scholarly journals Automatic segmentation of cerebral infarcts in follow-up computed tomography images with convolutional neural networks

2019 ◽  
Vol 12 (9) ◽  
pp. 848-852 ◽  
Author(s):  
Renan Sales Barros ◽  
Manon L Tolhuisen ◽  
Anna MM Boers ◽  
Ivo Jansen ◽  
Elena Ponomareva ◽  
...  
Keyword(s):  
Neural Networks ◽  
Ischemic Stroke ◽  
Acute Ischemic Stroke ◽  
Convolutional Neural Networks ◽  
Infarct Volume ◽  
Automatic Segmentation ◽  
Ct Images ◽  
Reference Standard ◽  
Computed Tomography Images

Background and purposeInfarct volume is a valuable outcome measure in treatment trials of acute ischemic stroke and is strongly associated with functional outcome. Its manual volumetric assessment is, however, too demanding to be implemented in clinical practice.ObjectiveTo assess the value of convolutional neural networks (CNNs) in the automatic segmentation of infarct volume in follow-up CT images in a large population of patients with acute ischemic stroke.Materials and methodsWe included CT images of 1026 patients from a large pooling of patients with acute ischemic stroke. A reference standard for the infarct segmentation was generated by manual delineation. We introduce three CNN models for the segmentation of subtle, intermediate, and severe hypodense lesions. The fully automated infarct segmentation was defined as the combination of the results of these three CNNs. The results of the three-CNNs approach were compared with the results from a single CNN approach and with the reference standard segmentations.ResultsThe median infarct volume was 48 mL (IQR 15–125 mL). Comparison between the volumes of the three-CNNs approach and manually delineated infarct volumes showed excellent agreement, with an intraclass correlation coefficient (ICC) of 0.88. Even better agreement was found for severe and intermediate hypodense infarcts, with ICCs of 0.98 and 0.93, respectively. Although the number of patients used for training in the single CNN approach was much larger, the accuracy of the three-CNNs approach strongly outperformed the single CNN approach, which had an ICC of 0.34.ConclusionConvolutional neural networks are valuable and accurate in the quantitative assessment of infarct volumes, for both subtle and severe hypodense infarcts in follow-up CT images. Our proposed three-CNNs approach strongly outperforms a more straightforward single CNN approach.

Abstract P538: A Detailed Analysis of Intracranial Hemorrhage After Endovascular Treatment in Acute Ischemic Stroke Due to Large Vessel Occlusion in the Escape-NA1 Trial

Stroke ◽  
2021 ◽  
Vol 52 (Suppl_1) ◽  
Author(s):  
Johanna Ospel ◽  
Michael D Hill ◽  
Nima Kashani ◽  
Arnuv Mayank ◽  
Nishita Singh ◽  
...  
Keyword(s):  
Ischemic Stroke ◽  
Acute Ischemic Stroke ◽  
Intracranial Hemorrhage ◽  
Intraventricular Hemorrhage ◽  
Infarct Volume ◽  
Prognostic Impact ◽  
Vessel Occlusion ◽  
Good Outcome

Purpose: We investigated the prevalence and prognostic impact on outcome of any intracranial hemorrhage, hemorrhage morphology, type and volume in acute ischemic stroke patients undergoing mechanical thrombectomy. Methods: Prevalence of intracranial hemorrhage, hemorrhage type, morphology and volume was determined on 24h follow-up imaging (non contrast head CT or gradient-echo/susceptibility-weighted MRI). Proportions of good outcome (mRS 0-2 at 90 days) were reported for patients with vs. without any intracranial hemorrhage. Multivariable logistic regression with adjustment for key minimization variables and total infarct volume was performed to obtain adjusted effect size estimates for hemorrhage type and volume on good outcome. Results: Hemorrhage on follow up-imaging was seen in 372/1097 (33.9%) patients, among them 126 (33.9%) with hemorrhagic infarction (HI) type 1, 108 (29.0%) with HI-2, 72 /19.4%) with parenchymal hematoma (PH) type 1, 37 (10.0) with PH2, 8 (2.2%) with remote PH and 21 (5.7%) with extra-parenchymal/intraventricular hemorrhage. Good outcomes were less often achieved by patients with hemorrhage on follow-up imaging (164/369 [44.4%] vs. 500/720 [69.4%]). Any type of intracranial hemorrhage was strongly associated with decreased chances of good outcome ( adj OR 0.62 [CI 95 0.44 - 0.87]). The effect of hemorrhage was driven by both PH hemorrhage sub-type [PH-1 ( adj OR 0.39 [CI 95 0.21 - 0.72]), PH-2 ( adj OR 0.15 [CI 95 0.05 - 0.50])] and extra-parenchymal/intraventricular hemorrhage ( adj OR 0.60 (0.20-1.78) Petechial hemorrhages (HI-1 and HI-2) were not associated with poorer outcomes. Hemorrhage volume ( adj OR 0.97 [CI 95 0.05 - 0.99] per ml increase) was significantly associated with decreased chances of good outcome. Conclusion: Presence of any hemorrhage on follow-up imaging was seen in one third of patients and strongly associated with decreased chances of good outcome.

