• Haydn Hoffman, Jacob Wood, John R Cote, Muhammad S Jalal, Fadar O Otite, Hesham E Masoud, and Grahame C Gould.
• Department of Neurosurgery, State University of New York Upstate Medical University, Syracuse, New York, USA. Electronic address: hhoffman@semmes-murphey.com.
• World Neurosurg. 2024 Feb 1; 182: e137e154e137-e154.

ObjectiveMechanical thrombectomy (MT) improves outcomes in patients with LVO but many still experience mortality or severe disability. We sought to develop machine learning (ML) models that predict 90-day outcomes after MT for LVO.MethodsConsecutive patients who underwent MT for LVO between 2015-2021 at a Comprehensive Stroke Center were reviewed. Outcomes included 90-day favorable functional status (mRS 0-2), severe disability (mRS 4-6), and mortality. ML models were trained for each outcome using prethrombectomy data (pre) and with thrombectomy data (post).ResultsThree hundred and fifty seven patients met the inclusion criteria. After model screening and hyperparameter tuning the top performing ML model for each outcome and timepoint was random forest (RF). Using only prethrombectomy features, the AUCs for the RFpre models were 0.73 (95% CI 0.62-0.85) for favorable functional status, 0.77 (95% CI 0.65-0.86) for severe disability, and 0.78 (95% CI 0.64-0.88) for mortality. All of these were better than a standard statistical model except for favorable functional status. Each RF model outperformed Pre, SPAN-100, THRIVE, and HIAT scores (P < 0.0001 for all). The most predictive features were premorbid mRS, age, and NIHSS. Incorporating MT data, the AUCs for the RFpost models were 0.80 (95% CI 0.67-0.90) for favorable functional status, 0.82 (95% CI 0.69-0.91) for severe disability, and 0.71 (95% CI 0.55-0.84) for mortality.ConclusionsRF models accurately predicted 90-day outcomes after MT and performed better than standard statistical and clinical prediction models.Copyright © 2023 Elsevier Inc. All rights reserved.

14 keywords...

hide…