Predict Tissue Outcome Research Articles

Annotation-free prediction of treatment-specific tissue outcome from 4D CT perfusion imaging in acute ischemic stroke

Acute ischemic stroke is a critical health condition that requires timely intervention. Following admission, clinicians typically use perfusion imaging to facilitate treatment decision-making. While deep learning models leveraging perfusion data have demonstrated the ability to predict post-treatment tissue infarction for individual patients, predictions are often represented as binary or probabilistic masks that are not straightforward to interpret or easy to obtain. Moreover, these models typically rely on large amounts of subjectively segmented data and non-standard perfusion analysis techniques. To address these challenges, we propose a novel deep learning approach that directly predicts follow-up computed tomography images from full spatio-temporal 4D perfusion scans through a temporal compression. The results show that this method leads to realistic follow-up image predictions containing the infarcted tissue outcomes. The proposed compression method achieves comparable prediction results to using perfusion maps as inputs but without the need for perfusion analysis or arterial input function selection. Additionally, separate models trained on 45 patients treated with thrombolysis and 102 treated with thrombectomy showed that each model correctly captured the different patient-specific treatment effects as shown by image difference maps. The findings of this work clearly highlight the potential of our method to provide interpretable stroke treatment decision support without requiring manual annotations.

Prediction of tissue outcome in acute ischemic stroke based on single-phase CT angiography at admission.

In acute ischemic stroke, prediction of the tissue outcome after reperfusion can be used to identify patients that might benefit from mechanical thrombectomy (MT). The aim of this work was to develop a deep learning model that can predict the follow-up infarct location and extent exclusively based on acute single-phase computed tomography angiography (CTA) datasets. In comparison to CT perfusion (CTP), CTA imaging is more widely available, less prone to artifacts, and the established standard of care in acute stroke imaging protocols. Furthermore, recent RCTs have shown that also patients with large established infarctions benefit from MT, which might not have been selected for MT based on CTP core/penumbra mismatch analysis. All patients with acute large vessel occlusion of the anterior circulation treated at our institution between 12/2015 and 12/2020 were screened (N = 404) and 238 patients undergoing MT with successful reperfusion were included for final analysis. Ground truth infarct lesions were segmented on 24 h follow-up CT scans. Pre-processed CTA images were used as input for a U-Net-based convolutional neural network trained for lesion prediction, enhanced with a spatial and channel-wise squeeze-and-excitation block. Post-processing was applied to remove small predicted lesion components. The model was evaluated using a 5-fold cross-validation and a separate test set with Dice similarity coefficient (DSC) as the primary metric and average volume error as the secondary metric. The mean ± standard deviation test set DSC over all folds after post-processing was 0.35 ± 0.2 and the mean test set average volume error was 11.5 mL. The performance was relatively uniform across models with the best model according to the DSC achieved a score of 0.37 ± 0.2 after post-processing and the best model in terms of average volume error yielded 3.9 mL. 24 h follow-up infarct prediction using acute CTA imaging exclusively is feasible with DSC measures comparable to results of CTP-based algorithms reported in other studies. The proposed method might pave the way to a wider acceptance, feasibility, and applicability of follow-up infarct prediction based on artificial intelligence.

Providing clinical context to the spatio-temporal analysis of 4D CT perfusion to predict acute ischemic stroke lesion outcomes

