scholarly journals Computed tomography angiography-based deep learning method for treatment selection and infarct volume prediction in anterior cerebral circulation large vessel occlusion

2021 ◽  
Vol 10 (11) ◽  
pp. 205846012110603
Author(s):  
Lasse Hokkinen ◽  
Teemu Mäkelä ◽  
Sauli Savolainen ◽  
Marko Kangasniemi
Keyword(s):  
Neural Network ◽  
Computed Tomography ◽  
Convolutional Neural Network ◽  
Computed Tomography Angiography ◽  
Time Window ◽  
Infarct Volume ◽  
Correlation Coefficients ◽  
Late Time ◽  
Large Vessel Occlusion ◽  
Vessel Occlusion

Background Computed tomography perfusion (CTP) is the mainstay to determine possible eligibility for endovascular thrombectomy (EVT), but there is still a need for alternative methods in patient triage. Purpose To study the ability of a computed tomography angiography (CTA)-based convolutional neural network (CNN) method in predicting final infarct volume in patients with large vessel occlusion successfully treated with endovascular therapy. Materials and Methods The accuracy of the CTA source image-based CNN in final infarct volume prediction was evaluated against follow-up CT or MR imaging in 89 patients with anterior circulation ischemic stroke successfully treated with EVT as defined by Thrombolysis in Cerebral Infarction category 2b or 3 using Pearson correlation coefficients and intraclass correlation coefficients. Convolutional neural network performance was also compared to a commercially available CTP-based software (RAPID, iSchemaView). Results A correlation with final infarct volumes was found for both CNN and CTP-RAPID in patients presenting 6–24 h from symptom onset or last known well, with r = 0.67 ( p < 0.001) and r = 0.82 ( p < 0.001), respectively. Correlations with final infarct volumes in the early time window (0–6 h) were r = 0.43 ( p = 0.002) for the CNN and r = 0.58 ( p < 0.001) for CTP-RAPID. Compared to CTP-RAPID predictions, CNN estimated eligibility for thrombectomy according to ischemic core size in the late time window with a sensitivity of 0.38 and specificity of 0.89. Conclusion A CTA-based CNN method had moderate correlation with final infarct volumes in the late time window in patients successfully treated with EVT.

scholarly journals Mobile Interventional Stroke Teams Improve Outcomes in the Early Time Window for Large Vessel Occlusion Stroke

Stroke ◽  
2021 ◽  
Author(s):  
Jacob R. Morey ◽  
Xiangnan Zhang ◽  
Naoum Fares Marayati ◽  
Stavros Matsoukas ◽  
Emily Fiano ◽  
...  
Keyword(s):  
Scale Score ◽  
Time Window ◽  
Early Time ◽  
National Institutes Of Health ◽  
Large Vessel ◽  
Late Time ◽  
Modified Rankin Scale ◽  
Large Vessel Occlusion ◽  
Vessel Occlusion ◽  
Endovascular Thrombectomy

Background and Purpose: Endovascular thrombectomy for large vessel occlusion stroke is a time-sensitive intervention. The use of a Mobile Interventional Stroke Team (MIST) traveling to Thrombectomy Capable Stroke Centers to perform endovascular thrombectomy has been shown to be significantly faster with improved discharge outcomes, as compared with the drip-and-ship (DS) model. The effect of the MIST model stratified by time of presentation has yet to be studied. We hypothesize that patients who present in the early window (last known well of ≤6 hours) will have better clinical outcomes in the MIST model. Methods: The NYC MIST Trial and a prospectively collected stroke database were assessed for patients undergoing endovascular thrombectomy from January 2017 to February 2020. Patients presenting in early and late time windows were analyzed separately. The primary end point was the proportion with a good outcome (modified Rankin Scale score of 0–2) at 90 days. Secondary end points included discharge National Institutes of Health Stroke Scale and modified Rankin Scale. Results: Among 561 cases, 226 patients fit inclusion criteria and were categorized into MIST and DS cohorts. Exclusion criteria included a baseline modified Rankin Scale score of >2, inpatient status, or fluctuating exams. In the early window, 54% (40/74) had a good 90-day outcome in the MIST model, as compared with 28% (24/86) in the DS model ( P <0.01). In the late window, outcomes were similar (35% versus 41%; P =0.77). The median National Institutes of Health Stroke Scale at discharge was 5.0 and 12.0 in the early window ( P <0.01) and 5.0 and 11.0 in the late window ( P =0.11) in the MIST and DS models, respectively. The early window discharge modified Rankin Scale was significantly better in the MIST model ( P <0.01) and similar in the late window ( P =0.41). Conclusions: The MIST model in the early time window results in better 90-day outcomes compared with the DS model. This may be due to the MIST capturing high-risk fast progressors at an earlier time point. REGISTRATION: URL: https://www.clinicaltrials.gov ; Unique identifier: NCT03048292.

