International Journal of Electronics and Communications Systems · 2021 · 36 citations · 17 references
ComputingBiomedical InformaticsHealth InformaticsFeature EngineeringMachine LearningPredictive AnalyticsClassificationData PreprocessingDecision TreesEnsemble LearningPredictive Health ModelingStroke RecognitionClassification AlgorithmsLogistic RegressionMachine Learning ClassificationMachine Learning For HealthcareMachine Learning Prediction ModelsMortality PredictionRandom Forest ClassificationRandom Forests
Stroke is one of the fatal brain diseases that cause death in 3 to 10 hours. However, most stroke mortality can be prevented by identifying the nature of the stroke and reacting to it promptly through smart health systems. In this paper, a machine learning model is approached for predicting the existence of stroke of a patient where the Random forest classifier outperforms the state-of-the-art models, including Logistic Regression, Decision Tree Classifier (DTC), K-NN. We conduct the experiments on datasets which has 5110 observations with 12 attributes. We also applied EDA for preprocessing and feature techniques for balancing the datasets. Finally, a cloud-based mobile app collects user data to analyze and provide the possibility of stroke for alerting the person with the accuracy of precision 96 percent, recall 96 percent, and F1-score 96 percent. This user-friendly system can be a lifesaver as the person gets an essential warning very easily by providing very little information from anywhere with a mobile device.
17
Prediction of stroke outcome with echoplanar perfusion- and diffusion-weighted MRI
P. Alan Barber, David Darby, Patricia Desmond et al. · Neurology · 1998 · 562 citations
Takayoshi Ohkubo, Kei Asayama, Masahiro Kikuya et al. · Journal of Hypertension · 2004 · 283 citations · Full text
An integrated machine learning approach to stroke prediction
Aditya Khosla, Yu Cao, Cliff Chiung-Yu Lin et al. · 2010 · 202 citations