Acute ischemic stroke is a leading cause of mortality and morbidity worldwide. Timely identification of the extent of a stroke is crucial for effective treatment, whereas spatio-temporal (4D) Computed Tomography Perfusion (CTP) imaging is playing a critical role in this process. Recently, the first deep learning-based methods that leverage the full spatio-temporal nature of perfusion imaging for predicting stroke lesion outcomes have been proposed. However, clinical information is typically not integrated into the learning process, which may be helpful to improve the tissue outcome prediction given the known influence of various factors (i.e., physiological, demographic, and treatment factors) on lesion growth. Cross-attention, a multimodal fusion strategy, has been successfully used to combine information from multiple sources, but it has yet to be applied to stroke lesion outcome prediction. Therefore, this work aimed to develop and evaluate a novel multimodal and spatio-temporal deep learning model that utilizes cross-attention to combine information from 4D CTP and clinical metadata simultaneously to predict stroke lesion outcomes. The proposed model was evaluated using a dataset of 70 acute ischemic stroke patients, demonstrating significantly improved volume estimates (mean error = 19 ml) compared to a baseline unimodal approach (mean error = 35 ml, p< 0.05). The proposed model allows generating attention maps and counterfactual outcome scenarios to investigate the relevance of clinical variables in predicting stroke lesion outcomes at a patient level, helping to provide a better understanding of the model’s decision-making process.

Artificial intelligence for treatment planning and soft tissue outcome prediction of orthognathic treatment: A systematic review.

The accuracy of artificial intelligence (AI) in treatment planning and outcome prediction in orthognathic treatment (OGT) has not been systematically reviewed. To determine the accuracy of AI in treatment planning and soft tissue outcome prediction in OGT. Systematic review. Unrestricted search of indexed databases and reference lists of included studies. Clinical studies that addressed the focused question 'Is AI useful for treatment planning and soft tissue outcome prediction in OGT?' were included. Study screening, selection and data extraction were performed independently by two authors. The risk of bias (RoB) was assessed using the Cochrane Collaboration's RoB and ROBINS-I tools for randomised and non-randomised clinical studies, respectively. Eight clinical studies (seven retrospective cohort studies and one randomised controlled study) were included. Four studies assessed the role of AI for treatment decision making; and four studies assessed the accuracy of AI in soft tissue outcome prediction after OGT. In four studies, the level of agreement between AI and non-AI decision making was found to be clinically acceptable (at least 90%). In four studies, it was shown that AI can be used for soft tissue outcome prediction after OGT; however, predictions were not clinically acceptable for the lip and chin areas. All studies had a low to moderate RoB. Due to high methodological inconsistencies among the included studies, it was not possible to conduct a meta-analysis and reporting biases assessment. AI can be a useful aid to traditional treatment planning by facilitating clinical treatment decision making and providing a visualisation tool for soft tissue outcome prediction in OGT. PROSPERO CRD42022366864.

Tissue Outcome Prediction in Patients with Proximal Vessel Occlusion and Mechanical Thrombectomy Using Logistic Models

Perfusion CT is established to aid selection of patients with proximal intracranial vessel occlusion for thrombectomy in the extended time window. Selection is mostly based on simple thresholding of perfusion parameter maps, which, however, does not exploit the full information hidden in the high-dimensional perfusion data. We implemented a multiparametric mass-univariate logistic model to predict tissue outcome based on data from 405 stroke patients with acute proximal vessel occlusion in the anterior circulation who underwent mechanical thrombectomy. Input parameters were acute multimodal CT imaging (perfusion, angiography, and non-contrast) as well as basic demographic and clinical parameters. The model was trained with the knowledge of recanalization status and final infarct localization. We found that perfusion parameter maps (CBF, CBV, and Tmax) were sufficient for tissue outcome prediction. Compared with single-parameter thresholding-based models, our logistic model had comparable volumetric accuracy, but was superior with respect to topographical accuracy (AUC of receiver operating characteristic). We also found higher spatial accuracy (Dice index) in an independent internal but not external cross-validation. Our results highlight the value of perfusion data compared with non-contrast CT, CT angiography and clinical information for tissue outcome-prediction. Multiparametric logistic prediction has high potential to outperform the single-parameter thresholding-based approach. In the future, the combination of tissue and functional outcome prediction might provide an individual biomarker for the benefit from mechanical thrombectomy in acute stroke care.

Predicting the tissue outcome of acute ischemic stroke from acute 4D computed tomography perfusion imaging using temporal features and deep learning.

