Fully automated stroke tissue estimation using random forest classifiers (FASTER)

Several clinical trials have recently proven the efficacy of mechanical thrombectomy for treating ischemic stroke, within a six-hour window for therapy. To move beyond treatment windows and toward personalized risk assessment, it is essential to accurately identify the extent of tissue-at-risk (“penumbra”). We introduce a fully automated method to estimate the penumbra volume using multimodal MRI (diffusion-weighted imaging, a T2w- and T1w contrast-enhanced sequence, and dynamic susceptibility contrast perfusion MRI). The method estimates tissue-at-risk by predicting tissue damage in the case of both persistent occlusion and of complete recanalization. When applied to 19 test cases with a thrombolysis in cerebral infarction grading of 1–2a, mean overestimation of final lesion volume was 30 ml, compared with 121 ml for manually corrected thresholding. Predicted tissue-at-risk volume was positively correlated with final lesion volume (p < 0.05). We conclude that prediction of tissue damage in the event of either persistent occlusion or immediate and complete recanalization, from spatial features derived from MRI, provides a substantial improvement beyond predefined thresholds. It may serve as an alternative method for identifying tissue-at-risk that may aid in treatment selection in ischemic stroke.

[1]  Douglas G. Altman,et al.  Measurement in Medicine: The Analysis of Method Comparison Studies , 1983 .

[2]  G. Schroth,et al.  Three-Month and Long-Term Outcomes and Their Predictors in Acute Basilar Artery Occlusion Treated With Intra-Arterial Thrombolysis , 2011, Stroke.

[3]  G. Schlaug,et al.  The ischemic penumbra: operationally defined by diffusion and perfusion MRI. , 1999, Neurology.

[4]  Homer H. Pien,et al.  Stroke Tissue Outcome Prediction Using A Spatially-Correlated Model , 2008 .

[5]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[6]  Xavier Robin,et al.  pROC: an open-source package for R and S+ to analyze and compare ROC curves , 2011, BMC Bioinformatics.

[7]  Stefan Bauer,et al.  Towards automatic MRI volumetry for treatment selection in acute ischemic stroke patients , 2014, 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[8]  Gerhard Schroth,et al.  Endovascular Therapy of 623 Patients With Anterior Circulation Stroke , 2012, Stroke.

[9]  Kaspar Anton Schindler,et al.  Adverse Effect of Early Epileptic Seizures in Patients Receiving Endovascular Therapy for Acute Stroke , 2012, Stroke.

[10]  H Handels,et al.  Comparison of 10 TTP and Tmax Estimation Techniques for MR Perfusion-Diffusion Mismatch Quantification in Acute Stroke , 2013, American Journal of Neuroradiology.

[11]  Nils Daniel Forkert,et al.  Multivariate Dynamic Prediction of Ischemic Infarction and Tissue Salvage as a Function of Time and Degree of Recanalization , 2015, Journal of cerebral blood flow and metabolism : official journal of the International Society of Cerebral Blood Flow and Metabolism.

[12]  Claus Kiefer,et al.  Factors that determine penumbral tissue loss in acute ischaemic stroke. , 2013, Brain : a journal of neurology.

[13]  Kim Mouridsen,et al.  Predicting Tissue Outcome From Acute Stroke Magnetic Resonance Imaging: Improving Model Performance by Optimal Sampling of Training Data , 2009, Stroke.

[14]  P. Schellinger,et al.  Strengthening the link: the critical role of children in the stroke chain of recovery. , 2008, Stroke.

[15]  Reto Meuli,et al.  Perfusion-CT Assessment of Infarct Core and Penumbra: Receiver Operating Characteristic Curve Analysis in 130 Patients Suspected of Acute Hemispheric Stroke , 2006, Stroke.

[16]  R. Bammer,et al.  Real‐time diffusion‐perfusion mismatch analysis in acute stroke , 2010, Journal of magnetic resonance imaging : JMRI.

[17]  Antonio Criminisi,et al.  Decision Forests: A Unified Framework for Classification, Regression, Density Estimation, Manifold Learning and Semi-Supervised Learning , 2012, Found. Trends Comput. Graph. Vis..

[18]  S. Allassonnière,et al.  Medical image analysis methods in MR/CT-imaged acute-subacute ischemic stroke lesion: Segmentation, prediction and insights into dynamic evolution simulation models. A critical appraisal , 2012, NeuroImage: Clinical.

[19]  Hester F. Lingsma,et al.  A randomized trial of intraarterial treatment for acute ischemic stroke. , 2015, The New England journal of medicine.