Abstract P514: Clinically Approximated Hypoperfused Tissue in Large Vessel Occlusion Stroke

Stroke ◽  
2021 ◽  
Vol 52 (Suppl_1) ◽  
Author(s):  
Shashvat Desai ◽  
Santiago Ortega ◽  
Sunil Sheth ◽  
Mudassir Farooqui ◽  
Victor Lopez Rivera ◽  
...  
Keyword(s):  
Conversion Factor ◽  
Time Window ◽  
Clinical Severity ◽  
Large Vessel ◽  
Late Time ◽  
Large Vessel Occlusion ◽  
Eligibility Criteria ◽  
Vessel Occlusion ◽  
Nihss Score ◽  
Derivation Cohort

Introduction: Patient selection for thrombectomy of acute ischemic stroke (AIS) caused by large vessel occlusion (LVO) in the delayed time window (>6 hours) is dependent on delineation of clinical-core mismatch or radiological target mismatch using perfusion imaging. Selection paradigms not involving advanced imaging and software processing may reduce time to treatment and broaden eligibility. We aim to develop a conversion factor to approximately determine the volume of hypoperfused tissue using the NIHSS score [CAT volume (clinically approximated tissue)] and explore its ability to identify patients eligible for thrombectomy in the late time window. Methods: We performed a retrospective analysis of anterior circulation LVO strokes at three comprehensive stroke centers. Demographic, clinical (NIHSS score, TLKW-time last known well) and imaging [computed tomography with perfusion (CTP) processed using RAPID, IschemaView] information was analyzed. A conversion factor, which is a multiple of the NIHSS score (one multiple for NIHSS score <10 and another for NIHSS score ≥10), was derived to calculate CAT volumes. Accuracy (sensitivity and specificity) of CAT-based thrombectomy eligibility criteria (similar to DEFUSE-3 criteria but using CAT volume instead of Tmax >6 seconds volume) was tested using DEFUSE-3 criteria eligibility as a gold standard. Result: Of the 309 LVO strokes [mean age of 70 ±14, 46% male, median NIHSS 16 (12-20)] included in this study, 38% of patients arrived beyond 6 hours of TLKW. Conversion factors derived (derivation cohort-center A:187) based on median (50 th percentile) values of Tmax >6s volume for NIHSS <10 subgroup was 15 and for NIHSS ≥10 subgroup was 6. Subsequently calculated CAT volume-based eligibility criteria yielded a sensitivity of 100% and specificity of 92% in detecting DEFUSE-3 eligible patients (AUC-0.92 CI-0.82-1) in the validation cohort (center B and C:122). Conclusions: Clinical severity of stroke (NIHSS score) may be used to calculate the volume of hypoperfused tissue during LVO stroke. Clinically approximated hypoperfused tissue (CAT) volumes for NIHSS score <10 (using a factor of 15) and ≥10 (using a factor of 6) subgroups can accurately identify DEFUSE-3 eligible patients.

