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

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.

Application of emerging technologies in ischemic stroke: from clinical study to basic research

Stroke is a primary cause of noncommunicable disease-related death and disability worldwide. The most common form, ischemic stroke, is increasing in incidence resulting in a significant burden on patients and society. Urgent action is thus needed to address preventable risk factors and improve treatment methods. This review examines emerging technologies used in the management of ischemic stroke, including neuroimaging, regenerative medicine, biology, and nanomedicine, highlighting their benefits, clinical applications, and limitations. Additionally, we suggest strategies for technological development for the prevention, diagnosis, and treatment of ischemic stroke.

Knowledge structure and global trends of machine learning in stroke over the past decade: A scientometric analysis

Objective

Machine learning (ML) models have been widely applied in stroke prediction, diagnosis, treatment, and prognosis assessment. We aimed to conduct a comprehensive scientometrics analysis of studies related to ML in stroke and reveal its current status, knowledge structure, and global trends.

Methods

All documents related to ML in stroke were retrieved from the Web of Science database on March 15, 2023. We refined the documents by including only original articles and reviews in the English language. The literature published over the past decade was imported into scientometrics software for influence detection and collaborative network analysis.

Results

2389 related publications were included. The annual publication outputs demonstrated explosive growth, with an average growth rate of 63.99 %. Among the 90 countries/regions involved, the United States (729 articles) and China (636 articles) were the most productive countries. Frontiers in Neurology was the most prolific journal with 94 articles. 234 highly cited articles, each with more than 31 citations, were detected. Keyword analysis revealed a total of 5333 keywords, with a predominant focus on the application of ML models in the early diagnosis, classification, and prediction of "acute ischemic stroke" and "atrial fibrillation-related stroke". The keyword "classification" had the first and longest burst, spanning from 2013 to 2018. 'Upport vector machine' got the strongest burst strength with 6.2. Keywords such as 'mechanical thrombectomy', 'expression', and 'prognosis' experienced bursts in 2022 and have continued to be prominent.

Conclusion

The applications of ML in stroke are increasingly diverse and extensive, with researchers showing growing interest over the past decade. However, the clinical application of ML in stroke is still in its early stages, and several limitations and challenges need to be addressed for its widespread adoption in clinical practice.

Recent developments in pre-hospital and in-hospital triage for endovascular stroke treatment

Triage describes the assignment of resources based on where they can be best used, are most needed, or are most likely to achieve success. Triage is of particular importance in time-critical conditions such as acute ischemic stroke. In this setting, one of the goals of triage is to minimize the delay to endovascular thrombectomy (EVT), without delaying intravenous thrombolysis or other time-critical treatments including patients who cannot benefit from EVT. EVT triage is highly context-specific, and depends on availability of financial resources, staff resources, local infrastructure, and geography. Furthermore, the EVT triage landscape is constantly changing, as EVT indications evolve and new neuroimaging methods, EVT technologies, and adjunctive medical treatments are developed and refined. This review provides an overview of recent developments in EVT triage at both the pre-hospital and in-hospital stages. We discuss pre-hospital large vessel occlusion detection tools, transport paradigms, in-hospital workflows, acute stroke neuroimaging protocols, and angiography suite workflows. The most important factor in EVT triage, however, is teamwork. Irrespective of any new technology, EVT triage will only reach optimal performance if all team members, including paramedics, nurses, technologists, emergency physicians, neurologists, radiologists, neurosurgeons, and anesthesiologists, are involved and engaged. Thus, building sustainable relationships through continuous efforts and hands-on training forms an integral part in ensuring rapid and efficient EVT triage.

