Clot Analog Attenuation in Non-contrast CT Predicts Histology: an Experimental Study Using Machine Learning

Abstract

Exact histological clot composition remains unknown. The purpose of this study was to identify the best imaging variables to be extrapolated on clot composition and clarify variability in the imaging of thrombi by non-contrast CT. Using a CT-phantom and covering a wide range of histologies, we analyzed 80 clot analogs with respect to X-ray attenuation at 24 and 48 h after production. The mean, maximum, and minimum HU values for the axial and coronal reconstructions were recorded. Each thrombus underwent a corresponding histological analysis, together with a laboratory analysis of water and iron contents. Decision trees, a type of supervised machine learning, were used to select the primary variable altering attenuation and the best parameter for predicting histology. The decision trees selected red blood cells (RBCs) for correlation with all attenuation parameters (p < 0.001). Conversely, maximum attenuation on axial CT offered the greatest accuracy for discriminating up to four groups of clot histology (p < 0.001). Similar RBC-rich thrombi displayed variable imaging associated with different iron (p = 0.023) and white blood cell contents (p = 0.019). Water content varied among the different histologies but did not in itself account for the differences in attenuation. Independent factors determining clot attenuation were the RBCs (β = 0.33, CI = 0.219–0.441, p < 0.001) followed by the iron content (β = 0.005, CI = 0.0002–0.009, p = 0.042). Our findings suggest that it is possible to extract more and valuable information from NCCT that can be extrapolated to provide insights into clot histological and chemical composition.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4

Abbreviations

NCCT:

Non-contrast CT

RBCs:

Red blood cells

HU:

Hounsfield units

ROI:

Region of interest

WBCs:

White blood cells

SD:

Standard deviation

IQR:

Interquartile range

AIS:

Acute ischemic stroke

References

  1. 1.

    Campbell BC, Mitchell PJ, Kleinig TJ, Dewey HM, Churilov L, Yassi N, et al. Endovascular therapy for ischemic stroke with perfusion-imaging selection. N Engl J Med. 2015;372(11):1009–18.

    Article  CAS  Google Scholar 

  2. 2.

    Goyal M, Demchuk AM, Menon BK, Eesa M, Rempel JL, Thornton J, et al. Randomized assessment of rapid endovascular treatment of ischemic stroke. N Engl J Med. 2015;372(11):1019–30.

    Article  CAS  Google Scholar 

  3. 3.

    Berkhemer OA, Fransen PS, Beumer D, van den Berg LA, Lingsma HF, Yoo AJ, et al. A randomized trial of intraarterial treatment for acute ischemic stroke. N Engl J Med. 2015;372(1):11–20.

    Article  CAS  Google Scholar 

  4. 4.

    Jovin TG, Chamorro A, Cobo E, de Miquel MA, Molina CA, Rovira A, et al. Thrombectomy within 8 hours after symptom onset in ischemic stroke. N Engl J Med. 2015;372(24):2296–306.

    Article  CAS  Google Scholar 

  5. 5.

    Velasco Gonzalez A, Buerke B, Gorlich D, Chapot R, Smagge L, Velasco MV, et al. Variability in the decision-making process of acute ischemic stroke in difficult clinical and radiological constellations: analysis based on a cross-sectional interview-administered stroke questionnaire. Eur Radiol. 2019;29(11):6275–84.

    Article  Google Scholar 

  6. 6.

    Niesten JM, van der Schaaf IC, van der Graaf Y, Kappelle LJ, Biessels GJ, Horsch AD, et al. Predictive value of thrombus attenuation on thin-slice non-contrast CT for persistent occlusion after intravenous thrombolysis. Cerebrovasc Dis. 2014;37(2):116–22.

    Article  CAS  Google Scholar 

  7. 7.

    Puig J, Pedraza S, Demchuk A, Daunis IEJ, Termes H, Blasco G, et al. Quantification of thrombus Hounsfield units on noncontrast CT predicts stroke subtype and early recanalization after intravenous recombinant tissue plasminogen activator. AJNR Am J Neuroradiol. 2012;33(1):90–6.

    Article  CAS  Google Scholar 

  8. 8.

    Brinjikji W, Duffy S, Burrows A, Hacke W, Liebeskind D, Majoie C, et al. Correlation of imaging and histopathology of thrombi in acute ischemic stroke with etiology and outcome: a systematic review. J Neurointerv Surg. 2017;9(6):529–34.

    Article  Google Scholar 

  9. 9.

