Machine Learning-Based Model for Prediction of Hemorrhage Transformation in Acute Ischemic Stroke After Alteplase

Circulating miRNA profiles and the risk of hemorrhagic transformation after thrombolytic treatment of acute ischemic stroke: a pilot study

Background

Hemorrhagic transformation (HT) in acute ischemic stroke is likely to occur in patients treated with intravenous thrombolysis (IVT) and may lead to neurological deterioration and symptomatic intracranial hemorrhage (sICH). Despite the complex inclusion and exclusion criteria for IVT and some useful tools to stratify HT risk, sICH still occurs in approximately 6% of patients because some of the risk factors for this complication remain unknown.

Objective

This study aimed to explore whether there are any differences in circulating microRNA (miRNA) profiles between patients who develop HT after thrombolysis and those who do not.

Methods

Using qPCR, we quantified the expression of 84 miRNAs in plasma samples collected prior to thrombolytic treatment from 10 individuals who eventually developed HT and 10 patients who did not. For miRNAs that were downregulated (fold change (FC) <0.67) or upregulated (FC >1.5) with p < 0.10, we investigated the tissue specificity and performed KEGG pathway annotation using bioinformatics tools. Owing to the small patient sample size, instead of multivariate analysis with all major known HT risk factors, we matched the results with the admission NIHSS scores only.

Results

We observed trends towards downregulation of miR-1-3p, miR-133a-3p, miR-133b and miR-376c-3p, and upregulation of miR-7-5p, miR-17-3p, and miR-296-5p. Previously, the upregulated miR-7-5p was found to be highly expressed in the brain, whereas miR-1, miR-133a-3p and miR-133b appeared to be specific to the muscles and myocardium.

Conclusion

miRNA profiles tend to differ between patients who develop HT and those who do not, suggesting that miRNA profiling, likely in association with other omics approaches, may increase the current power of tools predicting thrombolysis-associated sICH in acute ischemic stroke patients. This study represents a free hypothesis-approach pilot study as a continuation from our previous work. Herein, we showed that applying mathematical analyses to extract information from raw big data may result in the identification of new pathophysiological pathways and may complete standard design works.

Risk prediction models for intracranial hemorrhage in acute ischemic stroke patients receiving intravenous alteplase treatment: a systematic review

Objectives

To identify and compare published models that use related factors to predict the risk of intracranial hemorrhage (ICH) in acute ischemic stroke patients receiving intravenous alteplase treatment.

Methods

Risk prediction models for ICH in acute ischemic stroke patients receiving intravenous alteplase treatment were collected from PubMed, Embase, Web of Science, and the Cochrane Library up to April 7, 2023. A meta-analysis was performed using Stata 13.0, and the included models were evaluated using the Prediction Model Risk of Bias Assessment Tool (PROBAST).

Results

A total of 656 references were screened, resulting in 13 studies being included. Among these, one was a prospective cohort study. Ten studies used internal validation; five studies used external validation, with two of them using both. The area under the receiver operating characteristic (ROC) curve for subjects reported in the models ranged from 0.68 to 0.985. Common predictors in the prediction models include National Institutes of Health Stroke Scale (NIHSS) (OR = 1.17, 95% CI 1.09-1.25, p < 0.0001), glucose (OR = 1.54, 95% CI 1.09-2.17, p < 0.05), and advanced age (OR = 1.50, 95% CI 1.15-1.94, p < 0.05), and the meta-analysis shows that these are independent risk factors. After PROBAST evaluation, all studies were assessed as having a high risk of bias but a low risk of applicability concerns.

Conclusion

This study systematically reviews available evidence on risk prediction models for ICH in acute ischemic stroke patients receiving intravenous alteplase treatment. Few models have been externally validated, while the majority demonstrate significant discriminative power.

