Machine learning-based prediction of clinical outcomes after first-ever ischemic stroke

Will you please stop with this fucking useless research on predicting failure to recover! And just do what survivors want! EXACT 100% RECOVERY PROTOCOLS! I 'd have you all fired.

Hope you're OK with that when you are the 1 in 4 per WHO that has a stroke, you'll want full recovery, this doesn't get you there.

 

Machine learning-based prediction of clinical outcomes after first-ever ischemic stroke

Lea Fast¹, Uchralt Temuulen², Kersten Villringer², Anna Kufner^2,3,4, Huma Fatima Ali⁵, Eberhard Siebert⁶, Shufan Huo^2,4,7, Sophie K. Piper^3,8,9, Pia Sophie Sperber^2,10,11,12, Thomas Liman^2,7,13,14, Matthias Endres^{2,3,4,7,10,13} and Kerstin Ritter^1,15*

• ¹Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin and Humboldt-Universität zu Berlin, Department of Psychiatry and Psychotherapy, Berlin, Germany
• ²Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin and Humboldt-Universität zu Berlin, Center for Stroke Research Berlin (CSB), Berlin, Germany
• ³Berlin Institute of Health at Charité - Universitätsmedizin Berlin, Berlin, Germany
• ⁴Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin and Humboldt-Universität zu Berlin, Department of Neurology with Experimental Neurology, Berlin, Germany
• ⁵Berlin School of Mind and Brain, Humboldt-Universität zu Berlin, Berlin, Germany
• ⁶Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin and Humboldt-Universität zu Berlin, Department of Neuroradiology, Berlin, Germany
• ⁷German Center for Cardiovascular Research (Deutsches Zentrum für Herz-Kreislauferkrankungen, DZHK), Partner Site Berlin, Berlin, Germany
• ⁸Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin and Humboldt-Universität zu Berlin, Institute of Biometry and Clinical Epidemiology, Berlin, Germany
• ⁹Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin and Humboldt-Universität zu Berlin, Institute of Medical Informatics, Berlin, Germany
• ¹⁰Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin and Humboldt-Universität zu Berlin, NeuroCure Cluster of Excellence, NeuroCure Clinical Research Center (NCRC), Berlin, Germany
• ¹¹Experimental and Clinical Research Center, A Cooperation Between the Max Delbrück Center for Molecular Medicine in the Helmholtz Association and Charité – Universitätsmedizin Berlin, Berlin, Germany
• ¹²Max Delbrück Center for Molecular Medicine in the Helmholtz Association (MDC), Berlin, Germany
• ¹³German Center for Neurodegenerative Diseases (Deutsches Zentrum für Neurodegenerative Erkrankungen, DZNE), Partner Site Berlin, Berlin, Germany
• ¹⁴Department of Neurology, Evangelical Hospital Oldenburg, Carl von Ossietzky-University, Oldenburg, Germany
• ¹⁵Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin and Humboldt-Universität zu Berlin, Bernstein Center for Computational Neuroscience (BCCN), Berlin, Germany

Background: Accurate prediction of clinical outcomes in individual patients following acute stroke is vital for healthcare providers to optimize treatment strategies and plan further patient care. Here, we use advanced machine learning (ML) techniques to systematically compare the prediction of functional recovery, cognitive function, depression, and mortality of first-ever ischemic stroke patients and to identify the leading prognostic factors.

Methods: We predicted clinical outcomes for 307 patients (151 females, 156 males; 68 ± 14 years) from the PROSpective Cohort with Incident Stroke Berlin study using 43 baseline features. Outcomes included modified Rankin Scale (mRS), Barthel Index (BI), Mini-Mental State Examination (MMSE), Modified Telephone Interview for Cognitive Status (TICS-M), Center for Epidemiologic Studies Depression Scale (CES-D) and survival. The ML models included a Support Vector Machine with a linear kernel and a radial basis function kernel as well as a Gradient Boosting Classifier based on repeated 5-fold nested cross-validation. The leading prognostic features were identified using Shapley additive explanations.

Results: The ML models achieved significant prediction performance for mRS at patient discharge and after 1 year, BI and MMSE at patient discharge, TICS-M after 1 and 3 years and CES-D after 1 year. Additionally, we showed that National Institutes of Health Stroke Scale (NIHSS) was the top predictor for most functional recovery outcomes as well as education for cognitive function and depression.

Conclusion: Our machine learning analysis successfully demonstrated the ability to predict clinical outcomes after first-ever ischemic stroke and identified the leading prognostic factors that contribute to this prediction.

1. Introduction

Stroke is the second most common cause of death and a major cause of disability on a worldwide scale (1). It occurs when the blood supply to brain tissue is interrupted by either blockage (ischaemic stroke) or bleeding caused by rupture of cerebral blood vessels (haemorrhagic stroke) ultimately resulting in irreversible neuronal death (2). The incidence of stroke is set to rise due to the demographic shift affecting populations across the globe (3). Thus, it is paramount to identify parameters that can aid in accurate prediction of long-term clinical outcome post-stroke.

In recent years the move toward electronic health records and the application of machine learning (ML) techniques in the medical research field have opened new frontiers of personalized medicine and decision support. The key advantage is that—in contrast to traditional statistical analyses—not only can predictors and biomarkers be identified on a group level, but ML techniques also enable prediction on an individual patient level. In other words, the outcome for a single patients can be predicted by considering a vast array of variables (4). Numerous studies have successfully demonstrated the ability of ML models to predict specific clinical outcomes after stroke with remarkable accuracy and identified leading baseline factors that carry high prognostic value (5–8). Most studies so far have focused on the prediction of the modified Rankin Scale (mRS) (9) as it is the gold standard for determining functional recovery after stroke. While there are some studies investigating the ML-based prediction of the Barthel Index (BI) (10) and Modified Telephone Interview for Cognitive Status (TICS-M) (11), research regarding the Center for Epidemiologic Studies Depression Scale (CES-D) (12) and Mini-Mental State Examination (MMSE) (13) is sparse. In addition, the heterogeneity of ML techniques, clinical outcomes and datasets used in these studies makes it difficult to assess the broader implications of their findings (4).

The primary aim of the present study was therefore to conduct a systematic comparison of ML-based outcome prediction after first-ever ischemic stroke featuring measures of functional recovery (mRS, BI), cognitive function (MMSE, TICS-M), depression (CES-D), and mortality. The analysis was based on three powerful ML models and an array of baseline features including demographic, clinical, serological and MRI variables. As a secondary aim, we set out to identify to the key prognostic markers for each outcome using state-of-the-art visualization techniques.

More at link.