    Heo JH, Kim K, Yoo J, Kim YD, Nam HS, Kim EY. Computed tomography-based thrombus imaging for the prediction of recanalization after reperfusion therapy in stroke. J Stroke. 2017;19(1):40–9.

    Article  Google Scholar 

  10. 10.

    Froehler MT, Tateshima S, Duckwiler G, Jahan R, Gonzalez N, Vinuela F, et al. The hyperdense vessel sign on CT predicts successful recanalization with the Merci device in acute ischemic stroke. J Neurointerv Surg. 2013;5(4):289–93.

    Article  Google Scholar 

  11. 11.

    Mokin M, Morr S, Natarajan SK, Lin N, Snyder KV, Hopkins LN, et al. Thrombus density predicts successful recanalization with Solitaire stent retriever thrombectomy in acute ischemic stroke. J Neurointerv Surg. 2015;7(2):104–7.

    Article  Google Scholar 

  12. 12.

    Liebeskind DS, Sanossian N, Yong WH, Starkman S, Tsang MP, Moya AL, et al. CT and MRI early vessel signs reflect clot composition in acute stroke. Stroke. 2011;42(5):1237–43.

    Article  Google Scholar 

  13. 13.

    Moftakhar P, English JD, Cooke DL, Kim WT, Stout C, Smith WS, et al. Density of thrombus on admission CT predicts revascularization efficacy in large vessel occlusion acute ischemic stroke. Stroke. 2013;44(1):243–5.

    Article  Google Scholar 

  14. 14.

    Mair G, Boyd EV, Chappell FM, von Kummer R, Lindley RI, Sandercock P, et al. Sensitivity and specificity of the hyperdense artery sign for arterial obstruction in acute ischemic stroke. Stroke. 2015;46(1):102–7.

    Article  Google Scholar 

  15. 15.

    Brouwer PA, Brinjikji W, De Meyer SF. Clot pathophysiology: why is it clinically important? Neuroimaging Clin N Am. 2018;28(4):611–23.

    Article  Google Scholar 

  16. 16.

    Turk AS 3rd, Campbell JM, Spiotta A, Vargas J, Turner RD, Chaudry MI, et al. An investigation of the cost and benefit of mechanical thrombectomy for endovascular treatment of acute ischemic stroke. J Neurointerv Surg. 2014;6(1):77–80.

    Article  Google Scholar 

  17. 17.

    Shu L, Meyne J, Jansen O, Jensen-Kondering U. Manual thrombus density measurement depends on the method of thrombus delineation. J Stroke. 2018;20(3):411–2.

    Article  Google Scholar 

  18. 18.

    Angermaier A, Langner S. Thrombus density measurement is promising but technical standards are needed. J Neurointerv Surg. 2017. https://doi.org/10.1136/neurintsurg-2015-011866.

  19. 19.

    Bourcier R, Pautre R, Mirza M, Castets C, Darcourt J, Labreuche J, et al. MRI quantitative T2* mapping to predict dominant composition of in vitro thrombus. AJNR Am J Neuroradiol. 2019;40(1):59–64.

    Article  CAS  Google Scholar 

  20. 20.

    Bretzner M, Lopes R, McCarthy R, Corseaux D, Auger F, Gunning G, et al. Texture parameters of R2* maps are correlated with iron concentration and red blood cells count in clot analogs: a 7T micro-MRI study. J Neuroradiol. 2019. https://doi.org/10.1136/j.neurad.2019.10.004.

  21. 21.

    Janot K, Oliveira TR, Fromont-Hankard G, Annan M, Filipiak I, Barantin L, et al. Quantitative estimation of thrombus-erythrocytes using MRI. A phantom study with clot analogs and analysis by statistic regression models. J Neurointerv Surg. 2019. https://doi.org/10.1136/neurintsurg-2019-014950.

  22. 22.

    Fennell VS, Setlur Nagesh SV, Meess KM, Gutierrez L, James RH, Springer ME, et al. What to do about fibrin rich ‘tough clots’? Comparing the Solitaire stent retriever with a novel geometric clot extractor in an in vitro stroke model. J Neurointerv Surg. 2018;10(9):907–10.

    Article  Google Scholar 

  23. 23.

    Johnson S, Duffy S, Gunning G, Gilvarry M, McGarry JP, McHugh PE. Review of mechanical testing and modelling of thrombus material for vascular implant and device design. Ann Biomed Eng. 2017;45(11):2494–508.

    Article  CAS  Google Scholar 

  24. 24.

    Duffy S, Farrell M, McArdle K, Thornton J, Vale D, Rainsford E, et al. Novel methodology to replicate clot analogs with diverse composition in acute ischemic stroke. J Neurointerv Surg. 2017;9(5):486–91.

