پیش بینی نتایج سکته مغزی
Abstract
Graphical abstracts
۱٫ Introduction
۲٫ Materials and methods
۳٫ Results
۴٫ Discussion
۵٫ Conclusion
Contributors
Declaration of Competing Interest
Acknowledgements
Appendix C. Supplementary materials
Appendix A. List of Taiwan Stroke Registry Investigators
Appendix B. Details of Machine learning models
Appendix C
Research Data
Reference
Abstract
Introduction Being able to predict functional outcomes after a stroke is highly desirable for clinicians. This allows clinicians to set reasonable goals with patients and relatives, and to reach shared aftercare decisions for recovery or rehabilitation. The aim of this study was to apply various machine learning (ML) methods for 90-day stroke outcome predictions, using a nationwide disease registry. Methods This study used the Taiwan Stroke Registry (TSR) which has prospectively collected data from stroke patients since 2006. Three known ML models (support vector machine, random forest, and artificial neural network), and a hybrid artificial neural network were implemented and evaluated by 10-time repeated hold-out with 10-fold cross-validation. Results ML techniques present over 0.94 AUC in both ischemic and hemorrhagic stroke using preadmission and inpatient data. By adding follow-up data, the prediction ability improved to 0.97 AUC. We screened 206 clinical variables to identify 17 important features from the ischemic stroke dataset and 22 features from the hemorrhagic stroke dataset without losing much performance. Error analysis revealed that most prediction errors come from more severe stroke patients. Conclusion The study showed that ML techniques trained from large, cross-reginal registry datasets were able to predict functional outcome after stroke with high accuracy. The follow-up data is important which can further improve the predictive models’ performance. With similar performances among different ML techniques, the algorithm’s characteristics and performance on severe stroke patients will be the primary focus when we further develop inference models and artificial intelligence tools for potential medical.
Introduction
Stroke is the second leading cause of mortality in the world and the leading adult disability in developed countries [1, 2]. Many stroke survivors are left with various neurological deficits resulting in impaired quality of life of variable extent that has been a significant burden on patients, caregivers, and society [3]. More precise prediction of functional outcomes after a stroke may help clinicians in developing an appropriate long-term management plan. For example, plans based on better prediction of the extent of recovery with appropriate rehabilitative measures with patients’ domestic condition taken into consideration for reaching shared decisions with patients and family members [4-6]. Much effort has been devoted to determining predictors of functional outcome after stroke [6-8]. Several medical communities have created scores that can predict the patient’s functional outcome using data readily available at admission [9-11]. These scores use statistical analysis to identify the most relevant covariates from a set of pre-selected factors by domain experts. Recently, machine learning has become ubiquitous for solving complex problems in many scientific domains, especially in medical diagnosis or prognosis prediction [12, 13].