Predicting follow-up lesions from baseline CT perfusion (CTP) datasets in acute ischemic stroke patients is important for clinical decision making. Deep convolutional networks (DCNs) are assumed to be the current state-of-the-art for this task. However, many DCN classifiers have not been validated against the methods currently used in research (random decision forests, RDF) and clinical routine (Tmax thresholding). Specialized DCNs have even been designed to extract complex temporal features directly from spatiotemporal CTP data instead of using standard perfusion parameter maps. However, the benefits of applying deep learning to source or deconvolved CTP data compared to perfusion parameter maps have not been formally investigated so far. In this work, a modular UNet-based DCN is proposed that separates temporal feature extraction from tissue outcome prediction, allowing for both model validation using perfusion parameter maps as well as end-to-end learning from spatiotemporal CTP data. 145 retrospective datasets comprising baseline CTP imaging, perfusion parameter maps, and follow-up non-contrast CT with manual lesion segmentations were assembled from acute ischemic stroke patients treated with intravenous thrombolysis alone (IV; n = 43) or intra-arterial mechanical thrombectomy (IA; n = 102) with or without combined IV. Using the perfusion parameter maps as input, the proposed DCN (mean Dice: 0.287) outperformed the RDF (0.262) and simple Tmax-thresholding (0.249). The performance of the proposed DCN was approximately equal using features optimized from the deconvolved residual curves (0.286) compared to perfusion parameter maps (0.287), while using features optimized from the source concentration-time curves (0.296) provided the best tissue outcome predictions.

Predicting treatment-specific lesion outcomes in acute ischemic stroke from 4D CT perfusion imaging using spatio-temporal convolutional neural networks.

For the diagnosis and precise treatment of acute ischemic stroke, predicting the final location and volume of lesions is of great clinical interest. Current deep learning-based prediction methods mainly use perfusion parameter maps, which can be calculated from spatio-temporal (4D) CT perfusion (CTP) imaging data, to estimate the tissue outcome of an acute ischemic stroke. However, this calculation relies on a deconvolution operation, an ill-posed problem requiring strong regularization and definition of an arterial input function. Thus, improved predictions might be achievable if the deep learning models were applied directly to acute 4D CTP data rather than perfusion maps. In this work, a novel deep spatio-temporal convolutional neural network is proposed for predicting treatment-dependent stroke lesion outcomes by making full use of raw 4D CTP data. By merging a U-Net-like architecture with temporal convolutional networks, we efficiently process the spatio-temporal information available in CTP datasets to make a tissue outcome prediction. The proposed method was evaluated on 147 patients using a 10-fold cross validation, which demonstrated that the proposed 3D+time model (mean Dice=0.45) significantly outperforms both a 2D+time variant of our approach (mean Dice=0.43) and a state-of-the-art method that uses perfusion maps (mean Dice=0.38). These results show that 4D CTP datasets include more predictive information than perfusion parameter maps, and that the proposed method is an efficient approach to make use of this complex data.

U-net Models Based on Computed Tomography Perfusion Predict Tissue Outcome in Patients with Different Reperfusion Patterns.

Evaluation of cerebral perfusion is important for treatment selection in patients with acute large vessel occlusion (LVO). To assess ischemic core and tissue at risk more accurately, we developed a deep learning model named U-net using computed tomography perfusion (CTP) images. A total of 110 acute ischemic stroke patients undergoing endovascular treatment with major reperfusion (≥ 80%) or minimal reperfusion (≤ 20%) were included. Using baseline CTP, we developed two U-net models: one model in major reperfusion group to identify infarct core; the other in minimal reperfusion group to identify tissue at risk. The performance of fixed-thresholding methods was compared with that of U-net models. In the major reperfusion group, the model estimated infarct core with a Dice score coefficient (DSC) of 0.61 and an area under the curve (AUC) of 0.92, while fixed-thresholding methods had a DSC of 0.52. In the minimal reperfusion group, the model estimated tissue at risk with a DSC of 0.67 and an AUC of 0.93, while fixed-thresholding methods had a DSC of 0.51. In both groups, excellent volumetric consistency (intraclass correlation coefficient was 0.951 in major reperfusion and 0.746 in minimal reperfusion) was achieved between the estimated lesion and the actual lesion volume. Thus, in patients with anterior LVO, the CTP-based U-net models were able to identify infarct core and tissue at risk on baseline CTP superior to fixed-thresholding methods, providing individualized prediction of final lesion in patients with different reperfusion patterns.

