHET L, DAVIS F G, et al. The epidemiology of glioma in adults: a "state of the science" review[J]. Neuro Oncol, 2014, 16(7): 896-913. DOI: 10.1093/neuonc/nou087.
[2]
CHEN R, ZHANG H, WU W, et al. Antigen presentation machinery signature-derived CALR mediates migration, polarization of macrophages in glioma and predicts immunotherapy response[J/OL]. Front Immunol, 2022, 13: 833792 [2023-08-01]. https://pubmed.ncbi.nlm.nih.gov/35418980/. DOI: 10.3389/fimmu.2022.833792.
[3]
WANG S K, HAN X J, ZHU J Y, et al. Value of diffusion-weighted imaging combined with diffusion kurtosis imaging in the hierarchical diagnosis and prognosis assessment of glioma[J]. Chin J Magn Reson Imaging, 2022, 13(9): 86-90, 99. DOI: 10.12015/issn.1674-8034.2022.09.016.
[4]
ILHAN B, LIN K, GUNERI P, et al. Improving oral cancer outcomes with imaging and artificial intelligence[J]. J Dent Res, 2020, 99(3): 241-248. DOI: 10.1177/0022034520902128.
[5]
ITO Y, NAKAJIMA T, INAGE T, et al. Prediction of nodal metastasis in lung cancer using deep learning of endobronchial ultrasound images[J/OL]. Cancers (Basel), 2022, 14(14): 3334 [2023-08-01]. https://pubmed.ncbi.nlm.nih.gov/35884395/. DOI: 10.3390/cancers14143334.
[6]
XI I L, ZHAO Y, WANG R, et al. Deep learning to distinguish benign from malignant renal lesions based on routine MR imaging[J]. Clin Cancer Res, 2020, 26(8): 1944-1952. DOI: 10.1158/1078-0432.Ccr-19-0374.
[7]
AURNA N F, YOUSUF M A, TAHER K A, et al. A classification of MRI brain tumor based on two stage feature level ensemble of deep CNN models[J/OL]. Comput Biol Med, 2022, 146: 105539 [2023-08-01]. https://pubmed.ncbi.nlm.nih.gov/35483227/. DOI: 10.1016/j.compbiomed.2022.105539.
[8]
CHEN H, LI C, ZHENG L, et al. A machine learning-based survival prediction model of high grade glioma by integration of clinical and dose-volume histogram parameters[J]. Cancer Med, 2021, 10(8): 2774-2786. DOI: 10.1002/cam4.3838.
[9]
CHEN X, KANG R, KROEMER G, et al. Ferroptosis in infection, inflammation, and immunity[J/OL]. J Exp Med, 2021, 218(6): e20210518 [2023-08-01]. https://pubmed.ncbi.nlm.nih.gov/33978684/. DOI: 10.1084/jem.20210518.
[10]
LI J, SHU X, XU J, et al. S100A9-CXCL12 activation in BRCA1-mutant breast cancer promotes an immunosuppressive microenvironment associated with resistance to immunotherapy[J/OL]. Nat Commun, 2022, 13(1): 1481 [2023-08-01]. https://pubmed.ncbi.nlm.nih.gov/35304461/. DOI: 10.1038/s41467-022-29151-5.
[11]
CLAVREUL A, LEMÉE J M, SOULARD G, et al. A simple preoperative blood count to stratify prognosis in isocitrate dehydrogenase-wildtype glioblastoma patients treated with radiotherapy plus concomitant and adjuvant temozolomide[J/OL]. Cancers (Basel)2021, 13(22): 5778 [2023-08-01]. https://pubmed.ncbi.nlm.nih.gov/34830935/. DOI: 10.3390/cancers13225778.
[12]
LI S S, LIU X J, ZHANG D Y, et al. Correlation between NLR, PLR, fibrinogen, PDW and clinical pathological grade in glioma patients[J]. Journal of Regional Anatomy and Operative Surgery, 2019, 28(5): 383-386. DOI: 10.11659/jjssx.02E019019.
[13]
GARRETT C, BECKER T M, LYNCH D, et al. Comparison of neutrophil to lymphocyte ratio and prognostic nutritional index with other clinical and molecular biomarkers for prediction of glioblastoma multiforme outcome[J/OL]. PLoS One, 2021, 16(6): e0252614 [2023-08-01]. https://pubmed.ncbi.nlm.nih.gov/34138894/. DOI: 10.1371/journal.pone.0252614.