Abstract 3681: Early Anticoagulation Is Not Associated With Hemorrhagic Transformation Of Ischemic Stroke In Patients With Atrial Fibrillation

Stroke ◽  
2012 ◽  
Vol 43 (suppl_1) ◽  
Author(s):  
H. B Brouwers ◽  
Svetlana Lorenzano ◽  
Lyndsey H Starks ◽  
David M Greer ◽  
Steven K Feske ◽  
...  
Keyword(s):  
Atrial Fibrillation ◽  
Ischemic Stroke ◽  
Acute Ischemic Stroke ◽  
Clinical Data ◽  
Infarct Volume ◽  
Hemorrhagic Transformation ◽  
P Value ◽  
Long Term Outcome ◽  
Nihss Score

Purpose: Hemorrhagic transformation (HT) is a common and potentially devastating complication of ischemic stroke, however its prevalence, predictors, and outcome remain unclear. Early anticoagulation is thought to be a risk factor for HT which raises the clinical question when to (re)start anticoagulation in ischemic stroke patients who have a compelling indication, such as atrial fibrillation. We conducted a prospective cohort study to address this question and to identify association of hemorrhagic transformation with outcome measures in patients with atrial fibrillation in the setting of acute ischemic stroke. Materials and Methods: We performed a prospective study which enrolled consecutive patients admitted with acute ischemic stroke presenting to a single center over a three-year period. As part of the observational study, baseline clinical data and stroke characteristics as well as 3 month functional outcome were collected. For this sub-study, we restricted the analysis to subjects diagnosed with atrial fibrillation. CT and MRI scans were reviewed by experienced readers, blinded to clinical data, to assess for hemorrhagic transformation (using ECASS 2 criteria), microbleeds and infarct volumes in both admission and follow-up scans. Clinical and outcome data were analyzed for association with hemorrhagic transformation. Results: Of 94 patients, 63 had a history of atrial fibrillation (67.0%) and 31 had newly discovered atrial fibrillation (33.0%). We identified HT in 3 of 94 baseline scans (3.2%) and 22 of 48 follow-up scans (45.8%) obtained a median of 3 days post-stroke. In-hospital initiation of either anti-platelet (n = 36; OR 0.34 [95% CI 0.10-1.16], p-value = 0.09) or anticoagulation with unfractionated intravenous heparin or low molecular weight heparin (n = 72; OR 0.25 [95% CI 0.06-1.15], p-value = 0.08) was not associated with HT. Initial NIH Stroke Scale (NIHSS) score (median 13.0 [IQR 15.0] vs. 7.0 [IQR 10.0], p-value = 0.029) and baseline infarct volume (median 17 [IQR 42.03] vs. 5 [IQR 10.95], p-value = 0.011) were significantly higher in patients with HT compared to those without. Hemorrhagic transformation was associated with a significantly higher 48-hour median NIHSS score (20 [IQR 3.0] vs. 2 [IQR 3.25], p-value = 0.007) and larger final infarct volume (81.40 [IQR 82.75] vs. 9.95 [IQR 19.73], p-value < 0.001). Finally, we found a trend towards poorer 3-month modified Rankin Scale scores in subjects with HT (OR 11.25 [95% CI 0.97-130.22], p-value = 0.05). Conclusion: In patients with atrial fibrillation, initial NIHSS score and baseline infarct volume are associated with hemorrhagic transformation in acute ischemic stroke. Early initiation of antithrombotic therapy was not associated with hemorrhagic transformation. Patients with hemorrhagic transformation were found to have a poorer short and long term outcome and larger final infarct volumes.

