Dataset Information

Predicting final ischemic stroke lesions from initial diffusion-weighted images using a deep neural network.

ABSTRACT:

Background

For prognosis of stroke, measurement of the diffusion-perfusion mismatch is a common practice for estimating tissue at risk of infarction in the absence of timely reperfusion. However, perfusion-weighted imaging (PWI) adds time and expense to the acute stroke imaging workup. We explored whether a deep convolutional neural network (DCNN) model trained with diffusion-weighted imaging obtained at admission could predict final infarct volume and location in acute stroke patients.

Methods

In 445 patients, we trained and validated an attention-gated (AG) DCNN to predict final infarcts as delineated on follow-up studies obtained 3 to 7 days after stroke. The input channels consisted of MR diffusion-weighted imaging (DWI), apparent diffusion coefficients (ADC) maps, and thresholded ADC maps with values less than 620 × 10^-6 mm²/s, while the output was a voxel-by-voxel probability map of tissue infarction. We evaluated performance of the model using the area under the receiver-operator characteristic curve (AUC), the Dice similarity coefficient (DSC), absolute lesion volume error, and the concordance correlation coefficient (ρ_c) of the predicted and true infarct volumes.

Results

The model obtained a median AUC of 0.91 (IQR: 0.84-0.96). After thresholding at an infarction probability of 0.5, the median sensitivity and specificity were 0.60 (IQR: 0.16-0.84) and 0.97 (IQR: 0.93-0.99), respectively, while the median DSC and absolute volume error were 0.50 (IQR: 0.17-0.66) and 27 ml (IQR: 7-60 ml), respectively. The model's predicted lesion volumes showed high correlation with ground truth volumes (ρ_c = 0.73, p < 0.01).

Conclusion

An AG-DCNN using diffusion information alone upon admission was able to predict infarct volumes at 3-7 days after stroke onset with comparable accuracy to models that consider both DWI and PWI. This may enable treatment decisions to be made with shorter stroke imaging protocols.

SUBMITTER: Nazari-Farsani S

PROVIDER: S-EPMC9727698 | biostudies-literature | 2022 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Predicting final ischemic stroke lesions from initial diffusion-weighted images using a deep neural network.

Nazari-Farsani Sanaz S Yu Yannan Y Duarte Armindo Rui R Lansberg Maarten M Liebeskind David S DS Albers Gregory G Christensen Soren S Levin Craig S CS Zaharchuk Greg G

NeuroImage. Clinical 20221201

<h4>Background</h4>For prognosis of stroke, measurement of the diffusion-perfusion mismatch is a common practice for estimating tissue at risk of infarction in the absence of timely reperfusion. However, perfusion-weighted imaging (PWI) adds time and expense to the acute stroke imaging workup. We explored whether a deep convolutional neural network (DCNN) model trained with diffusion-weighted imaging obtained at admission could predict final infarct volume and location in acute stroke patients.< ...[more]

PMID: 36481696

Similar Datasets

Project description:ImportancePredicting infarct size and location is important for decision-making and prognosis in patients with acute stroke.ObjectivesTo determine whether a deep learning model can predict final infarct lesions using magnetic resonance images (MRIs) acquired at initial presentation (baseline) and to compare the model with current clinical prediction methods.Design, setting, and participantsIn this multicenter prognostic study, a specific type of neural network for image segmentation (U-net) was trained, validated, and tested using patients from the Imaging Collaterals in Acute Stroke (iCAS) study from April 14, 2014, to April 15, 2018, and the Diffusion Weighted Imaging Evaluation for Understanding Stroke Evolution Study-2 (DEFUSE-2) study from July 14, 2008, to September 17, 2011 (reported in October 2012). Patients underwent baseline perfusion-weighted and diffusion-weighted imaging and MRI at 3 to 7 days after baseline. Patients were grouped into unknown, minimal, partial, and major reperfusion status based on 24-hour imaging results. Baseline images acquired at presentation were inputs, and the final true infarct lesion at 3 to 7 days was considered the ground truth for the model. The model calculated the probability of infarction for every voxel, which can be thresholded to produce a prediction. Data were analyzed from July 1, 2018, to March 7, 2019.Main outcomes and measuresArea under the curve, Dice score coefficient (DSC) (a metric from 0-1 indicating the extent of overlap between the prediction and the ground truth; a DSC of ≥0.5 represents significant overlap), and volume error. Current clinical methods were compared with model performance in subgroups of patients with minimal or major reperfusion.ResultsAmong the 182 patients included in the model (97 women [53.3%]; mean [SD] age, 65 [16] years), the deep learning model achieved a median area under the curve of 0.92 (interquartile range [IQR], 0.87-0.96), DSC of 0.53 (IQR, 0.31-0.68), and volume error of 9 (IQR, -14 to 29) mL. In subgroups with minimal (DSC, 0.58 [IQR, 0.31-0.67] vs 0.55 [IQR, 0.40-0.65]; P = .37) or major (DSC, 0.48 [IQR, 0.29-0.65] vs 0.45 [IQR, 0.15-0.54]; P = .002) reperfusion for which comparison with existing clinical methods was possible, the deep learning model had comparable or better performance.Conclusions and relevanceThe deep learning model appears to have successfully predicted infarct lesions from baseline imaging without reperfusion information and achieved comparable performance to existing clinical methods. Predicting the subacute infarct lesion may help clinicians prepare for decompression treatment and aid in patient selection for neuroprotective clinical trials.