A weakly supervised deep learning model integrating noncontrasted computed tomography images and clinical factors facilitates haemorrhagic transformation prediction after intravenous thrombolysis in acute ischaemic stroke patients

BACKGROUND

Haemorrhage transformation (HT) is a serious complication of intravenous thrombolysis (IVT) in acute ischaemic stroke (AIS). Accurate and timely prediction of the risk of HT before IVT may change the treatment decision and improve clinical prognosis. We aimed to develop a deep learning method for predicting HT after IVT for AIS using noncontrast computed tomography (NCCT) images.

METHODS

We retrospectively collected data from 828 AIS patients undergoing recombinant tissue plasminogen activator (rt-PA) treatment within a 4.5-h time window (n = 665) or of undergoing urokinase treatment within a 6-h time window (n = 163) and divided them into the HT group (n = 69) and non-HT group (n = 759). HT was defined based on the criteria of the European Cooperative Acute Stroke Study-II trial. To address the problems of indiscernible features and imbalanced data, a weakly supervised deep learning (WSDL) model for HT prediction was constructed based on multiple instance learning and active learning using admission NCCT images and clinical information in addition to conventional deep learning models. Threefold cross-validation and transfer learning were performed to confirm the robustness of the network. Of note, the predictive value of the commonly used scales in clinics associated with NCCT images (i.e., the HAT and SEDAN score) was also analysed and compared to measure the feasibility of our proposed DL algorithms.

RESULTS

Compared to the conventional DL and ML models, the WSDL model had the highest AUC of 0.799 (95% CI 0.712-0.883). Significant differences were observed between the WSDL model and five ML models (P < 0.05). The prediction performance of the WSDL model outperforms the HAT and SEDAN scores at the optimal operating point (threshold = 1.5). Further subgroup analysis showed that the WSDL model performed better for symptomatic intracranial haemorrhage (AUC = 0.833, F1 score = 0.909).

CONCLUSIONS

Our WSDL model based on NCCT images had relatively good performance for predicting HT in AIS and may be suitable for assisting in clinical treatment decision-making.

Application and risk prediction of thrombolytic therapy in cardio-cerebrovascular diseases: a review

Cardiocerebrovascular diseases (CVDs) are the leading cause of death worldwide, consuming huge healthcare budget. For CVD patients, the prompt assessment and appropriate administration is the crux to save life and improve prognosis. Thrombolytic therapy, as a non-invasive approach to achieve recanalization, is the basic component of CVD treatment. Still, there are risks that limits its application. The objective of this review is to give an introduction on the utilization of thrombolytic therapy in cardiocerebrovascular blockage diseases, including coronary heart disease and ischemic stroke, and to review the development in risk assessment of thrombolytic therapy, comparing the performance of traditional scales and novel artificial intelligence-based risk assessment models.

Machine Learning Models for Predicting Early Neurological Deterioration and Risk Classification of Acute Ischemic Stroke

This study aimed to create machine learning models for predicting early neurological deterioration and risk classification in acute ischemic stroke (AIS) before intravenous thrombolysis (IVT). The study included 704 AIS patients categorized into END and non-END groups. The least absolute shrinkage and selection operator (LASSO) regression was employed to select the best predictors from clinical indicators, leading to the creation of Model 1. Univariate and multivariate logistic regression analyses identified independent predictive factors for END from inflammatory cell ratios. These factors were combined with clinical indicators, forming Model 2. Receiver operating characteristic (ROC) curves assessed the models' predictive performance. Key variables for Model 1 included the NIHSS score, systolic blood pressure, and lymphocyte percentage. Neutrophil-to-Lymphocyte ratio, Platelet-to-Neutrophil ratio, and Platelet-to-Lymphocyte ratio independently predicted END. Model 1 exhibited moderate predictive ability (AUC 0.721 in training, AUC 0.635 in test). Model 2, which integrated clinical indicators and inflammatory cell ratios, demonstrated strong performance in both training (AUC 0.862) and test (AUC 0.816). Machine learning models, combining clinical indicators and inflammatory cell ratios before IVT, accurately predict END and associated risk in AIS.