DOI QR코드

DOI QR Code

Noncontrast Computed Tomography-Based Radiomics Analysis in Discriminating Early Hematoma Expansion after Spontaneous Intracerebral Hemorrhage

  • Zuhua Song (Department of Radiology, The Second Affiliated Hospital of Chongqing Medical University) ;
  • Dajing Guo (Department of Radiology, The Second Affiliated Hospital of Chongqing Medical University) ;
  • Zhuoyue Tang (Department of Radiology, Chongqing General Hospital) ;
  • Huan Liu (GE Healthcare) ;
  • Xin Li (Department of Radiology, The Second Affiliated Hospital of Chongqing Medical University) ;
  • Sha Luo (Department of Radiology, The Second Affiliated Hospital of Chongqing Medical University) ;
  • Xueying Yao (Department of Radiology, The Second Affiliated Hospital of Chongqing Medical University) ;
  • Wenlong Song (Department of Radiology, The Second Affiliated Hospital of Chongqing Medical University) ;
  • Junjie Song (Department of Radiology, The Second Affiliated Hospital of Chongqing Medical University) ;
  • Zhiming Zhou (Department of Radiology, The Second Affiliated Hospital of Chongqing Medical University)
  • Received : 2020.03.10
  • Accepted : 2020.07.02
  • Published : 2021.03.01

Abstract

Objective: To determine whether noncontrast computed tomography (NCCT) models based on multivariable, radiomics features, and machine learning (ML) algorithms could further improve the discrimination of early hematoma expansion (HE) in patients with spontaneous intracerebral hemorrhage (sICH). Materials and Methods: We retrospectively reviewed 261 patients with sICH who underwent initial NCCT within 6 hours of ictus and follow-up CT within 24 hours after initial NCCT, between April 2011 and March 2019. The clinical characteristics, imaging signs and radiomics features extracted from the initial NCCT images were used to construct models to discriminate early HE. A clinical-radiologic model was constructed using a multivariate logistic regression (LR) analysis. Radiomics models, a radiomics-radiologic model, and a combined model were constructed in the training cohort (n = 182) and independently verified in the validation cohort (n = 79). Receiver operating characteristic analysis and the area under the curve (AUC) were used to evaluate the discriminative power. Results: The AUC of the clinical-radiologic model for discriminating early HE was 0.766. The AUCs of the radiomics model for discriminating early HE built using the LR algorithm in the training and validation cohorts were 0.926 and 0.850, respectively. The AUCs of the radiomics-radiologic model in the training and validation cohorts were 0.946 and 0.867, respectively. The AUCs of the combined model in the training and validation cohorts were 0.960 and 0.867, respectively. Conclusion: NCCT models based on multivariable, radiomics features and ML algorithm could improve the discrimination of early HE. The combined model was the best recommended model to identify sICH patients at risk of early HE.

Keywords

Acknowledgement

This study was supported by the medical research Key Program of the combination of Chongqing National health commission and Chongqing science and technology bureau, China (no 2019ZDXM010); the Basic and Frontier Research Project of Chongqing, China (no cstc2016jcyjA0294).