Tissue outcome prediction in hyperacute ischemic stroke: Comparison of machine learning models

Machine Learning (ML) has been proposed for tissue fate prediction after acute ischemic stroke (AIS), with the aim to help treatment decision and patient management. We compared three different ML models to the clinical method based on diffusion-perfusion thresholding for the voxel-based prediction of final infarct, using a large MRI dataset obtained in a cohort of AIS patients prior to recanalization treatment. Baseline MRI (MRI0), including diffusion-weighted sequence (DWI) and Tmax maps from perfusion-weighted sequence, and 24-hr follow-up MRI (MRI24h) were retrospectively collected in consecutive 394 patients AIS patients (median age = 70 years; final infarct volume = 28mL). Manually segmented DWI24h lesion was considered the final infarct. Gradient Boosting, Random Forests and U-Net were trained using DWI, apparent diffusion coefficient (ADC) and Tmax maps on MRI0 as inputs to predict final infarct. Tissue outcome predictions were compared to final infarct using Dice score. Gradient Boosting had significantly better predictive performance (median [IQR] Dice Score as for median age, maybe you can replace the comma with an equal sign for consistency 0.53 [0.29–0.68]) than U-Net (0.48 [0.18–0.68]), Random Forests (0.51 [0.27–0.66]), and clinical thresholding method (0.45 [0.25–0.62]) (P < 0.001). In this benchmark of ML models for tissue outcome prediction in AIS, Gradient Boosting outperformed other ML models and clinical thresholding method and is thus promising for future decision-making.

Time dependence in diffusion MRI predicts tissue outcome in ischemic stroke patients.

Purpose:Reperfusion therapy enables effective treatment of ischemic stroke presenting within 4–6 hours. However, tissue progression from ischemia to infarction is variable, and some patients benefit from treatment up until 24 hours. Improved imaging techniques are needed to identify these patients. Here, it was hypothesized that time dependence in diffusion MRI may predict tissue outcome in ischemic stroke.Methods:Diffusion MRI data were acquired with multiple diffusion times in five non-reperfused patients at 2, 9, and 100 days after stroke onset. Maps of “rate of kurtosis change” (k), mean kurtosis, ADC, and fractional anisotropy were derived. The ADC maps defined lesions, normal-appearing tissue, and the lesion tissue that would either be infarcted or remain viable by day 100. Diffusion parameters were compared (1) between lesions and normal-appearing tissue, and (2) between lesion tissue that would be infarcted or remain viable.Results:Positive values of k were observed within stroke lesions on day 2 (P = .001) and on day 9 (P = .023), indicating diffusional exchange. On day 100, high ADC values indicated infarction of 50 ± 20% of the lesion volumes. Tissue infarction was predicted by high k values both on day 2 (P = .026) and on day 9 (P = .046), by low mean kurtosis values on day 2 (P = .043), and by low fractional anisotropy values on day 9 (P = .029), but not by low ADC values.Conclusions:Diffusion time dependence predicted tissue outcome in ischemic stroke more accurately than the ADC, and may be useful for predicting reperfusion benefit.

Localized prediction of tissue outcome in acute ischemic stroke patients using diffusion- and perfusion-weighted MRI datasets.