Abstract WMP15: Automated Volumetric Assessment of Infarct Core in Non-Contrast Computed Tomography in Patients With Acute Ischemic Stroke Secondary to Large Vessel Occlusion

Stroke ◽  
2020 ◽  
Vol 51 (Suppl_1) ◽  
Author(s):  
Alvaro Garcia-Tornel ◽  
Matias Deck ◽  
Marc Ribo ◽  
David Rodriguez-Luna ◽  
Jorge Pagola ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Ischemic Stroke ◽  
Acute Ischemic Stroke ◽  
Functional Outcome ◽  
Infarct Volume ◽  
Large Vessel Occlusion ◽  
Infarct Core ◽  
Vessel Occlusion ◽  
Good Functional Outcome ◽  
Volumetric Assessment

Introduction: Perfusion imaging has emerged as an imaging tool to select patients with acute ischemic stroke (AIS) secondary to large vessel occlusion (LVO) for endovascular treatment (EVT). We aim to compare an automated method to assess the infarct ischemic core (IC) in Non-Contrast Computed Tomography (NCCT) with Computed Tomography Perfusion (CTP) imaging and its ability to predict functional outcome and final infarct volume (FIV). Methods: 494 patients with anterior circulation stroke treated with EVT were included. Volumetric assessment of IC in NCCT (eA-IC) was calculated using eASPECTS™ (Brainomix, Oxford). CTP was processed using availaible software considering CTP-IC as volume of Cerebral Blood Flow (CBF) <30% comparing with the contralateral hemisphere. FIV was calculated in patients with complete recanalization using a semiautomated method with a NCCT performed 48-72 hours after EVT. Complete recanalization was considered as modified Thrombolysis In Cerebral Ischemia (mTICI) ≥2B after EVT. Good functional outcome was defined as modified Rankin score (mRs) ≤2 at 90 days. Statistical analysis was performed to assess the correlation between EA-IC and CTP-IC and its ability to predict prognosis and FIV. Results: Median eA-IC and CTP-IC were 16 (IQR 7-31) and 8 (IQR 0-28), respectively. 419 patients (85%) achieved complete recanalization, and their median FIV was 17.5cc (IQR 5-52). Good functional outcome was achieved in 230 patients (47%). EA-IC and CTP-IC had moderate correlation between them (r=0.52, p<0.01) and similar correlation with FIV (r=0.52 and 0.51, respectively, p<0.01). Using ROC curves, both methods had similar performance in its ability to predict good functional outcome (EA-IC AUC 0.68 p<0.01, CTP-IC AUC 0.66 p<0.01). Multivariate analysis adjusted by confounding factors showed that eA-IC and CTP-IC predicted good functional outcome (for every 10cc and >40cc, OR 1.5, IC1.3-1.8, p<0.01 and OR 1.3, IC1.1-1.5, p<0.01, respectively). Conclusion: Automated volumetric assessment of infarct core in NCCT has similar performance predicting prognosis and final infarct volume than CTP. Prospective studies should evaluate a NCCT-core / vessel occlusion penumbra missmatch as an alternative method to select patients for EVT.

scholarly journals A Convolutional Neural Network for Anterior Intra-Arterial Thrombus Detection and Segmentation on Non-Contrast Computed Tomography of Patients with Acute Ischemic Stroke

2020 ◽  
Vol 10 (14) ◽  
pp. 4861
Author(s):  
Manon L. Tolhuisen ◽  
Elena Ponomareva ◽  
Anne M. M. Boers ◽  
Ivo G. H. Jansen ◽  
Miou S. Koopman ◽  
...  
Keyword(s):  
Neural Network ◽  
Computed Tomography ◽  
Ischemic Stroke ◽  
Convolutional Neural Network ◽  
Acute Ischemic Stroke ◽  
Ground Truth ◽  
Correct Identification ◽  
Anterior Circulation ◽  
Vessel Occlusion ◽  
Comprehensive Stroke Center