Project description:PurposeTo investigate radiomics based on DWI (diffusion-weighted imaging) for predicting 1-year ischemic stroke recurrence.MethodsA total of 1,580 ischemic stroke patients were enrolled in this retrospective study conducted from January 2018 to April 2021. Demographic and clinical characteristics were compared between recurrence and non-recurrence groups. On DWI, lesions were segmented using a 2D U-Net automatic segmentation network. Further, radiomics feature extraction was done using the segmented mask matrix on DWI and the corresponding ADC map. Additionally, radiomics features were extracted. The study participants were divided into a training cohort (n = 157, 57 recurrence patients, and 100 non-recurrence patients) and a test cohort (n = 846, 28 recurrence patients, 818 non-recurrence patients). A sparse representation feature selection model was performed to select features. Further classification was accomplished using a recurrent neural network (RNN). The area under the receiver operating characteristic curve values was obtained for model performance.ResultsA total of 1,003 ischemic stroke patients (682 men and 321 women; mean age: 65.90 ± 12.44 years) were included in the final analysis. About 85 patients (8.5%) recurred in 1 year, and patients in the recurrence group were older than the non-recurrence group (P = 0.003). The stroke subtype was significantly different between recurrence and non-recurrence groups, and cardioembolic stroke (11.3%) and large artery atherosclerosis patients (10.3%) showed a higher recurrence percentage (P = 0.005). Secondary prevention after discharge (statins, antiplatelets, and anticoagulants) was found significantly different between the two groups (P = 0.004). The area under the curve (AUC) of clinical-based model and radiomics-based model were 0.675 (95% CI: 0.643-0.707) and 0.779 (95% CI: 0.750-0.807), respectively. With an AUC of 0.847 (95% CI: 0.821-0.870), the model that combined clinical and radiomic characteristics performed better.ConclusionDWI-based radiomics could help to predict 1-year ischemic stroke recurrence.

Project description:PurposeTo compare the segmentation and detection performance of a deep learning model trained on a database of human-labeled clinical stroke lesions on diffusion-weighted (DW) images to a model trained on the same database enhanced with synthetic stroke lesions.Materials and methodsIn this institutional review board-approved study, a stroke database of 962 cases (mean patient age ± standard deviation, 65 years ± 17; 255 male patients; 449 scans with DW positive stroke lesions) and a normal database of 2027 patients (mean age, 38 years ± 24; 1088 female patients) were used. Brain volumes with synthetic stroke lesions on DW images were produced by warping the relative signal increase of real strokes to normal brain volumes. A generic three-dimensional (3D) U-Net was trained on four different databases to generate four different models: (a) 375 neuroradiologist-labeled clinical DW positive stroke cases (CDB); (b) 2000 synthetic cases (S2DB); (c) CDB plus 2000 synthetic cases (CS2DB); and (d) CDB plus 40 000 synthetic cases (CS40DB). The models were tested on 20% (n = 192) of the cases of the stroke database, which were excluded from the training set. Segmentation accuracy was characterized using Dice score and lesion volume of the stroke segmentation, and statistical significance was tested using a paired two-tailed Student t test. Detection sensitivity and specificity were compared with labeling done by three neuroradiologists.ResultsThe performance of the 3D U-Net model trained on the CS40DB (mean Dice score, 0.72) was better than models trained on the CS2DB (Dice score, 0.70; P < .001) or the CDB (Dice score, 0.65; P < .001). The deep learning model (CS40DB) was also more sensitive (91% [95% confidence interval {CI}: 89%, 93%]) than each of the three human readers (human reader 3, 84% [95% CI: 81%, 87%]; human reader 1, 78% [95% CI: 75%, 81%]; human reader 2, 79% [95% CI: 76%, 82%]), but was less specific (75% [95% CI: 72%, 78%]) than each of the three human readers (human reader 3, 96% [95% CI: 94%, 98%]; human reader 1, 92% [95% CI: 90%, 94%]; human reader 2, 89% [95% CI: 86%, 91%]).ConclusionDeep learning training for segmentation and detection of stroke lesions on DW images was significantly improved by enhancing the training set with synthetic lesions.Supplemental material is available for this article.© RSNA, 2020.