[20]  S. Jasko,et al.  Therapy , 1881, The American journal of dental science.

[21]  M. Krause,et al.  Endovascular therapy for ischemic stroke with perfusion-imaging selection. , 2015, The New England journal of medicine.

[22]  Hiroki Shirato,et al.  Difference in tracer delay-induced effect among deconvolution algorithms in CT perfusion analysis: quantitative evaluation with digital phantoms. , 2009, Radiology.

[23]  Max Wintermark,et al.  Imaging recommendations for acute stroke and transient ischemic attack patients: a joint statement by the American Society of Neuroradiology, the American College of Radiology and the Society of NeuroInterventional Surgery. , 2013, Journal of the American College of Radiology : JACR.

[24]  Ivana Galinovic,et al.  A Multicenter, Randomized, Double-Blind, Placebo-Controlled Trial to Test Efficacy and Safety of Magnetic Resonance Imaging-Based Thrombolysis in Wake-up Stroke (WAKE-UP) , 2014, International journal of stroke : official journal of the International Stroke Society.

[25]  Stefan Bauer,et al.  A Skull-Stripping Filter for ITK , 2013, The Insight Journal.

[26]  Xiao Hu,et al.  Regional Prediction of Tissue Fate in Acute Ischemic Stroke , 2012, Annals of Biomedical Engineering.

[27]  R. Higashida,et al.  Trial Design and Reporting Standards for Intra-Arterial Cerebral Thrombolysis for Acute Ischemic Stroke , 2003, Stroke.

[28]  G. Schroth,et al.  Recanalization and Outcome After Intra-Arterial Thrombolysis in Middle Cerebral Artery and Internal Carotid Artery Occlusion: Does Sex Matter? , 2007, Stroke.

[29]  A. Demchuk,et al.  Thrombectomy within 8 hours after symptom onset in ischemic stroke. , 2015, The New England journal of medicine.

[30]  Eric E. Smith,et al.  Randomized assessment of rapid endovascular treatment of ischemic stroke. , 2015, The New England journal of medicine.

[31]  Stefan Bauer,et al.  Patient-Specific Semi-supervised Learning for Postoperative Brain Tumor Segmentation , 2014, MICCAI.

[32]  Jan Sobesky,et al.  Identifying Thresholds for Penumbra and Irreversible Tissue Damage , 2004, Stroke.

[33]  J. Heo,et al.  Chronic cerebral hypoperfusion protects against acute focal ischemia, improves motor function, and results in vascular remodeling. , 2008, Current neurovascular research.

[34]  Z. Chaudhry,et al.  Infarct Volume Is a Pivotal Biomarker After Intra-Arterial Stroke Therapy , 2012, Stroke.

[35]  R. Meuli,et al.  The Acute STroke Registry and Analysis of Lausanne (ASTRAL): Design and Baseline Analysis of an Ischemic Stroke Registry Including Acute Multimodal Imaging , 2010, Stroke.

[36]  H. Diener,et al.  Stent-retriever thrombectomy after intravenous t-PA vs. t-PA alone in stroke. , 2015, The New England journal of medicine.

[37]  S. Bauer,et al.  Multi-Modal Glioblastoma Segmentation: Man versus Machine , 2014, PloS one.

[38]  Levin Häni,et al.  Segmenting the Ischemic Penumbra: A Decision Forest Approach with Automatic Threshold Finding , 2015, Brainles@MICCAI.

[39]  Roland Bammer,et al.  Geography, Structure, and Evolution of Diffusion and Perfusion Lesions in Diffusion and Perfusion Imaging Evaluation For Understanding Stroke Evolution (DEFUSE) , 2009, Stroke.

[40]  Fabien Scalzo,et al.  Deep learning of tissue fate features in acute ischemic stroke , 2015, 2015 IEEE International Conference on Bioinformatics and Biomedicine (BIBM).

[41]  B. Rosen,et al.  Infarct Prediction and Treatment Assessment with MRI-based Algorithms in Experimental Stroke Models , 2007, Journal of cerebral blood flow and metabolism : official journal of the International Society of Cerebral Blood Flow and Metabolism.

[42]  Manabu Inoue,et al.  MRI profile and response to endovascular reperfusion after stroke (DEFUSE 2): a prospective cohort study , 2012, The Lancet Neurology.

[43]  Scott Hamilton,et al.  Magnetic resonance imaging profiles predict clinical response to early reperfusion: The diffusion and perfusion imaging evaluation for understanding stroke evolution (DEFUSE) study , 2006, Annals of neurology.