BEA-Net: Body and Edge Aware Network With Multi-Scale Short-Term Concatenation for Medical Image Segmentation

Medical image segmentation is indispensable for diagnosis and prognosis of many diseases. To improve the segmentation performance, this study proposes a new 2D body and edge aware network with multi-scale short-term concatenation for medical image segmentation. Multi-scale short-term concatenation modules which concatenate successive convolution layers with different receptive fields, are proposed for capturing multi-scale representations with fewer parameters. Body generation modules with feature adjustment based on weight map computing via enlarging the receptive fields, and edge generation modules with multi-scale convolutions using Sobel kernels for edge detection, are proposed to separately learn body and edge features from convolutional features in decoders, making the proposed network be body and edge aware. Based on the body and edge modules, we design parallel body and edge decoders whose outputs are fused to achieve the final segmentation. Besides, deep supervision from the body and edge decoders is applied to ensure the effectiveness of the generated body and edge features and further improve the final segmentation. The proposed method is trained and evaluated on six public medical image segmentation datasets to show its effectiveness and generality. Experimental results show that the proposed method achieves better average Dice similarity coefficient and 95% Hausdorff distance than several benchmarks on all used datasets. Ablation studies validate the effectiveness of the proposed multi-scale representation learning modules, body and edge generation modules and deep supervision.

Core and penumbra estimation using deep learning-based AIF in association with clinical measures in computed tomography perfusion (CTP)

OBJECTIVES

To investigate whether utilizing a convolutional neural network (CNN)-based arterial input function (AIF) improves the volumetric estimation of core and penumbra in association with clinical measures in stroke patients.

METHODS

The study included 160 acute ischemic stroke patients (male = 87, female = 73, median age = 73 years) with approval from the institutional review board. The patients had undergone CTP imaging, NIHSS and ASPECTS grading. convolutional neural network (CNN) model was trained to fit a raw AIF curve to a gamma variate function. CNN AIF was utilized to estimate the core and penumbra volumes which were further validated with clinical scores.

RESULTS

Penumbra estimated by CNN AIF correlated positively with the NIHSS score (r = 0.69; p < 0.001) and negatively with the ASPECTS (r =  - 0.43; p < 0.001). The CNN AIF estimated penumbra and core volume matching the patient symptoms, typically in patients with higher NIHSS (> 20) and lower ASPECT score (< 5). In group analysis, the median CBF < 20%, CBF < 30%, rCBF < 38%, Tmax > 10 s, Tmax > 10 s volumes were statistically significantly higher (p < .05).

CONCLUSIONS

With inclusion of the CNN AIF in perfusion imaging pipeline, penumbra and core estimations are more reliable as they correlate with scores representing neurological deficits in stroke.

CRITICAL RELEVANCE STATEMENT

With CNN AIF perfusion imaging pipeline, penumbra and core estimations are more reliable as they correlate with scores representing neurological deficits in stroke.

Interpretable machine learning models for predicting in-hospital death in patients in the intensive care unit with cerebral infarction

BACKGROUND

Research on patients with cerebral infarction in the Intensive Care Unit (ICU) is still lacking. Our study aims to develop and validate multiple machine-learning (ML) models using two large ICU databases-Medical Information Mart for Intensive Care version III (MIMIC-III) and eICU Research Institute Database (eRI)-to guide clinical practice.

METHODS

We collected clinical data from patients with cerebral infarction in the MIMIC-III and eRI databases within 24 h of admission. The opinion of neurologists and the Least Absolute Shrinkage and Selection Operator regression was used to screen for relevant clinical features. Using eRI as the training set and MIMIC-III as the test set, we developed and validated six ML models. Based on the results of the model validation, we select the best model and perform the interpretability analysis on it.

RESULTS

A total of 4,338 patients were included in the study (eRI:3002, MIMIC-III:1336), resulting in a total of 18 clinical characteristics through screening. Model validation results showed that random forest (RF) was the best model, with AUC and F1 scores of 0.799 and 0.417 in internal validation and 0.733 and 0.498 in external validation, respectively; moreover, its sensitivity and recall were the highest of the six algorithms for both the internal and external validation. The explanatory analysis of the model showed that the three most important variables in the RF model were Acute Physiology Score-III, Glasgow Coma Scale score, and heart rate, and that the influence of each variable on the judgement of the model was consistent with medical knowledge.

CONCLUSION

Based on a large sample of patients and advanced algorithms, our study bridges the limitations of studies on this area. With our model, physicians can use the admission information of cerebral infarction patients in the ICU to identify high-risk groups among them who are prone to in-hospital death, so that they could be more alert to this group of patients and upgrade medical measures early to minimize the mortality of patients.