BackgroundAn accurate prediction of tissue outcome in acute ischemic stroke patients is of high interest for treatment decision making. To date, various machine learning models have been proposed that combine multi-parametric imaging data for this purpose. However, most of these machine learning models were trained using voxel information extracted from the whole brain, without taking differences in susceptibility to ischemia into account that exist between brain regions. The aim of this study was to develop and evaluate a local tissue outcome prediction approach, which makes predictions using locally trained machine learning models and thus accounts for regional differences.Material and methodsMulti-parametric MRI data from 99 acute ischemic stroke patients were used for the development and evaluation of the local tissue outcome prediction approach. Diffusion (ADC) and perfusion parameter maps (CBF, CBV, MTT, Tmax) and corresponding follow-up lesion masks for each patient were registered to the MNI brain atlas. Logistic regression (LR) and random forest (RF) models were trained employing a local approach, which makes predictions using models individually trained for each specific voxel position using the corresponding local data. A global approach, which uses a single model trained using all voxels of the brain, was used for comparison. Tissue outcome predictions resulting from the global and local RF and LR models, as well as a combined (hybrid) approach were quantitatively evaluated and compared using the area under the receiver operating characteristic curve (ROC AUC), the Dice coefficient, and the sensitivity and specificity metrics.ResultsStatistical analysis revealed the highest ROC AUC and Dice values for the hybrid approach. With 0.872 (ROC AUC; LR) and 0.353 (Dice; RF), these values were significantly higher (p < 0.01) than the values of the two other approaches. In addition, the local approach achieved the highest sensitivity of 0.448 (LR). Overall, the hybrid approach was only outperformed in sensitivity (LR) by the local approach and in specificity by both other approaches. However, in these cases the effect sizes were comparatively small.ConclusionThe results of this study suggest that using locally trained machine learning models can lead to better lesion outcome prediction results compared to a single global machine learning model trained using all voxel information independent of the location in the brain.

Apparent diffusion coefficients for predicting primary cholesteatoma risk of recurrence after surgical clearance

PurposeBeside the well-known accuracy of non-EPI DWI techniques and relative ADC maps in detecting cholesteatomatous tissue, ADC can also represent a useful tool for stratifying cholesteatoma risk of recurrence. Aim of this study is to test the role of ADC in determining risk of recurrence for primary middle ear cholesteatoma, proposing stratification based on pre-operative mean (mADC) and normalized (nADC) ADC values. MethodsIn this prospective study, 60 patients with primary unilateral middle ear cholesteatoma underwent a three-years-long follow-up to assess the presence of recurrent disease after macroscopically complete excisional surgery. Baseline MRI examination mADC and nADC values in the group with early evidence of recurrent cholesteatoma were compared to the group with no evidence of recurrence by using T statistics. ResultsADC values on pre-operative MRI examination were lower in cholesteatomas with early evidence of recurrence, and statistical significance was slightly higher for nADC compared to mADC measurements. We also determined a cut-off between the two groups, proposing stratification in high-risk of recurrence cholesteatomas (mADC≤ 1000 or nADC< 1.3) and low-risk cholesteatomas (mADC>1000 or nADC≥1.3). ConclusionsADC values resulted discriminating in identifying cholesteatomas with higher risk of early recurrence, both for mean and normalized ADC, with optimized tissue characterization and outcome prediction.

Abstract WMP13: Multiphase MRA Collateral Map: Prediction of Tissue Outcome and Penumbra in Patients With Acute Ischemic Stroke in the Anterior Circulation

