Machine learning models improve prediction of large vessel occlusion and mechanical thrombectomy candidacy in acute ischemic stroke. (September 2021)

• Record Type: Journal Article
• Authors: Thomas, Shon; de la Pena, Paula; Butler, Liam; Akbilgic, Oguz; Heiferman, Daniel M.; Garg, Ravi; Gill, Rick; Serrone, Joseph C.
• Is Part Of: Journal of clinical neuroscience. Volume 91(2021)
• Page Start: 383
• Page End: 390
• Publication Date: 2021-09
• Subjects: Acute ischemic stroke -- Large vessel occlusion -- Mechanical thrombectomy -- Machine learning model -- Stroke scale; LVO Large Vessel Occlusion -- MT Mechanical Thrombectomy -- AUC Area Under Curve -- tPA tissue Plasminogen Activator -- NIHSS National Institute of Health Stroke Scale -- CPSSS Cincinnati Prehospital Stroke Severity Scale -- RACE Rapid Arterial oCclusion Evaluate -- PASS Prehospital Acute Stroke Severity scale -- FAST-ED Field Assessment Stroke Triage for Emergency Destination -- G-FAST Gaze-Face-Arm-Speech-Time -- EMSA Emergency Medical Stroke Assessment -- 3I-SS 3-Item Stroke Scale -- FPSS Finnish Prehospital Stroke Scale -- SAVE Speech Arm Vision Eyes -- VAN Vision Aphasia Neglect -- XGB eXtreme Gradient Boosting -- RF Random Forest -- CART Classification And Regression Trees -- ROC Receiver Operating Characteristic; Brain -- Surgery -- Periodicals; Neurosciences -- Periodicals; Nervous system -- Surgery -- Periodicals; Brain -- surgery -- Periodicals; Neurosurgical Procedures -- Periodicals; Neurosciences -- Periodicals; Electronic journals; 616.8
• DOI: 10.1016/j.jocn.2021.07.021
• Languages: English
• ISSNs: 0967-5868

Abstract

Highlights: Early identification of large vessel occlusions (LVO) and timely recanalization are paramount to improved clinical outcomes in acute ischemic stroke. A stroke assessment that maximizes sensitivity and specificity for LVOs is needed to identify these cases and not overburden the health system with unnecessary transfers. Machine learning, used for predictive modeling in many aspects of stroke care, may have potential in predicting LVO presence and mechanical thrombectomy (MT) candidacy. Ischemic stroke patients treated at Loyola University Medical Center from July 2018 to June 2019 (N = 286) were included. Thirty-five clinical and demographic variables were analyzed using machine learning algorithmsto build models predictive of LVO presence and MT candidacy by area of the curve (AUC) analysis. The best performing model was compared with prior stroke scales. When using all 35 variables, RF best predicted LVO presence (AUC = 0.907 ± 0.856–0.957) while logistic regression best predicted MT candidacy (AUC = 0.930 ± 0.886–0.974). When compact models were evaluated, a 10-feature RF model best predicted LVO (AUC = 0.841 ± 0.778–0.904) and an 8-feature RF model best predicted MT candidacy (AUC = 0.862 ± 0.782–0.942). The compact RF models had SN Sn, Sp, NPV and PPV of 0.81, 0.87, 0.92, 0.72 for LVO and 0.87, 0.97, 0.97, 0.86 for MT, respectively. The 10-feature RF model was superior at predicting LVO to all previous stroke scales (AUC 0.944 vs 0.759–0.878) and the 8-feature RF model was superior at predicting MT (AUC 0.970 vs 0.746–0.834). Random forest machine learning models utilizing clinical and demographic variables predicts LVO presence and MT candidacy with a high degree of accuracy. Further validation of this strategy for triage of stroke patients requires prospective and external validation. Abstract: Background and Purpose: Early identification of large vessel occlusions (LVO) and timely recanalization are paramount to improved clinical outcomes in acute ischemic stroke. A stroke assessment that maximizes sensitivity and specificity for LVOs is needed to identify these cases and not overburden the health system with unnecessary transfers. Machine learning techniques are being used for predictive modeling in many aspects of stroke care and may have potential in predicting LVO presence and mechanical thrombectomy (MT) candidacy. Methods: Ischemic stroke patients treated at Loyola University Medical Center from July 2018 to June 2019 (N = 286) were included. Thirty-five clinical and demographic variables were analyzed using machine learning algorithms, including logistic regression, extreme gradient boosting, random forest (RF), and decision trees to build models predictive of LVO presence and MT candidacy by area of the curve (AUC) analysis. The best performing model was compared with prior stroke scales. Results: When using all 35 variables, RF best predicted LVO presence (AUC = 0.907 ± 0.856–0.957) while logistic regression best predicted MT candidacy (AUC = 0.930 ± 0.886–0.974). When compact models were evaluated, a 10-feature RF model best predicted LVO (AUC = 0.841 ± 0.778–0.904) and an 8-feature RF model best predicted MT candidacy (AUC = 0.862 ± 0.782–0.942). The compact RF models had sensitivity, specificity, negative predictive value and positive predictive value of 0.81, 0.87, 0.92, 0.72 for LVO and 0.87, 0.97, 0.97, 0.86 for MT, respectively. The 10-feature RF model was superior at predicting LVO to all previous stroke scales (AUC 0.944 vs 0.759–0.878) and the 8-feature RF model was superior at predicting MT (AUC 0.970 vs 0.746–0.834). Conclusion: Random forest machine learning models utilizing clinical and demographic variables predicts LVO presence and MT candidacy with a high degree of accuracy in an ischemic stroke cohort. Further validation of this strategy for triage of stroke patients requires prospective and external validation.