Desired Qualities of Endovascular Tools and Barriers to Treating Medium Vessel Occlusion MeVO : Insights from the MeVO-FRONTIERS International Survey

BACKGROUND

Endovascular treatment (EVT) for stroke due to medium vessel occlusion (MeVO) can be technically challenging and specific endovascular tools are needed to safely and effectively recanalize these relatively small and fragile vessels. We aimed to gain insight into availability and desired qualities of endovascular devices used in MeVO stroke and examined barriers to adoption of MeVO EVT in clinical practice on a global scale.

METHODS

We conducted a case-based international survey among neurointerventionalists. As a part of the survey, participants were asked whether they felt appropriate endovascular tools for MeVO stroke exist and are available to them in their clinical practice. We then examined barriers to adopting MeVO EVT and analyzed them by geographic regions.

RESULTS

A total of 263 neurointerventionists participated, of which 178 (67.7%) and 83 (31.6%) provided responses on desired qualities of MeVO EVT tools and on barriers to their adoption in local practice, respectively. The majority 121/178 (68%) felt there was substantial room for improvement regarding existing tools. A large proportion 131/178 (73.6%) felt they had appropriate access to existing tools. The most commonly mentioned barrier for adopting MeVO EVT in North America was "awaiting better tools" (9/28 responses, 32.1%), while "awaiting better evidence" (8/26 responses, 30.8%), and the need for improved "funding" (7/26 responses, 26.9%) were important barriers in Europe.

CONCLUSION

The majority of surveyed neurointerventionalists felt that dedicated MeVO EVT tools can be substantially improved upon. Different regions face various challenges in adoption of MeVO EVT, but overall, physicians are mostly awaiting better MeVO EVT tools.

Intracranial Hemorrhage Detection Using Parallel Deep Convolutional Models and Boosting Mechanism

Intracranial hemorrhage (ICH) can lead to death or disability, which requires immediate action from radiologists. Due to the heavy workload, less experienced staff, and the complexity of subtle hemorrhages, a more intelligent and automated system is necessary to detect ICH. In literature, many artificial-intelligence-based methods are proposed. However, they are less accurate for ICH detection and subtype classification. Therefore, in this paper, we present a new methodology to improve the detection and subtype classification of ICH based on two parallel paths and a boosting technique. The first path employs the architecture of ResNet101-V2 to extract potential features from windowed slices, whereas Inception-V4 captures significant spatial information in the second path. Afterwards, the detection and subtype classification of ICH is performed by the light gradient boosting machine (LGBM) using the outputs of ResNet101-V2 and Inception-V4. Thus, the combined solution, known as ResNet101-V2, Inception-V4, and LGBM (Res-Inc-LGBM), is trained and tested over the brain computed tomography (CT) scans of CQ500 and Radiological Society of North America (RSNA) datasets. The experimental results state that the proposed solution efficiently obtains 97.7% accuracy, 96.5% sensitivity, and 97.4% F1 score using the RSNA dataset. Moreover, the proposed Res-Inc-LGBM outperforms the standard benchmarks for the detection and subtype classification of ICH regarding the accuracy, sensitivity, and F1 score. The results prove the significance of the proposed solution for its real-time application.

Spatial accuracy of computed tomography perfusion to estimate the follow-up infarct on diffusion-weighted imaging after successful mechanical thrombectomy

BACKGROUND

Volumetric accuracy of using computed tomography perfusion (CTP) to estimate the post-treatment infarct in stroke patients with successful recanalization after mechanical thrombectomy (MT) has been studied a lot, however the spatial accuracy and its influence factors has not been fully investigated.