The aim of this study was to develop a convolutional neural network (CNN) that automatically detects and segments intra-arterial thrombi on baseline non-contrast computed tomography (NCCT) scans. We retrospectively collected computed tomography (CT)-scans of patients with an anterior circulation large vessel occlusion (LVO) from the Multicenter Randomized Clinical Trial of Endovascular Treatment for Acute Ischemic Stroke in the Netherlands trial, both for training (n = 86) and validation (n = 43). For testing we included patients with (n = 58) and without (n = 45) an LVO from our comprehensive stroke center. Ground truth was established by consensus between two experts using both CT angiography and NCCT. We evaluated the CNN for correct identification of a thrombus, its location and thrombus segmentation and compared these with the results of a neurologist in training and expert neuroradiologist. Sensitivity of the CNN thrombus detection was 0.86, vs. 0.95 and 0.79 for the neuroradiologists. Specificity was 0.65 for the network vs. 0.58 and 0.82 for the neuroradiologists. The CNN correctly identified the location of the thrombus in 79% of the cases, compared to 81% and 77% for the neuroradiologists. The sensitivity and specificity for thrombus identification and the rate for correct thrombus location assessment by the CNN were similar to those of expert neuroradiologists.

Collateral Circulation in Thrombectomy for Stroke After 6 to 24 Hours in the DAWN Trial

Stroke ◽  
2021 ◽  
Author(s):  
David S. Liebeskind ◽  
Hamidreza Saber ◽  
Bin Xiang ◽  
Ashutosh P. Jadhav ◽  
Tudor G. Jovin ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Scale Score ◽  
Collateral Circulation ◽  
Time Window ◽  
Infarct Volume ◽  
Late Time ◽  
Collateral Flow ◽  
Modified Rankin Scale ◽  
Diverse Range ◽  
Wide Range

Background and Purpose: Collaterals govern the pace and severity of cerebral ischemia, distinguishing fast or slow progressors and corresponding therapeutic opportunities. The fate of sustained collateral perfusion or collateral failure is poorly characterized. We evaluated the nature and impact of collaterals on outcomes in the late time window DAWN trial (Diffusion-Weighted Imaging or Computed Tomography Perfusion Assessment With Clinical Mismatch in the Triage of Wake-Up and Late Presenting Strokes Undergoing Neurointervention With Trevo). Methods: The DAWN Imaging Core Lab prospectively scored collateral grade on baseline computed tomography angiography (CTA; endovascular and control arms) and digital subtraction angiography (DSA; endovascular arm only), blinded to all other data. CTA collaterals were graded with the Tan scale and DSA collaterals were scored by ASITN grade (American Society of Interventional and Therapeutic Neuroradiology collateral score). Descriptive statistics characterized CTA collateral grade in all DAWN subjects and DSA collaterals in the endovascular arm. The relationship between collateral grade and day 90 outcomes were separately analyzed for each treatment arm. Results: Collateral circulation to the ischemic territory was evaluated on CTA (n=144; median 2, 0–3) and DSA (n=57; median 2, 1–4) before thrombectomy in 161 DAWN subjects (mean age 69.8±13.6 years; 55.3% women; 91 endovascular therapy, 70 control). CTA revealed a broad range of collaterals (Tan grade 3, n=64 [44%]; 2, n=45 [31%]; 1, n=31 [22%]; 0, n=4 [3%]). DSA also showed a diverse range of collateral grades (ASITN grade 4, n=4; 3, n=22; 2, n=27; 1, n=4). Across treatment arms, baseline demographics, clinical variables except atrial fibrillation (41.6% endovascular versus 25.0% controls, P =0.04), and CTA collateral grades were balanced. Differences were seen across the 3 levels of collateral flow (good, fair, poor) for baseline National Institutes of Health Stroke Scale, blood glucose <150, diabetes, previous ischemic stroke, baseline and 24-hour core infarct volume, baseline and 24-hour Alberta Stroke Program Early CT Score, dramatic infarct progression, final Thrombolysis in Cerebral Infarction 2b+, and death. Collateral flow was a significant predictor of 90-day modified Rankin Scale score of 0 to 2 in the endovascular arm, with 43.7% (31/71) of subjects with good collaterals, 30.8% (16/52) of subjects with fair collaterals, and 17.7% (6/34) of subjects with poor collaterals reaching modified Rankin Scale score of 0 to 2 at 90 days ( P =0.026). Conclusions: DAWN subjects enrolled at 6 to 24 hours after onset with limited infarct cores had a wide range of collateral grades on both CTA and DSA. Even in this late time window, better collaterals lead to slower stroke progression and better functional outcomes. REGISTRATION: URL: https://www.clinicaltrials.gov ; Unique identifier: NCT02142283.