[14]
LIU X, GE H, FENG X, et al. The combination of hemogram indexes to predict exacerbation in stable chronic obstructive pulmonary disease[J/OL]. Front Med (Lausanne), 2020, 7: 572435 [2023-08-01]. https://pubmed.ncbi.nlm.nih.gov/33381510/. DOI: 10.3389/fmed.2020.572435.
[15]
TANG Y, ZENG X, FENG Y, et al. Association of systemic immune-inflammation index with short-term mortality of congestive heart failure: A retrospective cohort study[J/OL]. Front Cardiovasc Med, 2021, 8: 753133 [2023-08-01]. https://pubmed.ncbi.nlm.nih.gov/34869661/. DOI: 10.3389/fcvm.2021.753133.
[16]
ZHANG S, LI Z, ZHOU H, et al. Advancing 3D medical image analysis with variable dimension transform based supervised 3D pre-training[J]. Neurocomputing, 2023, 529: 11-22. DOI: 10.1016/j.neucom.2023.01.012.
[17]
KIM H J, PARK J H, KIM H C, et al. Blood monocyte-derived CD169+ macrophages contribute to antitumor immunity against glioblastoma[J/OL]. Nat Commun, 2022, 13(1): 6211 [2023-08-01]. https://pubmed.ncbi.nlm.nih.gov/36266311/. DOI: 10.1038/s41467-022-34001-5.
[18]
WU W, KLOCKOW J L, ZHANG M, et al. Glioblastoma multiforme (GBM): An overview of current therapies and mechanisms of resistance[J/OL]. Pharmacol Res, 2021, 171: 105780 [2023-08-01]. https://pubmed.ncbi.nlm.nih.gov/34302977/. DOI: 10.1016/j.phrs.2021.105780.
[19]
ZHU M L, CHEN L, KONG X Y, et al. Clinical significance and prognostic value of systemic inflammation response index in breast cancer patients[J]. J Mod Oncol, 2022, 30(3): 442-450. DOI: 10.3969/j.issn.1672-4992.2022.03.014.
[20]
QI Q, ZHUANG L, SHEN Y, et al. A novel systemic inflammation response index (SIRI) for predicting the survival of patients with pancreatic cancer after chemotherapy[J]. Cancer, 2016, 122(14): 2158-2167. DOI: 10.1002/cncr.30057.
[21]
HUA X, LONG Z Q, HUANG X, et al. The preoperative systemic inflammation response index (SIRI) independently predicts survival in postmenopausal women with breast cancer[J/OL]. Curr Probl Cancer, 2020, 44(4): 100560 [2023-08-01]. https://pubmed.ncbi.nlm.nih.gov/32122667/. DOI: 10.1016/j.currproblcancer.2020.100560.
[22]
TOPKAN E, KUCUK A, OZDEMIR Y, et al. Systemic inflammation response index predicts survival outcomes in glioblastoma multiforme patients treated with standard stupp protocol[J/OL]. J Immunol Res, 2020, 2020: 8628540 [2023-08-01]. https://pubmed.ncbi.nlm.nih.gov/33274245/. DOI: 10.1155/2020/8628540.
[23]
XIN Y, ZHANG X, LI Y, et al. A systemic inflammation response index (SIRI)-based nomogram for predicting the recurrence of early stage hepatocellular carcinoma after radiofrequency ablation[J]. Cardiovasc Intervent Radiol, 2022, 45(1): 43-53. DOI: 10.1007/s00270-021-02965-4.
[24]
MOU L S, ZOU W F, ZHANG L, et al. Value of preoperative hematologic inflammatory markers in the diagnosis of different grades of glioma[J]. J Canc Control Treat, 2022, 35(11): 996-1000. DOI: 10.3969/j.issn.1674-0904.2022.11.003.
[25]
HE Q, LI L, REN Q. The prognostic value of preoperative systemic inflammatory response index (SIRI) in patients with high-grade glioma and the establishment of a nomogram[J/OL]. Front Oncol, 2021, 11: 671811 [2023-08-01]. https://pubmed.ncbi.nlm.nih.gov/34055639/. DOI: 10.3389/fonc.2021.671811.