METHODS

This retrospective study reviewed the data from consecutive anterior large vessel occlusion (LVO) patients who had baseline CTP, successful recanalization after MT, and post-treatment diffusion-weighed imaging (DWI). Ischemic core on baseline CTP was estimated using relative cerebral blood flood (CBF) of < 30%. The infarct area was outlined manually on post-treatment DWI, and registered to CTP. Spatial agreement was assessed using the Dice similarity coefficient (DSC) and average Hausdorff distance. According to the median DSC, the study population was dichotomized into high and low Dice groups. Univariable and multivariable regression analyses were used to determine the factors independently associated with the spatial agreement.

RESULTS

In 72 included patients, the median DSC was 0.26, and the median average Hausdorff distance was 1.77 mm. High Dice group showed significantly higher median ischemic core volume on baseline CTP (33.90 mL vs 3.40 mL, P < 0.001), lower proportion of moderate or severe leukoaraiosis [27.78% vs 52.78%, P = 0.031], and higher median infarct volume on follow-up DWI (51.17 mL vs 9.42 mL, P < 0.001) than low Dice group. Ischemic core volume on baseline CTP was found to be independently associated with the spatial agreement (OR, 1.092; P < 0.001).

CONCLUSIONS

CTP could help to spatially locate the post-treatment infarct in anterior LVO patients who achieving successful recanalization after MT. Ischemic core volume on baseline CTP was independently associated with the spatial agreement.

Neural network-derived perfusion maps: A model-free approach to computed tomography perfusion in patients with acute ischemic stroke

Objective

In this study, we investigate whether a Convolutional Neural Network (CNN) can generate informative parametric maps from the pre-processed CT perfusion data in patients with acute ischemic stroke in a clinical setting.

Methods

The CNN training was performed on a subset of 100 pre-processed perfusion CT dataset, while 15 samples were kept for testing. All the data used for the training/testing of the network and for generating ground truth (GT) maps, using a state-of-the-art deconvolution algorithm, were previously pre-processed using a pipeline for motion correction and filtering. Threefold cross validation had been used to estimate the performance of the model on unseen data, reporting Mean Squared Error (MSE). Maps accuracy had been checked through manual segmentation of infarct core and total hypo-perfused regions on both CNN-derived and GT maps. Concordance among segmented lesions was assessed using the Dice Similarity Coefficient (DSC). Correlation and agreement among different perfusion analysis methods were evaluated using mean absolute volume differences, Pearson correlation coefficients, Bland-Altman analysis, and coefficient of repeatability across lesion volumes.

Results

The MSE was very low for two out of three maps, and low in the remaining map, showing good generalizability. Mean Dice scores from two different raters and the GT maps ranged from 0.80 to 0.87. Inter-rater concordance was high, and a strong correlation was found between lesion volumes of CNN maps and GT maps (0.99, 0.98, respectively).

Conclusion

The agreement between our CNN-based perfusion maps and the state-of-the-art deconvolution-algorithm perfusion analysis maps, highlights the potential of machine learning methods applied to perfusion analysis. CNN approaches can reduce the volume of data required by deconvolution algorithms to estimate the ischemic core, and thus might allow the development of novel perfusion protocols with lower radiation dose deployed to the patient.

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.

Performance of Machine Learning for Tissue Outcome Prediction in Acute Ischemic Stroke: A Systematic Review and Meta-Analysis

Machine learning (ML) has been proposed for lesion segmentation in acute ischemic stroke (AIS). This study aimed to provide a systematic review and meta-analysis of the overall performance of current ML algorithms for final infarct prediction from baseline imaging. We made a comprehensive literature search on eligible studies developing ML models for core infarcted tissue estimation on admission CT or MRI in AIS patients. Eleven studies meeting the inclusion criteria were included in the quantitative analysis. Study characteristics, model methodology, and predictive performance of the included studies were extracted. A meta-analysis was conducted on the dice similarity coefficient (DSC) score by using a random-effects model to assess the overall predictive performance. Study heterogeneity was assessed by Cochrane Q and Higgins I² tests. The pooled DSC score of the included ML models was 0.50 (95% CI 0.39–0.61), with high heterogeneity observed across studies (I² 96.5%, p < 0.001). Sensitivity analyses using the one-study removed method showed the adjusted overall DSC score ranged from 0.47 to 0.52. Subgroup analyses indicated that the DL-based models outperformed the conventional ML classifiers with the best performance observed in DL algorithms combined with CT data. Despite the presence of heterogeneity, current ML-based approaches for final infarct prediction showed moderate but promising performance. Before well integrated into clinical stroke workflow, future investigations are suggested to train ML models on large-scale, multi-vendor data, validate on external cohorts and adopt formalized reporting standards for improving model accuracy and robustness.

