Predicting short-term outcomes in atrial-fibrillation-related stroke using machine learning
BackgroundPrognostic prediction and the identification of prognostic factors are critical during the early period of atrial-fibrillation (AF)-related strokes as AF is associated with poor outcomes in stroke patients.MethodsTwo independent datasets, namely, the Korean Atrial Fibrillation Evaluation R...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-11-01
|
| Series: | Frontiers in Neurology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fneur.2023.1243700/full |
| _version_ | 1797632960620396544 |
|---|---|
| author | Eun-Tae Jeon Seung Jin Jung Tae Young Yeo Woo-Keun Seo Jin-Man Jung Jin-Man Jung |
| author_facet | Eun-Tae Jeon Seung Jin Jung Tae Young Yeo Woo-Keun Seo Jin-Man Jung Jin-Man Jung |
| author_sort | Eun-Tae Jeon |
| collection | DOAJ |
| description | BackgroundPrognostic prediction and the identification of prognostic factors are critical during the early period of atrial-fibrillation (AF)-related strokes as AF is associated with poor outcomes in stroke patients.MethodsTwo independent datasets, namely, the Korean Atrial Fibrillation Evaluation Registry in Ischemic Stroke Patients (K-ATTENTION) and the Korea University Stroke Registry (KUSR), were used for internal and external validation, respectively. These datasets include common variables such as demographic, laboratory, and imaging findings during early hospitalization. Outcomes were unfavorable functional status with modified Rankin scores of 3 or higher and mortality at 3 months. We developed two machine learning models, namely, a tree-based model and a multi-layer perceptron (MLP), along with a baseline logistic regression model. The area under the receiver operating characteristic curve (AUROC) was used as the outcome metric. The Shapley additive explanation (SHAP) method was used to evaluate the contributions of variables.ResultsMachine learning models outperformed logistic regression in predicting both outcomes. For 3-month unfavorable outcomes, MLP exhibited significantly higher AUROC values of 0.890 and 0.859 in internal and external validation sets, respectively, than those of logistic regression. For 3-month mortality, both machine learning models exhibited significantly higher AUROC values than the logistic regression for internal validation but not for external validation. The most significant predictor for both outcomes was the initial National Institute of Health and Stroke Scale.ConclusionThe explainable machine learning model can reliably predict short-term outcomes and identify high-risk patients with AF-related strokes. |
| first_indexed | 2024-03-11T11:45:06Z |
| format | Article |
| id | doaj.art-ee17a6ab980b41d3a1d038fe926eda7f |
| institution | Directory Open Access Journal |
| issn | 1664-2295 |
| language | English |
| last_indexed | 2024-03-11T11:45:06Z |
| publishDate | 2023-11-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neurology |
| spelling | doaj.art-ee17a6ab980b41d3a1d038fe926eda7f2023-11-09T16:17:51ZengFrontiers Media S.A.Frontiers in Neurology1664-22952023-11-011410.3389/fneur.2023.12437001243700Predicting short-term outcomes in atrial-fibrillation-related stroke using machine learningEun-Tae Jeon0Seung Jin Jung1Tae Young Yeo2Woo-Keun Seo3Jin-Man Jung4Jin-Man Jung5Department of Neurology, Korea University Ansan Hospital, Korea University College of Medicine, Ansan, Republic of KoreaDepartment of Family Medicine, Gimpo Woori Hospital, Gimpo, Republic of KoreaDepartment of Neurology, Korea University Ansan Hospital, Korea University College of Medicine, Ansan, Republic of KoreaDepartment of Neurology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of KoreaDepartment of Neurology, Korea University Ansan Hospital, Korea University College of Medicine, Ansan, Republic of KoreaKorea University Zebrafish Translational Medical Research Center, Ansan, Republic of KoreaBackgroundPrognostic prediction and the identification of prognostic factors are critical during the early period of atrial-fibrillation (AF)-related strokes as AF is associated with poor outcomes in stroke patients.MethodsTwo independent datasets, namely, the Korean Atrial Fibrillation Evaluation Registry in Ischemic Stroke Patients (K-ATTENTION) and the Korea University Stroke Registry (KUSR), were used for internal and external validation, respectively. These datasets include common variables such as demographic, laboratory, and imaging findings during early hospitalization. Outcomes were unfavorable functional status with modified Rankin scores of 3 or higher and mortality at 3 months. We developed two machine learning models, namely, a tree-based model and a multi-layer perceptron (MLP), along with a baseline logistic regression model. The area under the receiver operating characteristic curve (AUROC) was used as the outcome metric. The Shapley additive explanation (SHAP) method was used to evaluate the contributions of variables.ResultsMachine learning models outperformed logistic regression in predicting both outcomes. For 3-month unfavorable outcomes, MLP exhibited significantly higher AUROC values of 0.890 and 0.859 in internal and external validation sets, respectively, than those of logistic regression. For 3-month mortality, both machine learning models exhibited significantly higher AUROC values than the logistic regression for internal validation but not for external validation. The most significant predictor for both outcomes was the initial National Institute of Health and Stroke Scale.ConclusionThe explainable machine learning model can reliably predict short-term outcomes and identify high-risk patients with AF-related strokes.https://www.frontiersin.org/articles/10.3389/fneur.2023.1243700/fullatrial fibrilationmachine learningoutcomeprediction modelischemic stroke |
| spellingShingle | Eun-Tae Jeon Seung Jin Jung Tae Young Yeo Woo-Keun Seo Jin-Man Jung Jin-Man Jung Predicting short-term outcomes in atrial-fibrillation-related stroke using machine learning Frontiers in Neurology atrial fibrilation machine learning outcome prediction model ischemic stroke |
| title | Predicting short-term outcomes in atrial-fibrillation-related stroke using machine learning |
| title_full | Predicting short-term outcomes in atrial-fibrillation-related stroke using machine learning |
| title_fullStr | Predicting short-term outcomes in atrial-fibrillation-related stroke using machine learning |
| title_full_unstemmed | Predicting short-term outcomes in atrial-fibrillation-related stroke using machine learning |
| title_short | Predicting short-term outcomes in atrial-fibrillation-related stroke using machine learning |
| title_sort | predicting short term outcomes in atrial fibrillation related stroke using machine learning |
| topic | atrial fibrilation machine learning outcome prediction model ischemic stroke |
| url | https://www.frontiersin.org/articles/10.3389/fneur.2023.1243700/full |
| work_keys_str_mv | AT euntaejeon predictingshorttermoutcomesinatrialfibrillationrelatedstrokeusingmachinelearning AT seungjinjung predictingshorttermoutcomesinatrialfibrillationrelatedstrokeusingmachinelearning AT taeyoungyeo predictingshorttermoutcomesinatrialfibrillationrelatedstrokeusingmachinelearning AT wookeunseo predictingshorttermoutcomesinatrialfibrillationrelatedstrokeusingmachinelearning AT jinmanjung predictingshorttermoutcomesinatrialfibrillationrelatedstrokeusingmachinelearning AT jinmanjung predictingshorttermoutcomesinatrialfibrillationrelatedstrokeusingmachinelearning |