Background and Purpose: We developed the MRA collateral map derived from dynamic MR angiography and grading methods with significant linear association with functional outcomes of patients with acute ischemic stroke (AIS). This study is to verify the value of the MRA collateral map for predicting tissue outcome and penumbra in patients with AIS. Materials and Methods: From a prospectively maintained registry, patients with AIS due to occlusion or stenosis of the unilateral ICA and/or M1 MCA within 8 hours of symptom onset were included. The collateral-perfusion grading based on the MRA collateral map was estimated using 6-scale MAC. Changes of infarct area were divided into two groups with and without infarct growth (IG + and IG - ). Areas of baseline DWI lesion, Tmax &gt; 6s, and decreased collateral-perfusion on each phases of the MRA collateral map, and infarct lesion on follow-up image were compared by visual assessment. Results: One hundred thirty-five patients, including 85 males (mean age, 69 years old), were included. Shorter onset-to-door times (OR=1.04, 95% CI=1.01-1.08) and successful early reperfusion (OR=0.19, 95% CI=0.05-0.66) were independently associated with IG - in multivariate analysis. In subgroup analysis, good collateral-perfusion status was associated with IG - (OR=0.30, 95% CI=0.10-0.91). In IG + group, the infarction grew within hypoperfused area on the phase of the MRA collateral map immediately before the phase that matches the baseline DWI lesion. There was no infarct growth beyond hypoperfused area on the capillary phase of the MRA collateral map in both IG + and IG - groups. The area of Tmax &gt; 6s matched with the hypoperfused area on capillary phase of the MRA collateral map in 83% of patients. Conclusion: In this study, tissue fate in AIS was dependent on early reperfusion. In case of unsuccessful early reperfusion, it was associated with collateral-perfusion status. We suggest that the extent of penumbra can be estimated by the MRA collateral map.

Improved multi-parametric prediction of tissue outcome in acute ischemic stroke patients using spatial features.

IntroductionIn recent years, numerous methods have been proposed to predict tissue outcome in acute stroke patients using machine learning methods incorporating multiparametric imaging data. Most methods include diffusion and perfusion parameters as image-based parameters but do not include any spatial information although these parameters are spatially dependent, e.g. different perfusion properties in white and gray brain matter. This study aims to investigate if including spatial features improves the accuracy of multi-parametric tissue outcome prediction.Materials and methodsAcute and follow-up multi-center MRI datasets of 99 patients were available for this study. Logistic regression, random forest, and XGBoost machine learning models were trained and tested using acute MR diffusion and perfusion features and known follow-up lesions. Different combinations of atlas coordinates and lesion probability maps were included as spatial information. The stroke lesion predictions were compared to the true tissue outcomes using the area under the receiver operating characteristic curve (ROC AUC) and the Dice metric.ResultsThe statistical analysis revealed that including spatial features significantly improves the tissue outcome prediction. Overall, the XGBoost and random forest models performed best in every setting and achieved state-of-the-art results regarding both metrics with similar improvements achieved including Montreal Neurological Institute (MNI) reference space coordinates or voxel-wise lesion probabilities.ConclusionSpatial features should be integrated to improve lesion outcome prediction using machine learning models.

Simulated perfusion MRI data to boost training of convolutional neural networks for lesion fate prediction in acute stroke

The problem of final tissue outcome prediction of acute ischemic stroke is assessed from physically realistic simulated perfusion magnetic resonance images. Different types of simulations with a focus on the arterial input function are discussed. These simulated perfusion magnetic resonance images are fed to convolutional neural network to predict real patients. Performances close to the state-of-the-art performances are obtained with a patient specific approach. This approach consists in training a model only from simulated images tuned to the arterial input function of a tested real patient. This demonstrates the added value of physically realistic simulated images to predict the final infarct from perfusion.

