Machine Learning-Based Model for Prediction of Outcomes in Acute Stroke

Abstract

Background and Purpose- Thepredictionof long-termoutcomesin ischemicstrokepatients may be useful in treatment decisions.Machinelearning techniques are being increasingly adapted for use in the medical field because of their high accuracy. This study investigated the applicability ofmachinelearning techniques topredictlong-termoutcomesin ischemicstrokepatients. Methods- This was a retrospective study using a prospective cohort that enrolled patients withacuteischemicstroke. Favorableoutcomewas defined as modified Rankin Scale score 0, 1, or 2 at 3 months. We developed 3machinelearningmodels(deep neural network, random forest, and logistic regression) and compared theirpredictability. To evaluate the accuracy of themachinelearningmodels, we also compared them to theAcute StrokeRegistry and Analysis of Lausanne (ASTRAL) score. Results- A total of 2604 patients were included in this study, and 2043 (78%) of them had favorableoutcomes. The area under the curve for the deep neural networkmodelwas significantly higher than that of the ASTRAL score (0.888 versus 0.839; P<0.001), while the areas under the curves of the random forest (0.857; P=0.136) and logistic regression (0.849; P=0.413)modelswere not significantly higher than that of the ASTRAL score. Using only the 6 variables that are used for the ASTRAL score, the performance of themachinelearningmodelsdid not significantly differ from that of the ASTRAL score. Conclusions-Machinelearning algorithms, particularly the deep neural network, can improve thepredictionof long-termoutcomesin ischemicstrokepatients.