References

  1. Ziai WC, Carhuapoma JR. Intracerebral hemorrhage. Continuum (Minneap Minn) 2018;24:1603-1622
  2. Brouwers HB, Chang Y, Falcone GJ, Cai X, Ayres AM, Battey TW, et al. Predicting hematoma expansion after primary intracerebral hemorrhage. JAMA Neurol 2014;71:158-164 https://doi.org/10.1001/jamaneurol.2013.5433
  3. Poon MT, Fonville AF, Al-Shahi Salman R. Long-term prognosis after intracerebral haemorrhage: systematic review and meta-analysis. J Neurol Neurosurg Psychiatry 2014;85:660-667 https://doi.org/10.1136/jnnp-2013-306476
  4. Hansen BM, Nilsson OG, Anderson H, Norrving B, Saveland H, Lindgren A. Long term (13 years) prognosis after primary intracerebral haemorrhage: a prospective population based study of long term mortality, prognostic factors and causes of death. J Neurol Neurosurg Psychiatry 2013;84:1150-1155 https://doi.org/10.1136/jnnp-2013-305200
  5. Brouwers HB, Greenberg SM. Hematoma expansion following acute intracerebral hemorrhage. Cerebrovasc Dis 2013;35:195-201 https://doi.org/10.1159/000346599
  6. Dowlatshahi D, Demchuk AM, Flaherty ML, Ali M, Lyden PL, Smith EE, et al. Defining hematoma expansion in intracerebral hemorrhage: relationship with patient outcomes. Neurology 2011;76:1238-1244 https://doi.org/10.1212/WNL.0b013e3182143317
  7. Davis SM, Broderick J, Hennerici M, Brun NC, Diringer MN, Mayer SA, et al. Hematoma growth is a determinant of mortality and poor outcome after intracerebral hemorrhage. Neurology 2006;66:1175-1181 https://doi.org/10.1212/01.wnl.0000208408.98482.99
  8. Wada R, Aviv RI, Fox AJ, Sahlas DJ, Gladstone DJ, Tomlinson G, et al. CT angiography "spot sign" predicts hematoma expansion in acute intracerebral hemorrhage. Stroke 2007;38:1257-1262 https://doi.org/10.1161/01.STR.0000259633.59404.f3
  9. Li Q, Zhang G, Xiong X, Wang XC, Yang WS, Li KW, et al. Black hole sign: novel imaging marker that predicts hematoma growth in patients with intracerebral hemorrhage. Stroke 2016;47:1777-1781 https://doi.org/10.1161/STROKEAHA.116.013186
  10. Yu Z, Zheng J, He M, Guo R, Ma L, You C, et al. Accuracy of swirl sign for predicting hematoma enlargement in intracerebral hemorrhage: a meta-analysis. J Neurol Sci 2019;399:155-160 https://doi.org/10.1016/j.jns.2019.02.032
  11. Li Q, Zhang G, Huang YJ, Dong MX, Lv FJ, Wei X, et al. Blend sign on computed tomography: novel and reliable predictor for early hematoma growth in patients with intracerebral hemorrhage. Stroke 2015;46:2119-2123 https://doi.org/10.1161/STROKEAHA.115.009185
  12. Boulouis G, Morotti A, Brouwers HB, Charidimou A, Jessel MJ, Auriel E, et al. Association between hypodensities detected by computed tomography and hematoma expansion in patients with intracerebral hemorrhage. JAMA Neurol 2016;73:961-968 https://doi.org/10.1001/jamaneurol.2016.1218
  13. Barras CD, Tress BM, Christensen S, MacGregor L, Collins M, Desmond PM, et al. Density and shape as CT predictors of intracerebral hemorrhage growth. Stroke 2009;40:1325-1331 https://doi.org/10.1161/STROKEAHA.108.536888
  14. Kumar V, Gu Y, Basu S, Berglund A, Eschrich SA, Schabath MB, et al. Radiomics: the process and the challenges. Magn Reson Imaging 2012;30:1234-1248 https://doi.org/10.1016/j.mri.2012.06.010
  15. Lambin P, Rios-Velazquez E, Leijenaar R, Carvalho S, van Stiphout RG, Granton P, et al. Radiomics: extracting more information from medical images using advanced feature analysis. Eur J Cancer 2012;48:441-446 https://doi.org/10.1016/j.ejca.2011.11.036
  16. Tian Q, Yan LF, Zhang X, Zhang X, Hu YC, Han Y, et al. Radiomics strategy for glioma grading using texture features from multiparametric MRI. J Magn Reson Imaging 2018;48:1518-1528 https://doi.org/10.1002/jmri.26010
  17. Ginsburg SB, Algohary A, Pahwa S, Gulani V, Ponsky L, Aronen HJ, et al. Radiomic features for prostate cancer detection on MRI differ between the transition and peripheral zones: preliminary findings from a multi-institutional study. J Magn Reson Imaging 2017;46:184-193 https://doi.org/10.1002/jmri.25562
  18. Ma C, Zhang Y, Niyazi T, Wei J, Guocai G, Liu J, et al. Radiomics for predicting hematoma expansion in patients with hypertensive intraparenchymal hematomas. Eur J Radiol 2019;115:10-15 https://doi.org/10.1016/j.ejrad.2019.04.001
  19. Shen Q, Shan Y, Hu Z, Chen W, Yang B, Han J, et al. Quantitative parameters of CT texture analysis as potential markers for early prediction of spontaneous intracranial hemorrhage enlargement. Eur Radiol 2018;28:4389-4396 https://doi.org/10.1007/s00330-018-5364-8
  20. Xie H, Ma S, Wang X, Zhang X. Noncontrast computer tomography-based radiomics model for predicting intracerebral hemorrhage expansion: preliminary findings and comparison with conventional radiological model. Eur Radiol 2020;30:87-98 https://doi.org/10.1007/s00330-019-06378-3
  21. Demchuk AM, Dowlatshahi D, Rodriguez-Luna D, Molina CA, Blas YS, Dzialowski I, et al. Prediction of haematoma growth and outcome in patients with intracerebral haemorrhage using the CT-angiography spot sign (PREDICT): a prospective observational study. Lancet Neurol 2012;11:307-314 https://doi.org/10.1016/S1474-4422(12)70038-8
  22. Marini S, Morotti A, Ayres AM, Crawford K, Kourkoulis CE, Lena UK, et al. Sex differences in intracerebral hemorrhage expansion and mortality. J Neurol Sci 2017;379:112-116 https://doi.org/10.1016/j.jns.2017.05.057
  23. Jurukovska-Nospal M, Arsova V, Levchanska J, SidovskaIvanovska B. Effects of statins (atorvastatin) on serum lipoprotein levels in patients with primary hyperlipidemia and coronary heart disease. Prilozi 2007;28:137-148
  24. Zhang F, Zhang S, Tao C, Yang Z, Li X, You C, et al. Association between serum glucose level and spot sign in intracerebral hemorrhage. Medicine (Baltimore) 2019;98:e14748
  25. Allen CL, Bayraktutan U. Antioxidants attenuate hyperglycaemia-mediated brain endothelial cell dysfunction and blood-brain barrier hyperpermeability. Diabetes Obes Metab 2009;11:480-490
  26. Asadi H, Dowling R, Yan B, Mitchell P. Machine learning for outcome prediction of acute ischemic stroke post intra-arterial therapy. PLoS One 2014;9:e88225
  27. Singal AG, Mukherjee A, Elmunzer BJ, Higgins PD, Lok AS, Zhu J, et al. Machine learning algorithms outperform conventional regression models in predicting development of hepatocellular carcinoma. Am J Gastroenterol 2013;108:1723-1730 https://doi.org/10.1038/ajg.2013.332
  28. Li H, Xie Y, Wang X, Chen F, Sun J, Jiang X. Radiomics features on non-contrast computed tomography predict early enlargement of spontaneous intracerebral hemorrhage. Clin Neurol Neurosurg 2019;185:105491