Acute Infarct Segmentation on Diffusion-Weighted Imaging Using Deep Learning Algorithm and RAPID MRI

Optimal Cerebral Blood Flow Thresholds for Ischemic Core Estimation Using Computed Tomography Perfusion and Diffusion-Weighted Imaging

OBJECTIVE

Computed tomography perfusion (CTP) imaging is crucial in quantifying cerebral blood flow (CBF) and thereby making an endovascular treatment (EVT) after large vessel occlusion. However, CTP is prone to overestimating the ischemic core. We sought to delineate the optimal regional CBF (rCBF) thresholds of pre-EVT CTP.

METHODS

We collected acute ischemic stroke patients due to large vessel occlusion who achieved successful recanalization with baseline CTP, immediate post-EVT diffusion-weighted image (DWI) within 3 hours, and delayed post-EVT DWI between 24 and 196 hours. Core volumes estimated by CTP at various rCBF thresholds were validated against immediate and delayed DWI lesion volumes.

RESULTS

A total of 175 acute large vessel occlusion patients were included. Baseline CTP was taken in a median of 24 minutes (interquartile range [IQR] 21-31 minutes) after arrival; after the CTP, groin puncture in a median of 37 minutes (IQR 28-52 minutes), immediate post-EVT DWI scans in a median of 1.6 hours (IQR 0.8-2.1 hours), and delayed DWI scans in a median of 89 hours (IQR 69-106 hours). The correlations between the rCBF thresholds were the best at rCBF <22% for immediate DWI (0.64; 95% CI 0.55-0.73) and at rCBF <30% for delayed DWI (0.69; 95% CI 0.61-0.76). The interval between CTP and recanalization was inversely correlated with the overestimation of ischemic core volume compared with the subsequent DWI.

INTERPRETATION

Optimal rCBF thresholds for estimating ischemic core using CTP depend significantly on the timing of DWI post-EVT and CTP to recanalization delay. The optimal rCBF thresholds for ischemic core estimation may vary depending on the clinical setting. ANN NEUROL 2025;97:919-929.

Multiphase CTA vs. MRA collateral map for predicting functional outcomes after acute ischemic stroke

PURPOSE

To compare the prognostic abilities of multiphase CT angiography (mCTA) and multiphase MR angiography (MRA) collateral map in acute anterior circulation ischemic stroke.

METHODS

This secondary analysis of a prospective observational study included data from participants with acute ischemic stroke due to steno-occlusion of the internal carotid artery and/or middle cerebral artery within 8 h of symptom onset between January 1, 2016, and March 31, 2021. The intermethod agreement of the collateral scores (CSs) from mCTA and the collateral perfusion scores (CPSs) from the MRA collateral map was analyzed. Multiple logistic regression analyses were conducted to determine the prognostic value of mCTA and MRA collateral maps.

RESULTS

169 participants (106 men and 63 women, mean age 69 years ± 13) were included. The agreement between the CSs of mCTA and the CPSs of the MRA collateral map (weighted kappa = 0.44, 95% confidence interval [CI]: 0.37-0.52) of 168 participants was moderate. Younger age (Odds ratio [OR], 0.51; 95% CI, 0.34-0.76; p = 0.001), lower baseline NIHSS scores (OR, 0.89; 95% CI, 0.81-0.99; p = 0.024), CPS 4 (OR, 36.66; 95% CI, 1.79-750.29; p = 0.019) and CPS 5 (OR, 144.10; 95% CI, 1.11-18788.93; p = 0.046) on the MRA collateral map, and successful reperfusion (OR, 9.63; 95% CI, 3.00-30.94; p < 0.001) were independently associated with favorable functional outcomes.

CONCLUSIONS

Only the MRA collateral map demonstrated clinical prognostic value in acute anterior circulation ischemic stroke patients, demonstrating the superiority of the MRA collateral map over mCTA in collateral assessment.

A multicenter validation and calibration of automated software package for detecting anterior circulation large vessel occlusion on CT angiography

PURPOSE

To validate JLK-LVO, a software detecting large vessel occlusion (LVO) on computed tomography angiography (CTA), within a multicenter dataset.