[26]
WANG Z, LI J, YUAN Y, et al. Prognostic significance of preoperative systemic inflammation response index in newly diagnosed glioblastoma patients underwent gross total resection: a propensity score matching analysis[J/OL]. World J Surg Oncol, 2022, 20(1): 137 [2023-08-01]. https://pubmed.ncbi.nlm.nih.gov/35488347/. DOI: 10.1186/s12957-022-02588-0.
[27]
SUN G Y, WANG S L, SONG Y W, et al. Radiation-induced lymphopenia predicts poorer prognosis in patients with breast cancer: A post hoc analysis of a randomized controlled trial of postmastectomy hypofractionated radiation therapy[J]. Int J Radiat Oncol Biol Phys, 2020, 108(1): 277-285. DOI: 10.1016/j.ijrobp.2020.02.633.
[28]
CORTESE N, CARRIERO R, LAGHI L, et al. Prognostic significance of tumor-associated macrophages: past, present and future[J/OL]. Semin Immunol, 2020, 48: 101408 [2023-08-01]. https://pubmed.ncbi.nlm.nih.gov/32943279/. DOI: 10.1016/j.smim.2020.101408.
[29]
SCHIFFMANN L M, FRITSCH M, GEBAUER F, et al. Tumour-infiltrating neutrophils counteract anti-VEGF therapy in metastatic colorectal cancer[J]. Br J Cancer, 2019, 120(1): 69-78. DOI: 10.1038/s41416-018-0198-3.
[30]
LIAO H, YANG J, LI Y, et al. One 3D VOI-based deep learning radiomics strategy, clinical model and radiologists for predicting lymph node metastases in pancreatic ductal adenocarcinoma based on multiphasic contrast-enhanced computer tomography[J/OL]. Front Oncol, 2022, 12: 990156 [2023-08-01]. https://pubmed.ncbi.nlm.nih.gov/36158647/. DOI: 10.3389/fonc.2022.990156.
[31]
XU Y, HE X, LI Y, et al. The nomogram of MRI-based radiomics with complementary visual features by machine learning improves stratification of glioblastoma patients: A multicenter study[J]. J Magn Reson Imaging, 2021, 54(2): 571-583. DOI: 10.1002/jmri.27536.
[32]
CHOI Y, NAM Y, JANG J, et al. Radiomics may increase the prognostic value for survival in glioblastoma patients when combined with conventional clinical and genetic prognostic models[J]. Eur Radiol, 2021, 31(4): 2084-2093. DOI: 10.1007/s00330-020-07335-1.
[33]
HAN W, QIN L, BAY C, et al. Deep transfer learning and radiomics feature prediction of survival of patients with high-grade gliomas[J]. AJNR Am J Neuroradiol, 2020, 41(1): 40-48. DOI: 10.3174/ajnr.A6365.
[34]
WEI H H, YANG Y, FU F F, et al. Diagnosis and prognosis prediction of glioma based on multimodal MRI radiomics and deep learning[J]. Chin J Magn Reson Imaging, 2023, 14(5): 175-180. DOI: 10.12015/issn.1674-8034.2023.05.031.
[35]
COSGRIFF C V, CELI L A. Deep learning for risk assessment: all about automatic feature extraction[J]. Br J Anaesth, 2020, 124(2): 131-133. DOI: 10.1016/j.bja.2019.10.017.
[36]
WU X, GUO Y, SA Y, et al. Contrast-enhanced spectral mammography-based prediction of non-sentinel lymph node metastasis and axillary tumor burden in patients with breast cancer[J/OL]. Front Oncol, 2022, 12: 823897 [2023-08-01]. https://pubmed.ncbi.nlm.nih.gov/35615151/. DOI: 10.3389/fonc.2022.823897.
[37]
WANG J, ZHENG X, ZHANG J, et al. An MRI-based radiomics signature as a pretreatment noninvasive predictor of overall survival and chemotherapeutic benefits in lower-grade gliomas[J]. Eur Radiol, 2021, 31(4): 1785-1794. DOI: 10.1007/s00330-020-07581-3.

PREV A study on rs-fMRI dynamic functional network connectivity in patients with type 2 diabetic
NEXT Percent amplitude of fluctuation study in primary angle-closure glaucoma based on resting-state fMRI
  


LINK


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