    Article  Google Scholar 

  25. 25.

    Brinjikji W, Michalak G, Kadirvel R, Dai D, Gilvarry M, Duffy S, et al. Utility of single-energy and dual-energy computed tomography in clot characterization: an in-vitro study. Interv Neuroradiol. 2017;23(3):279–84.

    Article  Google Scholar 

  26. 26.

    Cecchin E, De Marchi S, Querin F, Marin MG, Fiorentino R, Tesio F. Efficacy of hepatic computed tomography to detect iron overload in chronic hemodialysis. Kidney Int. 1990;37(3):943–50.

    Article  CAS  Google Scholar 

  27. 27.

    Jang S, Graffy PM, Ziemlewicz TJ, Lee SJ, Summers RM, Pickhardt PJ. Opportunistic osteoporosis screening at routine abdominal and thoracic CT: normative L1 trabecular attenuation values in more than 20 000 adults. Radiology. 2019;291(2):360–7.

    Article  Google Scholar 

  28. 28.

    New PF, Aronow S. Attenuation measurements of whole blood and blood fractions in computed tomography. Radiology. 1976;121(3 Pt. 1):635–40.

    Article  CAS  Google Scholar 

  29. 29.

    Voter WA, Lucaveche C, Erickson HP. Concentration of protein in fibrin fibers and fibrinogen polymers determined by refractive index matching. Biopolymers. 1986;25(12):2375–84.

    Article  CAS  Google Scholar 

  30. 30.

    Brown AE, Litvinov RI, Discher DE, Purohit PK, Weisel JW. Multiscale mechanics of fibrin polymer: gel stretching with protein unfolding and loss of water. Science. 2009;325(5941):741–4.

    Article  CAS  Google Scholar 

  31. 31.

    Gunning GM, McArdle K, Mirza M, Duffy S, Gilvarry M, Brouwer PA. Clot friction variation with fibrin content; implications for resistance to thrombectomy. J Neurointerv Surg. 2018;10(1):34–8.

    Article  Google Scholar 

  32. 32.

    Weafer FM, Duffy S, Machado I, Gunning G, Mordasini P, Roche E, et al. Characterization of strut indentation during mechanical thrombectomy in acute ischemic stroke clot analogs. J Neurointerv Surg. 2019;11:891–7.

    Article  Google Scholar 

  33. 33.

    van der Marel K, Chueh JY, Brooks OW, King RM, Marosfoi MG, Langan ET, et al. Quantitative assessment of device-clot interaction for stent retriever thrombectomy. J Neurointerv Surg. 2016;8(12):1278–82.

    Article  Google Scholar 

  34. 34.

    Chueh JY, Kuhn AL, Puri AS, Wilson SD, Wakhloo AK, Gounis MJ. Reduction in distal emboli with proximal flow control during mechanical thrombectomy: a quantitative in vitro study. Stroke. 2013;44(5):1396–401.

    Article  Google Scholar 

Download references

Acknowledgments

The authors thank the team headed by Dr. Ray McCarthy for providing clot analogs weekly as the source material for our study free of charge. Aglaé Velasco Gonzalez (Neuroradiology) performed this study in collaboration with the Institute of Biostatistics and Clinical Research and Institute of Neuropathology at the Faculty of Medicine, Westfälische Wilhelms-Universität Münster (WWU). We also thank Dr. Senner Volker for coordinating and Mrs. Andrea Rothaus of the neuropathology laboratory for preparing and staining the clot analog samples. Special thanks go to our X-ray technicians who always found a way to complete the experiments on time and whose support was invaluable. Finally, the authors thank the University of Muenster for giving us the time to complete this project.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Aglae Velasco Gonzalez.

Ethics declarations

Conflict of Interest

Aglaé Velasco González performed this study in the context of a program for research backed by the WWU University. One of the authors (Ray McCarthy) is an employee of Cerenovus. Authors who neither advise nor work for the industry had exclusive control over designing and performing the experiment, the data, and data analysis. This study received no industry financial support. All authors have approved the final manuscript.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

ESM 1

(DOCX 2356 kb)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Velasco Gonzalez, A., Buerke, B., Görlich, D. et al. Clot Analog Attenuation in Non-contrast CT Predicts Histology: an Experimental Study Using Machine Learning. Transl. Stroke Res. 11, 940–949 (2020). https://doi.org/10.1007/s12975-019-00766-z

Download citation

Keywords

  • Blood clot
  • Helical CT
  • Red blood cells
  • Decision trees
  • Iron