METHODS

From 2021 to 2023, we enrolled patients with ischemic stroke who underwent CTA within 24-hour of onset at six university hospitals for validation and calibration datasets and at another university hospital for an independent dataset for testing model calibration. The diagnostic performance was evaluated using area under the curve (AUC), sensitivity, and specificity across the entire study population and specifically in patients with isolated middle cerebral artery (MCA)-M2 occlusion. We calibrated LVO probabilities using logistic regression and by grouping LVO probabilities based on observed frequency.

RESULTS

After excluding 168 patients, 796 remained; the mean (SD) age was 68.9 (13.7) years, and 57.7% were men. LVO was present in 193 (24.3%) of patients, and the median interval from last-known-well to CTA was 5.7 h (IQR 2.5-12.1 h). The software achieved an AUC of 0.944 (95% CI 0.926-0.960), with a sensitivity of 89.6% (84.5-93.6%) and a specificity of 90.4% (87.7-92.6%). In isolated MCA-M2 occlusion, the AUROC was 0.880 (95% CI 0.824-0.921). Due to sparse data between 20 and 60% of LVO probabilities, recategorization into unlikely (0-20% LVO scores), less likely (20-60%), possible (60-90%), and suggestive (90-100%) provided a reliable estimation of LVO compared with mathematical calibration. The category of LVO probabilities was associated with follow-up infarct volumes and functional outcome.

CONCLUSION

In this multicenter study, we proved the clinical efficacy of the software in detecting LVO on CTA.

The CT collateral map: collateral perfusion estimation and baseline lesion assessment after acute anterior circulation ischemic stroke

PURPOSE

To investigate the clinical feasibility of a CT collateral map compared with an MRA collateral map, focusing on collateral perfusion (CP) estimation and baseline lesion assessment in acute ischemic stroke (AIS).

MATERIALS AND METHODS

This retrospective analysis used selected data from a prospectively collected database. We generated CT collateral maps derived from CT perfusion, encompassing images of arterial, capillary, early venous (CMEV), late venous, and delay phases. Three raters assessed CP scores from MRA and CT collateral maps and CMEV lesion volumes. Lesion volumes of baseline diffusion-weighted imaging (bDWI) and cerebral blood flow rate (CBF) < 30% were automatically measured by the software. The agreement between MRA and CT collateral maps in CP estimation and the correlation between lesion volumes with a CBF < 30% and the CMEV for bDWI lesion volumes were analyzed.

RESULTS

One-hundred ten patients (mean age ± standard deviation, 71 ± 14; 60 women) with AIS due to steno-occlusion of the internal carotid and/or middle cerebral arteries were included. The agreement between the MRA and CT collateral maps in CP grading was excellent (weighted κ = 0.93; 95% CI, 0.90-0.97). The concordance correlation coefficients (CCCs) of the CBF < 30% and CMEV for bDWI lesion volumes were 0.76 (95% CI, 0.60-0.91) and 0.97 (0.95-0.98), respectively.

CONCLUSION

The clinical feasibility of the CT collateral map is demonstrated by its significant correlation with the MRA collateral map in CP estimation and baseline lesion assessment.

Clinical feasibility of deep learning-driven magnetic resonance angiography collateral map in acute anterior circulation ischemic stroke

To validate the clinical feasibility of deep learning-driven magnetic resonance angiography (DL-driven MRA) collateral map in acute ischemic stroke. We employed a 3D multitask regression and ordinal regression deep neural network, called as 3D-MROD-Net, to generate DL-driven MRA collateral maps. Two raters graded the collateral perfusion scores of both conventional and DL-driven MRA collateral maps and measured the grading time. They also qualitatively assessed the image quality of both collateral maps. Interrater and inter-method agreements for collateral perfusion grading between the two collateral maps were analyzed, along with a comparison of grading time and image quality. In the analysis of the 296 acute ischemic stroke patients, the inter-method agreement for collateral perfusion grading was almost perfect (κ = 0.96, 95% CI: 0.95-0.98). Compared to conventional MRA collateral maps, the time taken for collateral perfusion grading on DL-driven MRA collateral maps was shorter (P < 0.001 for rater 1 and P = 0.003 for rater 2), and the image quality of the DL-driven MRA collateral maps was superior (P < 0.001 for rater 1 and P = 0.002 for rater 2). The DL-driven MRA collateral map demonstrates clinical feasibility for collateral perfusion grading in acute ischemic stroke, with the added benefits of reduced generation and interpretation time, along with improved image quality of the MRA collateral map.