Multi-Mode Imaging Scale for Endovascular Therapy in Patients with Acute Ischemic Stroke (META)

Background: With the guidance of multi-mode imaging, the time window for endovascular thrombectomy (EVT) has been expanded to 24 h. However, poor clinical outcomes are still not uncommon. We aimed to develop a multi-mode imaging scale for endovascular therapy in patients with acute ischemic stroke (META) to predict the neurological outcome in patients receiving endovascular thrombectomy (EVT). Methods: We included consecutive acute ischemic stroke patients with occlusion of middle cerebral artery and/or internal carotid artery who underwent EVT. Poor outcome was defined as modified Rankin Scale (mRS) score of 3−6 at 3 months. A five-point META score was constructed based on clot burden score, multi-segment clot, the Alberta Stroke Program early computed tomography score of cerebral blood volume (CBV-ASPECTS), and collateral status. We evaluated the META score performance using area under the curve (AUC) calculations. Results: A total of 259 patients were included. A higher META score was independently correlated with poor outcomes at 3 months (odds ratio, 1.690, 95% CI, 1.340 to 2.132, p < 0.001) after adjusting for age, hypertension, baseline National Institutes of Health Stroke Scale (NIHSS) score, and baseline blood glucose. Patients with a META score ≥ 2 were less likely to benefit from EVT (mRS 3−6: 60.8% vs. 29.2%, p < 0.001). The META score predicted poor outcomes with an AUC of 0.714, higher than the Pittsburgh Response to Endovascular therapy (PRE) score, the totaled health risks in vascular events (THRIVE) score (AUC: 0.566, 0.706), and the single imaging marker in the scale. Conclusions: The novel META score could refine the predictive accuracy of prognosis after EVT, which might provide a promising avenue for future automatic imaging analysis to help decision making.

Use of Machine Learning Algorithms to Predict the Outcomes of Mechanical Thrombectomy in Acute Ischemic Stroke Patients With an Extended Therapeutic Time Window

OBJECTIVE

The aim of this study was to evaluate the performance of machine learning (ML) algorithms in predicting the functional outcome of mechanical thrombectomy (MT) outside the 6-hour therapeutic time window in patients with acute ischemic stroke (AIS).

METHODS

One hundred seventy-seven consecutive AIS patients with large-vessel occlusion in the anterior circulation who underwent MT in the extended time window were enrolled. Clinical, neuroimaging, and treatment variables that could be obtained quickly in the real-world emergency settings were collected. Four machine learning algorithms (random forests, regularized logistic regression, support vector machine, and naive Bayes) were used to predict good outcomes (modified Rankin Scale scores of 0-2) at 90 days by using (1) only variables at admission and (2) both baseline and treatment variables. The performance of each model was evaluated using receiver operating characteristic (ROC) curve analysis. Feature importance was ranked using random forest algorithms.

RESULTS

Eighty patients (45.2%) had a favorable 90-day outcome. Machine learning models including baseline clinical and neuroimaging characteristics predicted 90-day modified Rankin Scale with an area under the ROC curve of 0.80-0.81, sensitivity of 0.60-0.71 and specificity of 0.71-0.76. Further inclusion the treatment variables significantly improved the predictive performance (mean area under the ROC curve, 0.89-0.90; sensitivity, 0.77-0.85; specificity, 0.75-0.87). The most important characteristics for predicting 90-day outcomes were age, hypoperfusion intensity ratio at admission, and National Institutes of Health Stroke Scale score at 24 hours after MT.

CONCLUSIONS

Machine learning algorithms may facilitate prediction of 90-day functional outcomes in AIS patients with an extended therapeutic time window.

The Assessment of Endovascular Therapies in Ischemic Stroke: Management, Problems and Future Approaches

Ischemic stroke accounts for over 80% of all strokes and is one of the leading causes of mortality and permanent disability worldwide. Intravenous administration of recombinant tissue plasminogen activator (rt-PA) is an approved treatment strategy for acute ischemic stroke of large arteries within 4.5 h of onset, and mechanical thrombectomy can be used for large arteries occlusion up to 24 h after onset. Improving diagnostic work up for acute treatment, reducing onset-to-needle time and urgent radiological access angiographic CT images (angioCT) and Magnetic Resonance Imaging (MRI) are real problems for many healthcare systems, which limits the number of patients with good prognosis in real world compared to the results of randomized controlled trials. The applied endovascular procedures demonstrated high efficacy, but some cellular mechanisms, following reperfusion, are still unknown. Changes in the morphology and function of mitochondria associated with reperfusion and ischemia-reperfusion neuronal death are still understudied research fields. Moreover, future research is needed to elucidate the relationship between continuously refined imaging techniques and the variable structure or physical properties of the clot along with vascular permeability and the pleiotropism of ischemic reperfusion lesions in the penumbra, in order to define targeted preventive procedures promoting long-term health benefits.

Prediction of Stroke Infarct Growth Rates by Baseline Perfusion Imaging

BACKGROUND AND PURPOSE

Computed tomography perfusion imaging allows estimation of tissue status in patients with acute ischemic stroke. We aimed to improve prediction of the final infarct and individual infarct growth rates using a deep learning approach.

METHODS

We trained a deep neural network to predict the final infarct volume in patients with acute stroke presenting with large vessel occlusions based on the native computed tomography perfusion images, time to reperfusion and reperfusion status in a derivation cohort (MR CLEAN trial [Multicenter Randomized Clinical Trial of Endovascular Treatment for Acute Ischemic Stroke in the Netherlands]). The model was internally validated in a 5-fold cross-validation and externally in an independent dataset (CRISP study [CT Perfusion to Predict Response to Recanalization in Ischemic Stroke Project]). We calculated the mean absolute difference between the predictions of the deep learning model and the final infarct volume versus the mean absolute difference between computed tomography perfusion imaging processing by RAPID software (iSchemaView, Menlo Park, CA) and the final infarct volume. Next, we determined infarct growth rates for every patient.

RESULTS

We included 127 patients from the MR CLEAN (derivation) and 101 patients of the CRISP study (validation). The deep learning model improved final infarct volume prediction compared with the RAPID software in both the derivation, mean absolute difference 34.5 versus 52.4 mL, and validation cohort, 41.2 versus 52.4 mL (P<0.01). We obtained individual infarct growth rates enabling the estimation of final infarct volume based on time and grade of reperfusion.

CONCLUSIONS

We validated a deep learning-based method which improved final infarct volume estimations compared with classic computed tomography perfusion imaging processing. In addition, the deep learning model predicted individual infarct growth rates which could enable the introduction of tissue clocks during the management of acute stroke.

Machine Learning-Based Approaches for Prediction of Patients’ Functional Outcome and Mortality after Spontaneous Intracerebral Hemorrhage

Spontaneous intracerebral hemorrhage (SICH) has been common in China with high morbidity and mortality rates. This study aims to develop a machine learning (ML)-based predictive model for the 90-day evaluation after SICH. We retrospectively reviewed 751 patients with SICH diagnosis and analyzed clinical, radiographic, and laboratory data. A modified Rankin scale (mRS) of 0–2 was defined as a favorable functional outcome, while an mRS of 3–6 was defined as an unfavorable functional outcome. We evaluated 90-day functional outcome and mortality to develop six ML-based predictive models and compared their efficacy with a traditional risk stratification scale, the intracerebral hemorrhage (ICH) score. The predictive performance was evaluated by the areas under the receiver operating characteristic curves (AUC). A total of 553 patients (73.6%) reached the functional outcome at the 3rd month, with the 90-day mortality rate of 10.2%. Logistic regression (LR) and logistic regression CV (LRCV) showed the best predictive performance for functional outcome (AUC = 0.890 and 0.887, respectively), and category boosting presented the best predictive performance for the mortality (AUC = 0.841). Therefore, ML might be of potential assistance in the prediction of the prognosis of SICH.

Prognosis of ischemic stroke predicted by machine learning based on multi-modal MRI radiomics

OBJECTIVE

Increased risk of stroke is highly associated with psychiatric disorders. We aimed to conduct the machine learning model based on multi-modal magnetic resonance imaging (MRI) radiomics predicting the prognosis of ischemic stroke.

METHODS

This study retrospectively analyzed 148 patients with acute ischemic stroke due to anterior circulation artery occlusion. Based on the modified Rankin Scale (mRS) score, patients were divided into good (mRS ≤ 2) and poor (mRS > 2) outcome groups. Segmentation of the infarct region was performed by manually outlining a mask of the lesion on diffusion-weighted images (DWI) using MRIcron software. The apparent diffusion coefficient (ADC), fluid decay inversion recoverage (FLAIR), susceptibility weighted imaging (SWI) and T1-weighted (T1w) images were aligned to the DWI images and the radiomic features within the lesion area were extracted for each image modality. The calculations were done using pyradiomics software and a total of 4,744 stroke-related imaging features were automatically calculated. Next, feature selection based on recursive feature elimination was used for each modality and three radiomic features were extracted from each modality plus one feature from the lesion mask, for a total of 16 radiomic features. At last, five machine learning (ML) models were trained and tested to predict stroke prognosis, calculate the received operating characteristic (ROC) curves and other parameters, evaluate the performance of the models and validate their predictive efficacy by five-fold cross-validation.

RESULTS

Sixteen radiomic features were selected to construct the ML models for prognostic classification. By five-fold cross-validation, light gradient boosting machine (LightGBM) model-based muti-modal MRI radiomic features performed best in binary prognostic classification with accuracy of 0.831, sensitivity of 0.739, specificity of 0.902, F1-score of 0.788 and an area under the curve (AUC) of 0.902.

CONCLUSION

The ML models based on muti-modal MRI radiomics are of high value for predicting clinical outcomes in acute stroke patients.

Computed Tomography Perfusion-Based Prediction of Core Infarct and Tissue at Risk: Can Artificial Intelligence Help Reduce Radiation Exposure?

BACKGROUND AND PURPOSE

We explored the feasibility of automated, arterial input function independent, vendor neutral prediction of core infarct, and penumbral tissue using complete and partial computed tomographic perfusion data sets through neural networks.

METHODS

Using retrospective computed tomographic perfusion data from 57 patients, split as training/validation (60%/40%), we developed and validated separate 2-dimensional U-net models for cerebral blood flow (CBF) and time to maximum (Tmax) maps calculation to predict core infarct and tissue at risk, respectively. Once trained, the full sets of 28 input images were sequentially reduced to equitemporal 14, 10, and 7 time points. The averaged structural similarity index measure between the model-derived images and ground truth perfusion maps was compared. Volumes for core infarct and Tmax were compared using the Pearson correlation coefficient.

RESULTS

Both CBF and Tmax maps derived using 28 and 14 time points had similar structural similarity index measure (0.80-0.81; P>0.05) when compared with ground truth images. The Pearson correlation for the CBF and Tmax volumes derived from the model using 28-tp with ground truth volumes derived from the RAPID software was 0.69 for CBF and 0.74 for Tmax. The predicted maps were fully concordant in terms of laterality to the commercial perfusion maps. The mean Dice scores were 0.54 for the core infarct and 0.63 for the hypoperfusion maps.

CONCLUSIONS

Artificial intelligence model-derived volumes show good correlation with RAPID-derived volumes for CBF and Tmax. Within the constraints of a small sample size, the perfusion map quality is similar when using 14-tp instead of 28-tp. Our findings provide proof of concept that vendor neutral artificial intelligence models for computed tomographic perfusion processing using complete or partial image data sets appear feasible. The model accuracy could be further optimized using larger data sets.