scholarly journals Computed Tomography Perfusion–Based Machine Learning Model Better Predicts Follow-Up Infarction in Patients With Acute Ischemic Stroke

Stroke ◽  
2021 ◽  
Vol 52 (1) ◽  
pp. 223-231
Author(s):  
Hulin Kuang ◽  
Wu Qiu ◽  
Anna M. Boers ◽  
Scott Brown ◽  
Keith Muir ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Blood Flow ◽  
Cerebral Blood Flow ◽  
Ischemic Stroke ◽  
Acute Ischemic Stroke ◽  
Infarct Volume ◽  
Interquartile Range ◽  
Time Dependent ◽  
Computed Tomography Perfusion

Background and Purpose: Prediction of infarct extent among patients with acute ischemic stroke using computed tomography perfusion is defined by predefined discrete computed tomography perfusion thresholds. Our objective is to develop a threshold-free computed tomography perfusion–based machine learning (ML) model to predict follow-up infarct in patients with acute ischemic stroke. Methods: Sixty-eight patients from the PRoveIT study (Measuring Collaterals With Multi-Phase CT Angiography in Patients With Ischemic Stroke) were used to derive a ML model using random forest to predict follow-up infarction voxel by voxel, and 137 patients from the HERMES study (Highly Effective Reperfusion Evaluated in Multiple Endovascular Stroke Trials) were used to test the derived ML model. Average map, T max , cerebral blood flow, cerebral blood volume, and time variables including stroke onset-to-imaging and imaging-to-reperfusion time, were used as features to train the ML model. Spatial and volumetric agreement between the ML model predicted follow-up infarct and actual follow-up infarct were assessed. Relative cerebral blood flow <0.3 threshold using RAPID software and time-dependent T max thresholds were compared with the ML model. Results: In the test cohort (137 patients), median follow-up infarct volume predicted by the ML model was 30.9 mL (interquartile range, 16.4–54.3 mL), compared with a median 29.6 mL (interquartile range, 11.1–70.9 mL) of actual follow-up infarct volume. The Pearson correlation coefficient between 2 measurements was 0.80 (95% CI, 0.74–0.86, P <0.001) while the volumetric difference was −3.2 mL (interquartile range, −16.7 to 6.1 mL). Volumetric difference with the ML model was smaller versus the relative cerebral blood flow <0.3 threshold and the time-dependent T max threshold ( P <0.001). Conclusions: A ML using computed tomography perfusion data and time estimates follow-up infarction in patients with acute ischemic stroke better than current methods.

scholarly journals Endovascular Treatment Effect Diminishes With Increasing Thrombus Perviousness

Stroke ◽  
2021 ◽  
Author(s):  
Manon Kappelhof ◽  
Manon L. Tolhuisen ◽  
Kilian M. Treurniet ◽  
Bruna G. Dutra ◽  
Heitor Alves ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Ischemic Stroke ◽  
Endovascular Treatment ◽  
Acute Ischemic Stroke ◽  
Functional Outcome ◽  
Odds Ratio ◽  
Infarct Volume ◽  
Treatment Allocation ◽  
Common Odds Ratio

Background and Purpose: Thrombus perviousness estimates residual flow along a thrombus in acute ischemic stroke, based on radiological images, and may influence the benefit of endovascular treatment for acute ischemic stroke. We aimed to investigate potential endovascular treatment (EVT) effect modification by thrombus perviousness. Methods: We included 443 patients with thin-slice imaging available, out of 1766 patients from the pooled HERMES (Highly Effective Reperfusion Evaluated in Multiple Endovascular Stroke trials) data set of 7 randomized trials on EVT in the early window (most within 8 hours). Control arm patients (n=233) received intravenous alteplase if eligible (212/233; 91%). Intervention arm patients (n=210) received additional EVT (prior alteplase in 178/210; 85%). Perviousness was quantified by thrombus attenuation increase on admission computed tomography angiography compared with noncontrast computed tomography. Multivariable regression analyses were performed including multiplicative interaction terms between thrombus attenuation increase and treatment allocation. In case of significant interaction, subgroup analyses by treatment arm were performed. Our primary outcome was 90-day functional outcome (modified Rankin Scale score), resulting in an adjusted common odds ratio for a one-step shift towards improved outcome. Secondary outcomes were mortality, successful reperfusion (extended Thrombolysis in Cerebral Infarction score, 2B–3), and follow-up infarct volume (in mL). Results: Increased perviousness was associated with improved functional outcome. After adding a multiplicative term of thrombus attenuation increase and treatment allocation, model fit improved significantly ( P =0.03), indicating interaction between perviousness and EVT benefit. Control arm patients showed significantly better outcomes with increased perviousness (adjusted common odds ratio, 1.2 [95% CI, 1.1–1.3]). In the EVT arm, no significant association was found (adjusted common odds ratio, 1.0 [95% CI, 0.9–1.1]), and perviousness was not significantly associated with successful reperfusion. Follow-up infarct volume (12% [95% CI, 7.0–17] per 5 Hounsfield units) and chance of mortality (adjusted odds ratio, 0.83 [95% CI, 0.70–0.97]) decreased with higher thrombus attenuation increase in the overall population, without significant treatment interaction. Conclusions: Our study suggests that the benefit of best medical care including alteplase, compared with additional EVT, increases in patients with more pervious thrombi.

