Optimization of Large Vessel Occlusion Detection in Acute Ischemic Stroke Using Machine Learning Methods

Abstract

The early detection of large-vessel occlusion (LVO) strokes is increasingly important as these patients are potential candidates for endovascular therapy, the availability of which is limited. Prehospital LVO detection scales mainly contain symptom variables only; however, recent studies revealed that other types of variables could be useful as well. Our aim was to comprehensively assess the predictive ability of several clinical variables for LVO prediction and to develop an optimal combination of them using machine learning tools. We have retrospectively analysed data from a prospectively collected multi-centre stroke registry. Data on 41 variables were collected and divided into four groups (baseline vital parameters/demographic data, medical history, laboratory values, and symptoms). Following the univariate analysis, the LASSO method was used for feature selection to select an optimal combination of variables, and various machine learning methods (random forest (RF), logistic regression (LR), elastic net method (ENM), and simple neural network (SNN)) were applied to optimize the performance of the model. A total of 526 patients were included. Several neurological symptoms were more common and more severe in the group of LVO patients. Atrial fibrillation (AF) was more common, and serum white blood cell (WBC) counts were higher in the LVO group, while systolic blood pressure (SBP) was lower among LVO patients. Using the LASSO method, nine variables were selected for modelling (six symptom variables, AF, chronic heart failure, and WBC count). When applying machine learning methods and 10-fold cross validation using the selected variables, all models proved to have an AUC between 0.736 (RF) and 0.775 (LR), similar to the performance of National Institutes of Health Stroke Scale (AUC: 0.790). Our study highlights that, although certain neurological symptoms have the best ability to predict an LVO, other variables (such as AF and CHF in medical history and white blood cell counts) should also be included in multivariate models to optimize their efficiency.

Systematic Review and Meta‐Analysis of Prehospital Machine Learning Scores as Screening Tools for Early Detection of Large Vessel Occlusion in Patients With Suspected Stroke

“…Eight studies met the inclusion criteria and were included (Figure 1 ). 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 …”

“…These studies included a total of 32 prehospital stroke scales and 21 ML models. Among them, 9 prehospital stroke scales 15 , 16 , 18 , 19 , 20 , 21 and 9 ML models 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 were eligible for meta‐analysis. Included studies were published from 2018 14 onwards and conducted globally (United States, 16 , 19 Japan, 15 , 20 China 14 , 21 (2 studies each) and 1 each from Hungary 18 and Taiwan 17 ).…”

“…Among them, 9 prehospital stroke scales 15 , 16 , 18 , 19 , 20 , 21 and 9 ML models 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 were eligible for meta‐analysis. Included studies were published from 2018 14 onwards and conducted globally (United States, 16 , 19 Japan, 15 , 20 China 14 , 21 (2 studies each) and 1 each from Hungary 18 and Taiwan 17 ). There was large variation in the total number of participants, from 268 19 to 19 580 21 .…”

An Automated Method of Causal Inference of the Underlying Cause of Death of Citizens

“…To comprehensively assess the predictive ability of several clinical variables for large-vessel occlusion (LVO) prediction, Tarkanyi et al retrospectively analyzed prospectively collected multicenter stroke registry data. After a univariate analysis, they used the LASSO method for feature selection to select an optimal combination of variables, and various machine learning methods (random forest (RF), logistic regression (LR), elastic net method (ENM), and simple neural network (SNN)) were applied to optimize the performance of the model [ 18 ]. Jafari et al used magnetic resonance imaging (MRI) and machine learning methods to diagnose the severity of brain tumors.…”

Enhancing Heart Disease Prediction through Ensemble Learning Techniques with Hyperparameter Optimization

“…Extensive medical research has identified various risk factors associated with heart disease, including physical inactivity, poor lifestyle choices, obesity, and unhealthy diet. Additionally, certain conditions, such as high blood pressure, smoking, family history, hypertension, stress, diabetes, and high cholesterol, can further elevate the risk of heart disease [7][8][9][10][11][12]. Given the complexity and multifactorial nature of heart disease, effective prediction and prevention strategies are essential for mitigating its impact.…”