Project description:Background and purposeLesions on diffusion-weighted imaging (DWI) occasionally appear on follow-up magnetic resonance imaging (MRI) among initially DWI-negative but clinically suspicious stroke patients. We established the prevalence of positive conversion in DWI-negative stroke and determined the clinical factors associated with it.MethodsThis retrospective, observational, single-center study included 5,271 patients hospitalized due to stroke/transient ischemic attack (TIA) in a single university hospital during 2010 to 2017. Patients without initial DWI lesions underwent follow-up DWI imaging as a routine practice. Adjusted hazard ratios (aHRs) for recurrent stroke risk according to positive conversion were determined using Cox proportional hazard regression. Adjusted odds ratios (aORs) and 95% confidence intervals (CIs) for positive conversion among initially DWI-negative patients were estimated.ResultsIn total, 694 (13.2%) patients (mean±standard deviation age, 62.9±13.7 years; male, 404 [58.2%]) were initially DWI-negative. Among them, 22.5% had positive-conversion on follow-up DWI. Positive conversion was associated with a higher risk of recurrent stroke (aHR, 3.12; 95% CI, 1.56 to 6.26). Early neurologic deterioration (aOR, 15.1; 95% CI, 5.71 to 47.66), atrial fibrillation (aOR, 6.17; 95% CI, 3.23 to 12.01), smoking (aOR, 3.76; 95% CI, 2.19 to 6.63), pre-stroke dependency (aOR, 1.62; 95% CI, 1.15 to 2.27), objective hemiparesis (aOR, 4.39; 95% CI, 1.90 to 10.32), longer symptom duration (aOR, 2.17; 95% CI, 1.57 to 3.08), high cholesterol (aOR, 4.70; 95% CI, 1.78 to 12.77), National Institutes of Health Stroke Scale score (aOR, 1.44; 95% CI, 1.08 to 1.91), and high systolic blood pressure (aOR, 1.01; 95% CI, 1.00 to 1.02) were associated with a higher incidence of lesions with delayed appearance. Regarding the location of lesions on follow-up DWI, 34.6% and 21.2% were in the cortex and brainstem, respectively.ConclusionsIn DWI-negative stroke/TIA, positive conversion is associated with a higher risk of recurrent stroke. DWI-negative stroke with factors related to positive conversion may require follow-up MRI for a definitive diagnosis.

Project description:Magnetic resonance (MR) imaging (MRI) is commonly used to diagnose, assess and monitor stroke. Accurate and timely segmentation of stroke lesions provides the anatomico-structural information that can aid physicians in predicting prognosis, as well as in decision making and triaging for various rehabilitation strategies. To segment stroke lesions, MR protocols, including diffusion-weighted imaging (DWI) and T2-weighted fluid attenuated inversion recovery (FLAIR) are often utilized. These imaging sequences are usually acquired with different spatial resolutions due to time constraints. Within the same image, voxels may be anisotropic, with reduced resolution along slice direction for diffusion scans in particular. In this study, we evaluate the ability of 2D and 3D U-Net Convolutional Neural Network (CNN) architectures to segment ischemic stroke lesions using single contrast (DWI) and dual contrast images (T2w FLAIR and DWI). The predicted segmentations correlate with post-stroke motor outcome measured by the National Institutes of Health Stroke Scale (NIHSS) and Fugl-Meyer Upper Extremity (FM-UE) index based on the lesion loads overlapping the corticospinal tracts (CST), which is a neural substrate for motor movement and function. Although the four methods performed similarly, the 2D multimodal U-Net achieved the best results with a mean Dice of 0.737 (95% CI: 0.705, 0.769) and a relatively high correlation between the weighted lesion load and the NIHSS scores (both at baseline and at 90 days). A monotonically constrained quintic polynomial regression yielded R2 = 0.784 and 0.875 for weighted lesion load versus baseline and 90-Days NIHSS respectively, and better corrected Akaike information criterion (AICc) scores than those of the linear regression. In addition, using the quintic polynomial regression model to regress the weighted lesion load to the 90-Days FM-UE score results in an R2 of 0.570 with a better AICc score than that of the linear regression. Our results suggest that the multi-contrast information enhanced the accuracy of the segmentation and the prediction accuracy for upper extremity motor outcomes. Expanding the training dataset to include different types of stroke lesions and more data points will help add a temporal longitudinal aspect and increase the accuracy. Furthermore, adding patient-specific data may improve the inference about the relationship between imaging metrics and functional outcomes.

Dataset Information

Predicting final ischemic stroke lesions from initial diffusion-weighted images using a deep neural network.

Background

Methods

Results

Conclusion

Publications

Predicting final ischemic stroke lesions from initial diffusion-weighted images using a deep neural network.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets