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

Ensemble Machine Learning to Predict Futile Recanalization After Mechanical Thrombectomy based on Non-Contrast CT Imaging

OBJECTIVES

Despite successful recanalization after Mechanical Thrombectomy (MT), approximately 25% of patients with Acute Ischemic Stroke (AIS) due to Large Vessel Occlusion (LVO) show unfavorable clinical outcomes, namely Futile Recanalization (FR). We aimed to use a Machine Learning (ML) Non-Contrast brain CT (NCCT) imaging predictive model to identify FR in patients undergoing MT.

MATERIALS & METHODS

Between July 2022 and December 2022, 70 consecutive patients with LVO undergoing a complete recanalization (eTICI 3) with MT within 8 hours from onset at our Centre were analyzed. Two NCCT images per patient of middle cerebral artery vascular territory and patients' clinical characteristics were classified by the presence of ischemic features on 24h NCCT after MT. Each slice was segmented with "Mazda" software ver.4.6 by placing a Region Of Interest (ROI) on the whole brain by two radiologists in consensus. A total of 381 features were extracted for each slice. The dataset was split into train and test set with a 70:30 ratio.

RESULTS

Eleven classification models were trained. An Ensemble Machine Learning (EML) model was obtained by averaging the predictions of models with accuracy on a test set >70%, with and without patients' clinical characteristics. The EML model combined with clinical data showed an accuracy of 0.76, a sensitivity of 0.88, a specificity of 0.69 with a NPV of 0.90, a PPV of 0.64, with AUC of 0.84.

CONCLUSION

NCCT and ML analysis shows promise in predicting FR after complete recanalization following MT in AIS patients. Larger studies are required to confirm these preliminary results.

From data to decisions: AI and functional connectivity for diagnosis, prognosis, and recovery prediction in stroke

Stroke is a severe medical condition which may lead to permanent disability conditions. The initial 8 weeks following a stroke are crucial for rehabilitation, as most recovery occurs during this period. Personalized approaches and predictive biomarkers are needed for tailored rehabilitation. In this context, EEG brain connectivity and Artificial Intelligence (AI) can play a crucial role in diagnosing and predicting stroke outcomes efficiently. In the present study, 127 patients with subacute ischemic lesions and 90 age- and gender-matched healthy controls were enrolled. EEG recordings were obtained from each participant within 15 days of stroke onset. Clinical evaluations were performed at baseline and at 40-days follow-up using the National Institutes of Health Stroke Scale (NIHSS). Functional connectivity analysis was conducted using Total Coherence (TotCoh) and Small Word (SW). Quadratic support vector machines (SVM) algorithms were implemented to classify healthy subjects compared to stroke patients (Healthy vs Stroke), determine the affected hemisphere (Left vs Right Hemisphere), and predict functional recovery (Functional Recovery Prediction). In the classification for Functional Recovery Prediction, an accuracy of 94.75%, sensitivity of 96.27% specificity of 92.33%, and AUC of 0.95 were achieved; for Healthy vs Stroke, an accuracy of 99.09%, sensitivity of 100%, specificity of 98.46%, and AUC of 0.99 were achieved. For Left vs Right Hemisphere classification, accuracy was 86.77%, sensitivity was 91.44%, specificity was 80.33%, and AUC was 0.87. These findings highlight the potential of utilizing functional connectivity measures based on EEG in combination with AI algorithms to improve patient outcomes by targeted rehabilitation interventions.

Machine learning-based nomogram: integrating MRI radiomics and clinical indicators for prognostic assessment in acute ischemic stroke

Background

Acute Ischemic Stroke (AIS) remains a leading cause of mortality and disability worldwide. Rapid and precise prognostication of AIS is crucial for optimizing treatment strategies and improving patient outcomes. This study explores the integration of machine learning-derived radiomics signatures from multi-parametric MRI with clinical factors to forecast AIS prognosis.

Objective

To develop and validate a nomogram that combines a multi-MRI radiomics signature with clinical factors for predicting the prognosis of AIS.

Methods

This retrospective study involved 506 AIS patients from two centers, divided into training (n = 277) and validation (n = 229) cohorts. 4,682 radiomic features were extracted from T1-weighted, T2-weighted, and diffusion-weighted imaging. Logistic regression analysis identified significant clinical risk factors, which, alongside radiomics features, were used to construct a predictive clinical-radiomics nomogram. The model's predictive accuracy was evaluated using calibration and ROC curves, focusing on distinguishing between favorable (mRS ≤ 2) and unfavorable (mRS > 2) outcomes.

Results

Key findings highlight coronary heart disease, platelet-to-lymphocyte ratio, uric acid, glucose levels, homocysteine, and radiomics features as independent predictors of AIS outcomes. The clinical-radiomics model achieved a ROC-AUC of 0.940 (95% CI: 0.912-0.969) in the training set and 0.854 (95% CI: 0.781-0.926) in the validation set, underscoring its predictive reliability and clinical utility.

Conclusion

The study underscores the efficacy of the clinical-radiomics model in forecasting AIS prognosis, showcasing the pivotal role of artificial intelligence in fostering personalized treatment plans and enhancing patient care. This innovative approach promises to revolutionize AIS management, offering a significant leap toward more individualized and effective healthcare solutions.

Ischemic stroke outcome prediction with diversity features from whole brain tissue using deep learning network

Objectives

This study proposed an outcome prediction method to improve the accuracy and efficacy of ischemic stroke outcome prediction based on the diversity of whole brain features, without using basic information about patients and image features in lesions.

Design

In this study, we directly extracted dynamic radiomics features (DRFs) from dynamic susceptibility contrast perfusion-weighted imaging (DSC-PWI) and further extracted static radiomics features (SRFs) and static encoding features (SEFs) from the minimum intensity projection (MinIP) map, which was generated from the time dimension of DSC-PWI images. After selecting whole brain features Ffuse from the combinations of DRFs, SRFs, and SEFs by the Lasso algorithm, various machine and deep learning models were used to evaluate the role of Ffuse in predicting stroke outcomes.

Results

The experimental results show that the feature Ffuse generated from DRFs, SRFs, and SEFs (Resnet 18) outperformed other single and combination features and achieved the best mean score of 0.971 both on machine learning models and deep learning models and the 95% CI were (0.703, 0.877) and (0.92, 0.983), respectively. Besides, the deep learning models generally performed better than the machine learning models.

Conclusion

The method used in our study can achieve an accurate assessment of stroke outcomes without segmentation of ischemic lesions, which is of great significance for rapid, efficient, and accurate clinical stroke treatment.

A comprehensive review for artificial intelligence on neuroimaging in rehabilitation of ischemic stroke

Stroke is the second leading cause of death worldwide, with ischemic stroke accounting for a significant proportion of morbidity and mortality among stroke patients. Ischemic stroke often causes disability and cognitive impairment in patients, which seriously affects the quality of life of patients. Therefore, how to predict the recovery of patients can provide support for clinical intervention in advance and improve the enthusiasm of patients for rehabilitation treatment. With the popularization of imaging technology, the diagnosis and treatment of ischemic stroke patients are often accompanied by a large number of imaging data. Through machine learning and Deep Learning, information from imaging data can be used more effectively. In this review, we discuss recent advances in neuroimaging, machine learning, and Deep Learning in the rehabilitation of ischemic stroke.

Predicting long-term outcomes for acute ischemic stroke using multi-model MRI radiomics and clinical variables

Purpose

The objective of this study was to create and validate a novel prediction model that incorporated both multi-modal radiomics features and multi-clinical features, with the aim of accurately identifying acute ischemic stroke (AIS) patients who faced a higher risk of poor outcomes.

Methods

A cohort of 461 patients diagnosed with AIS from four centers was divided into a training cohort and a validation cohort. Radiomics features were extracted and selected from diffusion-weighted imaging (DWI) and apparent diffusion coefficient (ADC) images to create a radiomic signature. Prediction models were developed using multi-clinical and selected radiomics features from DWI and ADC.

Results

A total of 49 radiomics features were selected from DWI and ADC images by the least absolute shrinkage and selection operator (LASSO). Additionally, 20 variables were collected as multi-clinical features. In terms of predicting poor outcomes in validation set, the area under the curve (AUC) was 0.727 for the DWI radiomics model, 0.821 for the ADC radiomics model, 0.825 for the DWI + ADC radiomics model, and 0.808 for the multi-clinical model. Furthermore, a prediction model was built using all selected features, the AUC for predicting poor outcomes increased to 0.86.

Conclusion

Radiomics features extracted from DWI and ADC images can serve as valuable biomarkers for predicting poor clinical outcomes in patients with AIS. Furthermore, when these radiomics features were combined with multi-clinical features, the predictive performance was enhanced. The prediction model has the potential to provide guidance for tailoring rehabilitation therapies based on individual patient risks for poor outcomes.

Current status and quality of radiomics studies for predicting outcome in acute ischemic stroke patients: a systematic review and meta-analysis

Background

Pre-treatment prediction of reperfusion and long-term prognosis in acute ischemic stroke (AIS) patients is crucial for effective treatment and decision-making. Recent studies have demonstrated that the inclusion of radiomics data can improve the performance of predictive models. This paper reviews published studies focused on radiomics-based prediction of reperfusion and long-term prognosis in AIS patients.

Methods

We systematically searched PubMed, Web of Science, and Cochrane databases up to September 9, 2023, for studies on radiomics-based prediction of AIS patient outcomes. The methodological quality of the included studies was evaluated using the phase classification criteria, the radiomics quality scoring (RQS) tool, and the Prediction model Risk Of Bias Assessment Tool (PROBAST). Two separate meta-analyses were performed of these studies that predict long-term prognosis and reperfusion in AIS patients.

Results

Sixteen studies with sample sizes ranging from 67 to 3,001 were identified. Ten studies were classified as phase II, and the remaining were categorized as phase 0 (n = 2), phase I (n = 1), and phase III (n = 3). The mean RQS score of all studies was 39.41%, ranging from 5.56 to 75%. Most studies (87.5%, 14/16) were at high risk of bias due to their retrospective design. The remaining two studies were categorized as low risk and unclear risk, respectively. The pooled area under the curve (AUC) was 0.88 [95% confidence interval (CI) 0.84-0.92] for predicting the long-term prognosis and 0.80 (95% CI 0.74-0.86) for predicting reperfusion in AIS.

Conclusion

Radiomics has the potential to predict immediate reperfusion and long-term outcomes in AIS patients. Further external validation and evaluation within the clinical workflow can facilitate personalized treatment for AIS patients. This systematic review provides valuable insights for optimizing radiomics prediction systems for both reperfusion and long-term outcomes in AIS patients.

Systematic review registration

https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42023461671, identifier CRD42023461671.

Textural Analysis of the Hyperdense Artery Sign in Patients with Acute Ischemic Stroke Predicts the Outcome of Thrombectomy

Textural analysis is pivotal in augmenting the diagnosis and outcomes of endovascular procedures for stroke patients. Due to the detection of changes imperceptible to the human eye, this type of analysis can potentially aid in deciding the optimal type of endovascular treatment. We included 40 patients who suffered from acute ischemic stroke caused by large vessel occlusion, and calculated 130 different textural features based on the non-enhanced CT scan using an open-source software (3D Slicer). Using chi-squared and Mann-Whitney tests and receiver operating characteristics analysis, we identified a total of 21 different textural parameters capable of predicting the outcome of thrombectomy (quantified as the mTICI score), with variable sensitivity (50-97.9%) and specificity (64.6-99.4%) rates. In conclusion, CT-based radiomics features are potential factors that can predict the outcome of thrombectomy in patients suffering from acute ischemic stroke, aiding in the decision between aspiration, mechanical, or combined thrombectomy procedure.

Assessment Precision of CT Perfusion Imaging in the Detection of Acute Ischemic Stroke: A Systematic Review and Meta-Analysis

Stroke, a prevalent medical emergency, comprises ischemic and hemorrhagic subtypes, with acute ischemic stroke (AIS) being a predominant type. The application of computed tomography perfusion (CTP) imaging has gained prominence due to its rapidity and accessibility in stroke evaluation. This study systematically reviews and conducts a meta-analysis of existing literature to assess the diagnostic accuracy of CTP in detecting AIS and predicting hemorrhagic transformation (HT). Employing Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, an extensive search was conducted across electronic databases and relevant radiology journals. Studies conducted between 2007 and 2023 that fulfilled predetermined inclusion criteria underwent quality assessment using the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS 2) tool. Cochrane diagnostic accuracy tools were used for data extraction. Thirteen studies involving a total of 1014 patients were included in the analysis. The diagnostic performance of CTP in predicting HT demonstrated high sensitivity (86.7%) and moderate specificity (77.8%), resulting in an overall accuracy of 79.1%. The negative predictive value (NPV) was notably high (92.9%), signifying its efficacy in excluding patients at risk of HT. The positive predictive value (PPV) was comparatively lower (60.3%), highlighting the need for clinical context when making thrombolysis decisions. The false positive rate was 16.2%, while the false negative rate was minimal (9.8%). Subgroup analysis underscored consistent sensitivity and specificity across diverse imaging metrics. The findings of this study emphasize the promising diagnostic accuracy of CTP imaging in predicting HT subsequent to AIS. This non-invasive technique can aid treatment decisions and patient management strategies. By effectively assessing perfusion status and offering predictive insights, CTP imaging improves stroke intervention choices, especially in identifying patients with a lower risk of HT.

Artificial Intelligence for Automated DWI/FLAIR Mismatch Assessment on Magnetic Resonance Imaging in Stroke: A Systematic Review

We conducted this Systematic Review to create an overview of the currently existing Artificial Intelligence (AI) methods for Magnetic Resonance Diffusion-Weighted Imaging (DWI)/Fluid-Attenuated Inversion Recovery (FLAIR)-mismatch assessment and to determine how well DWI/FLAIR mismatch algorithms perform compared to domain experts. We searched PubMed Medline, Ovid Embase, Scopus, Web of Science, Cochrane, and IEEE Xplore literature databases for relevant studies published between 1 January 2017 and 20 November 2022, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. We assessed the included studies using the Quality Assessment of Diagnostic Accuracy Studies 2 tool. Five studies fit the scope of this review. The area under the curve ranged from 0.74 to 0.90. The sensitivity and specificity ranged from 0.70 to 0.85 and 0.74 to 0.84, respectively. Negative predictive value, positive predictive value, and accuracy ranged from 0.55 to 0.82, 0.74 to 0.91, and 0.73 to 0.83, respectively. In a binary classification of ±4.5 h from stroke onset, the surveyed AI methods performed equivalent to or even better than domain experts. However, using the relation between time since stroke onset (TSS) and increasing visibility of FLAIR hyperintensity lesions is not recommended for the determination of TSS within the first 4.5 h. An AI algorithm on DWI/FLAIR mismatch assessment focused on treatment eligibility, outcome prediction, and consideration of patient-specific data could potentially increase the proportion of stroke patients with unknown onset who could be treated with thrombolysis.