Enhancing performance of a computed tomography perfusion software for improved prediction of final infarct volume in acute ischemic stroke patients

2021 ◽  
pp. 197140092098866
Author(s):  
Ryan A Rava ◽  
Kenneth V Snyder ◽  
Maxim Mokin ◽  
Muhammad Waqas ◽  
Alexander R Podgorsak ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Ischemic Stroke ◽  
Conservative Treatment ◽  
Acute Ischemic Stroke ◽  
Mechanical Thrombectomy ◽  
Infarct Volume ◽  
Correlation Coefficients ◽  
Intravenous Thrombolysis ◽  
Vessel Occlusion ◽  
Computed Tomography Perfusion

Computed tomography perfusion (CTP) is crucial for acute ischemic stroke (AIS) patient diagnosis. To improve infarct prediction, enhanced image processing and automated parameter selection have been implemented in Vital Images’ new CTP+ software. We compared CTP+ with its previous version, commercially available software (RAPID and Sphere), and follow-up diffusion-weighted imaging (DWI). Data from 191 AIS patients between March 2019 and January 2020 was retrospectively collected and allocated into endovascular intervention ( n = 81) and conservative treatment ( n = 110) cohorts. Intervention patients were treated for large vessel occlusion, underwent mechanical thrombectomy, and achieved successful reperfusion of thrombolysis in cerebral infarction 2b/2c/3. Conservative treatment patients suffered large or small vessel occlusion and did not receive intravenous thrombolysis or mechanical thrombectomy. Infarct and penumbra were assessed using intervention and conservative treatment patients, respectively. Infarct and penumbra volumes were segmented from CTP+ and compared with 24-h DWI along with RAPID, Sphere, and Vitrea. Mean infarct differences (95% confidence intervals) and Spearman correlation coefficients (SCCs) between DWI and each CTP software product for intervention patients are: CTP+  = (5.8 ± 5.9 ml, 0.62), RAPID = (10.0  ± 5.2 ml, 0.73), Sphere = (3.0 ± 6.0 ml, 0.56), Vitrea = (7.2 ± 4.9 ml, 0.66). For conservative treatment patients, mean infarct differences and SCCs are: CTP+ = (–8.0 ± 5.4 ml, 0.64), RAPID = (–25.6 ± 11.5 ml, 0.60), Sphere = (–25.6 ± 8.0 ml, 0.66), Vitrea = (1.3 ± 4.0 ml, 0.72). CTP+ performed similarly to RAPID and Sphere in addition to its semi-automated predecessor, Vitrea, when assessing intervention patient infarct volumes. For conservative treatment patients, CTP+ outperformed RAPID and Sphere in assessing penumbra. Semi-automated Vitrea remains the most accurate in assessing penumbra, but CTP+ provides an improved workflow from its predecessor.

scholarly journals Mechanical Thrombectomy of the Fetal Posterior Cerebral Artery

2021 ◽  
Author(s):  
Mohamad Abdalkader ◽  
Anurag Sahoo ◽  
Adam A. Dmytriw ◽  
Waleed Brinjikji ◽  
Guilherme Dabus ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Endovascular Treatment ◽  
Cerebral Artery ◽  
Computed Tomography Angiography ◽  
Mechanical Thrombectomy ◽  
Posterior Cerebral Artery ◽  
Large Vessel ◽  
Large Vessel Occlusion ◽  
Anterior Circulation ◽  
Vessel Occlusion