scholarly journals Intravenous fibrinolysis plus endovascular thrombectomy versus direct endovascular thrombectomy for anterior circulation acute ischemic stroke: clinical and infarct volume results.

2019 ◽  
Author(s):  
Massimo Gamba ◽  
Nicola Gilberti ◽  
Enrico Premi ◽  
Angelo Costa ◽  
Michele Frigerio ◽  
...  
Keyword(s):  
Ischemic Stroke ◽  
Acute Ischemic Stroke ◽  
Safety Profile ◽  
Infarct Volume ◽  
Intravenous Thrombolysis ◽  
Functional Independence ◽  
Anterior Circulation ◽  
Endovascular Thrombectomy ◽  
Symptomatic Intracerebral Hemorrhage

Abstract Background and Purpose endovascular therapy (ET) is the standard of care for anterior circulation acute ischemic stroke (AIS) caused by large vessel occlusion (LVO). The role of adjunctive intravenous thrombolysis (IVT) in these patients is still unclear. The present study aims to test whether IVT plus ET (CoT, combined therapy) provides additional benefits over direct ET for anterior circulation AIS by LVO. Methods we performed a single center retrospective observational study of patients with AIS caused by anterior circulation LVO, referred to our center between January 2014 and January 2017 and treated with ET. The patients were divided in 2 groups based on the treatment they received: CoT and, if IVT contraindicated, direct ET. We compared functional recovery (modified Rankin at 3-months follow-up), recanalization rate (thrombolysis in cerebral infarction [TICI] score) and time, early follow-up infarct volume (EFIV) (for recanalized patients only) as well as safety profile, defined as symptomatic intracerebral hemorrhage (sICH) and 3-month mortality, between groups. Results 145 subjects were included in the study, 70 in direct ET group and 75 in CoT group. Patients who received CoT presented more frequently a functional independence at 3-months follow-up compared to patients who received direct ET (mRS score 0-1: 48.5% vs 18.6%; P<0.001. mRS score 0-2: 67.1% vs 37.3%; P<0.001), higher first-pass success rate (62.7% vs 38.6%, P<0.05), higher recanalization rate (84.3% vs 65.3%; P=0.009) and, in recanalized subjects, smaller EFIV (16.4ml vs 62.3ml; P=0.003). The safety profile was similar for the 2 groups. In multivariable regression analysis, low baseline NIHSS score (P<0.05), vessel recanalization (P=0.05) and CoT (P=0.03) were indipendent predictors of 3-month favorable outcome. Conclusions CoT appears more effective than ET alone for anterior circulation AIS with LVO, with similar safety profile.

scholarly journals Mediation of the Relationship Between Endovascular Therapy and Functional Outcome by Follow-up Infarct Volume in Patients With Acute Ischemic Stroke

2019 ◽  
Vol 76 (2) ◽  
pp. 194 ◽  
Author(s):  
Anna M. M. Boers ◽  
Ivo G. H. Jansen ◽  
Scott Brown ◽  
Hester F. Lingsma ◽  
Ludo F. M. Beenen ◽  
...  
Keyword(s):  
Ischemic Stroke ◽  
Acute Ischemic Stroke ◽  
Functional Outcome ◽  
Endovascular Therapy ◽  
Infarct Volume ◽  
The Relationship
Sign in / Sign up

Export Citation Format

Share Document