Early-stage 11C-Flumazenil PET predicts day-14 selective neuronal loss in a rodent model of transient focal cerebral ischemia

Predicting tissue outcome early after stroke is an important goal. MRI >3 h accurately predicts infarction but is insensitive to selective neuronal loss (SNL). Previous studies suggest that chronic-stage 11 C-flumazenil PET (FMZ-PET) is a validated marker of SNL in rats, while early-stage FMZ-PET may predict infarction. Whether early FMZ-PET also predicts SNL is unknown. Following 45-min distal MCA occlusion, adult rats underwent FMZ-PET at 1 h and 48 h post-reperfusion to map distribution volume (VT), which reflects GABA-A receptor binding. NeuN immunohistochemistry was performed at Day 14. In each rat, VT and %NeuN loss were determined in 44 ROIs spanning the hemisphere. NeuN revealed isolated SNL and cortical infarction in five and one rats, respectively. In the SNL subgroup, VT-1 h was mildly reduced and only weakly predicted SNL, while VT-48 h was significantly increased and predicted SNL both individually (p < 0.01, Kendall) and across the group (p < 0.001), i.e. the higher the VT, the stronger the SNL. Similar correlations were found in the rat with infarction. Our findings suggest GABA-A receptors are still present on injured neurons at the 48 h timepoint, and the increased 48 h VT observed here is consistent with earlier rat studies showing early GABA-A receptor upregulation. That FMZ binding at 48 h was predictive of SNL may have clinical implications.

Technical considerations of multi-parametric tissue outcome prediction methods in acute ischemic stroke patients

Decisions regarding acute stroke treatment rely heavily on imaging, but interpretation can be difficult for physicians. Machine learning methods can assist clinicians by providing tissue outcome predictions for different treatment approaches based on acute multi-parametric imaging. To produce such clinically viable machine learning models, factors such as classifier choice, data normalization, and data balancing must be considered. This study gives comprehensive consideration to these factors by comparing the agreement of voxel-based tissue outcome predictions using acute imaging and clinical parameters with manual lesion segmentations derived from follow-up imaging. This study considers random decision forest, generalized linear model, and k-nearest-neighbor machine learning classifiers in conjunction with three data normalization approaches (non-normalized, relative to contralateral hemisphere, and relative to contralateral VOI), and two data balancing strategies (full dataset and stratified subsampling). These classifier settings were evaluated based on 90 MRI datasets from acute ischemic stroke patients. Distinction was made between patients recanalized using intraarterial and intravenous methods, as well as those without successful recanalization. For primary quantitative comparison, the Dice metric was computed for each voxel-based tissue outcome prediction and its corresponding follow-up lesion segmentation. It was found that the random forest classifier outperformed the generalized linear model and the k-nearest-neighbor classifier, that normalization did not improve the Dice score of the lesion outcome predictions, and that the models generated lesion outcome predictions with higher Dice scores when trained with balanced datasets. No significant difference was found between the treatment groups (intraarterial vs intravenous) regarding the Dice score of the tissue outcome predictions.

Predicting ischemic stroke tissue fate using a deep convolutional neural network on source magnetic resonance perfusion images.

Predicting infarct volume from magnetic resonance perfusion-weighted imaging can provide helpful information to clinicians in deciding how aggressively to treat acute stroke patients. Models have been developed to predict tissue fate, yet these models are mostly built using hand-crafted features (e.g., time-to-maximum) derived from perfusion images, which are sensitive to deconvolution methods. We demonstrate the application of deep convolution neural networks (CNNs) on predicting final stroke infarct volume using only the source perfusion images. We propose a deep CNN architecture that improves feature learning and achieves an area under the curve of , outperforming existing tissue fate models. We further validate the proposed deep CNN with existing 2-D and 3-D deep CNNs for images/video classification, showing the importance of the proposed architecture. Our work leverages deep learning techniques in stroke tissue outcome prediction, advancing magnetic resonance imaging perfusion analysis one step closer to an operational decision support tool for stroke treatment guidance.