Automated Segmentation of MRI White Matter Hyperintensities in 8421 Patients with Acute Ischemic Stroke

BACKGROUND AND PURPOSE

To date, only a few small studies have attempted deep learning-based automatic segmentation of white matter hyperintensity (WMH) lesions in patients with cerebral infarction; this issue is complicated because stroke-related lesions can obscure WMH borders. We developed and validated deep learning algorithms to segment WMH lesions accurately in patients with cerebral infarction using multisite data sets involving 8421 patients with acute ischemic stroke.

MATERIALS AND METHODS

We included 8421 patients with stroke from 9 centers in Korea. 2D UNet and squeeze-and-excitation (SE)-UNet models were trained using 2408 FLAIR MRIs from 3 hospitals and validated using 6013 FLAIR MRIs from 6 hospitals. WMH segmentation performance was assessed by calculating the Dice similarity coefficient (DSC), the correlation coefficient, and the concordance correlation coefficient compared with a human-segmented criterion standard. In addition, we obtained an uncertainty index that represents overall ambiguity in the voxel classification for WMH segmentation in each patient based on the Kullback-Leibler divergence.

RESULTS

In the training data set, the mean age was 67.4 (SD, 13.0) years, and 60.4% were men. The mean (95% CI) DSCs for UNet in internal testing and external validation were, respectively, 0.659 (0.649-0.669) and 0.710 (0.707-0.714), which were slightly lower than the reliability between humans (DSC = 0.744; 95% CI, 0.738-0.751; P = .031). Compared with the UNet, the SE-UNet demonstrated better performance, achieving a mean DSC of 0.675 (95% CI, 0.666-0.685; P < .001) in the internal testing and 0.722 (95% CI, 0.719-0.726; P < .001) in the external validation; moreover, it achieved high DSC values (ranging from 0.672 to 0.744) across multiple validation data sets. We observed a significant correlation between WMH volumes that were segmented automatically and manually for the UNet (r = 0.917, P < .001), and it was even stronger for the SE-UNet (r = 0.933, P < .001). The SE-UNet also attained a high concordance correlation coefficient (ranging from 0.841 to 0.956) in the external test data sets. In addition, the uncertainty indices in most patients (86%) in the external data sets were <0.35, with an average DSC of 0.744 in these patients.

CONCLUSIONS

We developed and validated deep learning algorithms to segment WMH in patients with acute cerebral infarction using the largest-ever MRI data sets. In addition, we showed that the uncertainty index can be used to identify cases in which automatic WMH segmentation is less accurate and requires human review.

Assessment of Deep Learning-Based Triage Application for Acute Ischemic Stroke on Brain MRI in the ER

RATIONALE AND OBJECTIVES

To assess a deep learning application (DLA) for acute ischemic stroke (AIS) detection on brain magnetic resonance imaging (MRI) in the emergency room (ER) and the effect of T2-weighted imaging (T2WI) on its performance.

MATERIALS AND METHODS

We retrospectively analyzed brain MRIs taken through the ER from March to October 2021 that included diffusion-weighted imaging (DWI) and fluid-attenuated inversion recovery (FLAIR) sequences. MRIs were processed by the DLA, and sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUROC) were evaluated, with three neuroradiologists establishing the gold standard for detection performance. In addition, we examined the impact of axial T2WI, when available, on the accuracy and processing time of DLA.

RESULTS

The study included 947 individuals (mean age ± standard deviation, 64 years ± 16; 461 men, 486 women), with 239 (25%) positive for AIS. The overall performance of DLA was as follows: sensitivity, 90%; specificity, 89%; accuracy, 89%; and AUROC, 0.95. The average processing time was 24 s. In the subgroup with T2WI, T2WI did not significantly impact MRI assessments but did result in longer processing times (35 s without T2WI compared to 48 s with T2WI, p < 0.001).

CONCLUSION

The DLA successfully identified AIS in the ER setting with an average processing time of 24 s. The absence of performance acquire with axial T2WI suggests that the DLA can diagnose AIS with just axial DWI and FLAIR sequences, potentially shortening the exam duration in the ER.

Comparison of two automated CT perfusion software packages in patients with ischemic stroke presenting within 24 h of onset

Background

We compared the ischemic core and hypoperfused tissue volumes estimated by RAPID and JLK-CTP, a newly developed automated computed tomography perfusion (CTP) analysis package. We also assessed agreement between ischemic core volumes by two software packages against early follow-up infarct volumes on diffusion-weighted images (DWI).

Methods

This retrospective study analyzed 327 patients admitted to a single stroke center in Korea from January 2021 to May 2023, who underwent CTP scans within 24 h of onset. The concordance correlation coefficient (ρ) and Bland-Altman plots were utilized to compare the volumes of ischemic core and hypoperfused tissue volumes between the software packages. Agreement with early (within 3 h from CTP) follow-up infarct volumes on diffusion-weighted imaging (n = 217) was also evaluated.

Results

The mean age was 70.7 ± 13.0 and 137 (41.9%) were female. Ischemic core volumes by JLK-CTP and RAPID at the threshold of relative cerebral blood flow (rCBF) < 30% showed excellent agreement (ρ = 0.958 [95% CI, 0.949 to 0.966]). Excellent agreement was also observed for time to a maximum of the residue function (T max) > 6 s between JLK-CTP and RAPID (ρ = 0.835 [95% CI, 0.806 to 0.863]). Although early follow-up infarct volume showed substantial agreement in both packages (JLK-CTP, ρ = 0.751 and RAPID, ρ = 0.632), ischemic core volumes at the threshold of rCBF <30% tended to overestimate ischemic core volumes.

Conclusion

JLK-CTP and RAPID demonstrated remarkable concordance in estimating the volumes of the ischemic core and hypoperfused area based on CTP within 24 h from onset.

Deep Learning-Based Automatic Classification of Ischemic Stroke Subtype Using Diffusion-Weighted Images

BACKGROUND AND PURPOSE

Accurate classification of ischemic stroke subtype is important for effective secondary prevention of stroke. We used diffusion-weighted image (DWI) and atrial fibrillation (AF) data to train a deep learning algorithm to classify stroke subtype.

METHODS

Model development was done in 2,988 patients with ischemic stroke from three centers by using U-net for infarct segmentation and EfficientNetV2 for subtype classification. Experienced neurologists (n=5) determined subtypes for external test datasets, while establishing a consensus for clinical trial datasets. Automatically segmented infarcts were fed into the model (DWI-only algorithm). Subsequently, another model was trained, with AF included as a categorical variable (DWI+AF algorithm). These models were tested: (1) internally against the opinion of the labeling experts, (2) against fresh external DWI data, and (3) against clinical trial dataset.

RESULTS

In the training-and-validation datasets, the mean (±standard deviation) age was 68.0±12.5 (61.1% male). In internal testing, compared with the experts, the DWI-only and the DWI+AF algorithms respectively achieved moderate (65.3%) and near-strong (79.1%) agreement. In external testing, both algorithms again showed good agreements (59.3%-60.7% and 73.7%-74.0%, respectively). In the clinical trial dataset, compared with the expert consensus, percentage agreements and Cohen's kappa were respectively 58.1% and 0.34 for the DWI-only vs. 72.9% and 0.57 for the DWI+AF algorithms. The corresponding values between experts were comparable (76.0% and 0.61) to the DWI+AF algorithm.

CONCLUSION

Our model trained on a large dataset of DWI (both with or without AF information) was able to classify ischemic stroke subtypes comparable to a consensus of stroke experts.