Jude P J Savarraj1, Georgene W Hergenroeder1, Liang Zhu1, Tiffany Chang1, Soojin Park1, Murad Megjhani1, Farhaan S Vahidy1, Zhongming Zhao1, Ryan S Kitagawa1, H Alex Choi2. 1. From the Departments of Neurosurgery (J.P.J.S., G.W.H., T.C., R.S.K., A.C.), Internal Medicine (L.Z.), and Neurology (F.S.V.), McGovern Medical School, Center for Precision Health, School of Biomedical Informatics (Z.Z.), and Human Genetics Center, School of Public Health (Z.Z.), The University of Texas Health Science Center at Houston; and Department of Neurology (S.P., M.M.), Columbia University, NY. 2. From the Departments of Neurosurgery (J.P.J.S., G.W.H., T.C., R.S.K., A.C.), Internal Medicine (L.Z.), and Neurology (F.S.V.), McGovern Medical School, Center for Precision Health, School of Biomedical Informatics (Z.Z.), and Human Genetics Center, School of Public Health (Z.Z.), The University of Texas Health Science Center at Houston; and Department of Neurology (S.P., M.M.), Columbia University, NY. Huimahn.A.Choi@uth.tmc.edu jude.p.savarraj@uth.tmc.edu.
Abstract
OBJECTIVE: To determine whether machine learning (ML) algorithms can improve the prediction of delayed cerebral ischemia (DCI) and functional outcomes after subarachnoid hemorrhage (SAH). METHODS: ML models and standard models (SMs) were trained to predict DCI and functional outcomes with data collected within 3 days of admission. Functional outcomes at discharge and at 3 months were quantified using the modified Rankin Scale (mRS) for neurologic disability (dichotomized as good [mRS ≤ 3] vs poor [mRS ≥ 4] outcomes). Concurrently, clinicians prospectively prognosticated 3-month outcomes of patients. The performance of ML, SMs, and clinicians were retrospectively compared. RESULTS: DCI status, discharge, and 3-month outcomes were available for 399, 393, and 240 participants, respectively. Prospective clinician (an attending, a fellow, and a nurse) prognostication of 3-month outcomes was available for 90 participants. ML models yielded predictions with the following area under the receiver operating characteristic curve (AUC) scores: 0.75 ± 0.07 (95% confidence interval [CI] 0.64-0.84) for DCI, 0.85 ± 0.05 (95% CI 0.75-0.92) for discharge outcome, and 0.89 ± 0.03 (95% CI 0.81-0.94) for 3-month outcome. ML outperformed SMs, improving AUC by 0.20 (95% CI -0.02 to 0.4) for DCI, by 0.07 ± 0.03 (95% CI -0.0018 to 0.14) for discharge outcomes, and by 0.14 (95% CI 0.03-0.24) for 3-month outcomes and matched physician's performance in predicting 3-month outcomes. CONCLUSION: ML models significantly outperform SMs in predicting DCI and functional outcomes and has the potential to improve SAH management.
OBJECTIVE: To determine whether machine learning (ML) algorithms can improve the prediction of delayed cerebral ischemia (DCI) and functional outcomes after subarachnoid hemorrhage (SAH). METHODS: ML models and standard models (SMs) were trained to predict DCI and functional outcomes with data collected within 3 days of admission. Functional outcomes at discharge and at 3 months were quantified using the modified Rankin Scale (mRS) for neurologic disability (dichotomized as good [mRS ≤ 3] vs poor [mRS ≥ 4] outcomes). Concurrently, clinicians prospectively prognosticated 3-month outcomes of patients. The performance of ML, SMs, and clinicians were retrospectively compared. RESULTS: DCI status, discharge, and 3-month outcomes were available for 399, 393, and 240 participants, respectively. Prospective clinician (an attending, a fellow, and a nurse) prognostication of 3-month outcomes was available for 90 participants. ML models yielded predictions with the following area under the receiver operating characteristic curve (AUC) scores: 0.75 ± 0.07 (95% confidence interval [CI] 0.64-0.84) for DCI, 0.85 ± 0.05 (95% CI 0.75-0.92) for discharge outcome, and 0.89 ± 0.03 (95% CI 0.81-0.94) for 3-month outcome. ML outperformed SMs, improving AUC by 0.20 (95% CI -0.02 to 0.4) for DCI, by 0.07 ± 0.03 (95% CI -0.0018 to 0.14) for discharge outcomes, and by 0.14 (95% CI 0.03-0.24) for 3-month outcomes and matched physician's performance in predicting 3-month outcomes. CONCLUSION: ML models significantly outperform SMs in predicting DCI and functional outcomes and has the potential to improve SAH management.
Authors: Eric S Rosenthal; Siddharth Biswal; Sahar F Zafar; Kathryn L O'Connor; Sophia Bechek; Apeksha V Shenoy; Emily J Boyle; Mouhsin M Shafi; Emily J Gilmore; Brandon P Foreman; Nicolas Gaspard; Thabele M Leslie-Mazwi; Jonathan Rosand; Daniel B Hoch; Cenk Ayata; Sydney S Cash; Andrew J Cole; Aman B Patel; M Brandon Westover Journal: Ann Neurol Date: 2018-05-16 Impact factor: 10.422
Authors: Mark J Hannon; Francis M Finucane; Mark Sherlock; Amar Agha; Christopher J Thompson Journal: J Clin Endocrinol Metab Date: 2012-02-22 Impact factor: 5.958
Authors: Mervyn D I Vergouwen; Marinus Vermeulen; Jan van Gijn; Gabriel J E Rinkel; Eelco F Wijdicks; J Paul Muizelaar; A David Mendelow; Seppo Juvela; Howard Yonas; Karel G Terbrugge; R Loch Macdonald; Michael N Diringer; Joseph P Broderick; Jens P Dreier; Yvo B W E M Roos Journal: Stroke Date: 2010-08-26 Impact factor: 7.914
Authors: Emmanuel Carrera; J Michael Schmidt; Mauro Oddo; Luis Fernandez; Jan Claassen; David Seder; Kiwon Lee; Neeraj Badjatia; E Sander Connolly; Stephan A Mayer Journal: Neurosurgery Date: 2009-08 Impact factor: 4.654
Authors: Markus Holling; Astrid Jeibmann; Joachim Gerss; Bernhard R Fischer; Hansdetlef Wassmann; Werner Paulus; Martin Hasselblatt; Friedrich K Albert Journal: J Neurosurg Date: 2009-03 Impact factor: 5.115
Authors: Sung-Ho Ahn; Jude P Savarraj; Mubashir Pervez; Wesley Jones; Jin Park; Sang-Beom Jeon; Sun U Kwon; Tiffany R Chang; Kiwon Lee; Dong H Kim; Arthur L Day; H Alex Choi Journal: Neurosurgery Date: 2018-07-01 Impact factor: 4.654
Authors: J Joep van der Harst; Gert-Jan R Luijckx; Jan Willem J Elting; Thijs Lammers; Reinoud P H Bokkers; Walter M van den Bergh; Omid S Eshghi; Jan D M Metzemaekers; Rob J M Groen; Aryan Mazuri; Nic J G M Veeger; J Marc C van Dijk; Maarten Uyttenboogaart Journal: Eur J Neurol Date: 2021-11-07 Impact factor: 6.288