Abstract WMP84: Predicting Cortical Selective Neuronal Loss Using 11 C-flumazenil PET After Temporary Focal Ischemia in Spontaneously Hypertensive Rats

Background: Predicting ischemic brain damage early after stroke is an important goal. T2-Weighted MRI &gt;3hrs after stroke onset accurately predicts infarction but is insensitive to selective neuronal loss (SNL). Chronic-stage 11 C-flumazenil PET (FMZ-PET) is a positive marker of SNL in rats (Ejaz, Neurobiol. Dis. 2013), and previous studies in man, baboons and cats suggest early-stage FMZ-PET could predict infarction. Here we assessed whether early FMZ-PET quantitatively predicts SNL. Methods: Six adult male spontaneously hypertensive rats underwent 45min distal MCA occlusion, and dynamic FMZ-PET was obtained at 1hr and 48hrs post-reperfusion to map distribution volume (DV), which reflects specific binding. Following perfusion-fixation at Day 14, immunohistochemistry using NeuN as neuronal marker was performed. Comparison of normalized FMZ DV to %NeuN loss (determined visually) was done by means of Pearson correlations across 44 cytoarchitectonic ROIs from 8 coronal sections spanning the hemisphere and matching Paxinos atlas, according to previously published methods (Hughes, NeuroImage 2012; Emmrich, JCBFM 2015). Results: NeuN revealed cortical infarction and isolated SNL in 1 and 5 rats, respectively. 1-hr post-reperfusion FMZ-DV was unaffected and did not predict tissue outcome. In contrast, 48-hr FMZ-DV was significantly increased and across the 44 ROIs negatively predicted SNL intensity both individually in 3/5 rats (r range: 0.38-0.60; trend in 2/5 rats) and across the 5 rats (r=0.59, p &lt;0.001), i.e. the higher the DV, the more severe the SNL; similar but stronger correlation (r=0.76) was found in the rat with infarction. Conclusion: The lack of FMZ changes 1hr after reperfusion suggests post-synaptic GABA- A receptors are still present onto dying neurons. Regardless of the underlying mechanism, e.g. receptor upregulation or increased binding via BBB damage, the FMZ binding increase present at 48hrs - which is consistent with earlier rat studies - was predictive of tissue outcome, including SNL - a new finding with potential clinical relevance.

Prediction of Tissue Outcome and Assessment of Treatment Effect in Acute Ischemic Stroke Using Deep Learning.

Treatment options for patients with acute ischemic stroke depend on the volume of salvageable tissue. This volume assessment is currently based on fixed thresholds and single imagine modalities, limiting accuracy. We wish to develop and validate a predictive model capable of automatically identifying and combining acute imaging features to accurately predict final lesion volume. Using acute magnetic resonance imaging, we developed and trained a deep convolutional neural network (CNNdeep) to predict final imaging outcome. A total of 222 patients were included, of which 187 were treated with rtPA (recombinant tissue-type plasminogen activator). The performance of CNNdeep was compared with a shallow CNN based on the perfusion-weighted imaging biomarker Tmax (CNNTmax), a shallow CNN based on a combination of 9 different biomarkers (CNNshallow), a generalized linear model, and thresholding of the diffusion-weighted imaging biomarker apparent diffusion coefficient (ADC) at 600×10-6 mm2/s (ADCthres). To assess whether CNNdeep is capable of differentiating outcomes of ±intravenous rtPA, patients not receiving intravenous rtPA were included to train CNNdeep,-rtpa to access a treatment effect. The networks' performances were evaluated using visual inspection, area under the receiver operating characteristic curve (AUC), and contrast. CNNdeep yields significantly better performance in predicting final outcome (AUC=0.88±0.12) than generalized linear model (AUC=0.78±0.12; P=0.005), CNNTmax (AUC=0.72±0.14; P<0.003), and ADCthres (AUC=0.66±0.13; P<0.0001) and a substantially better performance than CNNshallow (AUC=0.85±0.11; P=0.063). Measured by contrast, CNNdeep improves the predictions significantly, showing superiority to all other methods (P≤0.003). CNNdeep also seems to be able to differentiate outcomes based on treatment strategy with the volume of final infarct being significantly different (P=0.048). The considerable prediction improvement accuracy over current state of the art increases the potential for automated decision support in providing recommendations for personalized treatment plans.