Abstract BACKGROUND Fetal posterior cerebral artery (FPCA) occlusion is a rare but potentially disabling cause of stroke. While endovascular treatment is established for acute large vessel occlusion stroke, FPCA occlusions were excluded from acute ischemic stroke trials. We aim to report the feasibility, safety, and outcome of mechanical thrombectomy in acute FPCA occlusions. METHODS We performed a multicenter retrospective review of consecutive patients who underwent mechanical thrombectomy of acute FPCA occlusion. Primary FPCA occlusion was defined as an occlusion that was identified on the pre‐procedure computed tomography angiography or baseline angiogram whereas a secondary FPCA occlusion was defined as an occlusion that occurred secondary to embolization to a new territory after recanalization of a different large vessel occlusion. Demographics, clinical presentation, imaging findings, endovascular treatment, and outcome were reviewed. RESULTS There were 25 patients with acute FPCA occlusion who underwent mechanical thrombectomy, distributed across 14 centers. Median National Institutes of Health Stroke Scale on presentation was 16. There were 76% (19/25) of patients who presented with primary FPCA occlusion and 24% (6/25) of patients who had a secondary FPCA occlusion. The configuration of the FPCA was full in 64% patients and partial or “fetal‐type” in 36% of patients. FPCA occlusion was missed on initial computed tomography angiography in 21% of patients with primary FPCA occlusion (4/19). The site of occlusion was posterior communicating artery in 52%, P2 segment in 40% and P3 in 8% of patients. Thrombolysis in cerebral infarction 2b/3 reperfusion was achieved in 96% of FPCA patients. There were no intraprocedural complications. At 90 days, 48% (12/25) were functionally independent as defined by modified Rankin scale≤2. CONCLUSIONS Endovascular treatment of acute FPCA occlusion is safe and technically feasible. A high index of suspicion is important to detect occlusion of the FPCA in patients presenting with anterior circulation stroke syndrome and patent anterior circulation. Novelty and significance This is the first multicenter study showing that thrombectomy of FPCA occlusion is feasible and safe.

scholarly journals Deep Learning–Derived High-Level Neuroimaging Features Predict Clinical Outcomes for Large Vessel Occlusion

Stroke ◽  
2020 ◽  
Vol 51 (5) ◽  
pp. 1484-1492 ◽  
Author(s):  
Hidehisa Nishi ◽  
Naoya Oishi ◽  
Akira Ishii ◽  
Isao Ono ◽  
Takenori Ogura ◽  
...  
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Convolutional Neural Network ◽  
Network Model ◽  
Neural Network Model ◽  
Large Vessel ◽  
Large Vessel Occlusion ◽  
Vessel Occlusion ◽  
Neuroimaging Biomarkers ◽  
High Level

Background and Purpose— For patients with large vessel occlusion, neuroimaging biomarkers that evaluate the changes in brain tissue are important for determining the indications for mechanical thrombectomy. In this study, we applied deep learning to derive imaging features from pretreatment diffusion-weighted image data and evaluated the ability of these features in predicting clinical outcomes for patients with large vessel occlusion. Methods— This multicenter retrospective study included patients with anterior circulation large vessel occlusion treated with mechanical thrombectomy between 2013 and 2018. We designed a 2-output deep learning model based on convolutional neural networks (the convolutional neural network model). This model employed encoder-decoder architecture for the ischemic lesion segmentation, which automatically extracted high-level feature maps in its middle layers, and used its information to predict the clinical outcome. Its performance was internally validated with 5-fold cross-validation, externally validated, and the results compared with those from the standard neuroimaging biomarkers Alberta Stroke Program Early CT Score and ischemic core volume. The prediction target was a good clinical outcome, defined as a modified Rankin Scale score at 90-day follow-up of 0 to 2. Results— The derivation cohort included 250 patients, and the validation cohort included 74 patients. The convolutional neural network model showed the highest area under the receiver operating characteristic curve: 0.81±0.06 compared with 0.63±0.05 and 0.64±0.05 for the Alberta Stroke Program Early CT Score and ischemic core volume models, respectively. In the external validation, the area under the curve for the convolutional neural network model was significantly superior to those for the other 2 models. Conclusions— Compared with the standard neuroimaging biomarkers, our deep learning model derived a greater amount of prognostic information from pretreatment neuroimaging data. Although a confirmatory prospective evaluation is needed, the high-level imaging features derived by deep learning may offer an effective prognostic imaging biomarker.

Sign in / Sign up

Export Citation Format

Share Document