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

Lea Fast; Uchralt Temuulen; Kersten Villringer; Anna Kufner; Huma Fatima Ali; Eberhard Siebert; Shufan Huo; Sophie K Piper; Pia Sophie Sperber; Thomas Liman; Matthias Endres; Kerstin Ritter

doi:10.3389/fneur.2023.1114360

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

Front Neurol. 2023 Feb 21:14:1114360. doi: 10.3389/fneur.2023.1114360. eCollection 2023.

Authors

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}, Kerstin Ritter^{1

15}

Affiliations

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

Abstract

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.

Keywords: cognitive impairment; functional outcome; machine learning; mortality; outcome prediction; post-stroke depression; stroke.

Grants and funding

We acknowledge support from the German Research Foundation (DFG, 389563835; 402170461-TRR 265; 414984028-CRC 1404; 42075332-RU 5187) and the Manfred and Ursula-Müller Stiftung. ME received funding from DFG under Germany's Excellence Strategy–EXC-2049–390688087, Collaborative Research Center ReTune TRR 295-424778381, Bundesministerium für Bildung und Forschung (BMBF), Deutsches Zentrum für Neurodegenerative Erkrankungen (DZNE), Deutsches Zentrum für Herz-Kreislauferkrankungen (DZHK), EU, Corona Foundation, and Fondation Leducq.