ABSTRACT :
A stroke is a medical condition in which poor blood flow to the brain results in cell death. It is now a day a leading cause of death all over the world. Several risk factors believe to be related to the cause of stroke has been found by inspecting the affected individuals. Using these risk factors, a number of works have been carried out for predicting and classifying stroke diseases. Most of the models are based on data mining and machine learning algorithms. In this work, we have used four machine learning algorithms to detect the type of stroke that can possibly occur or occurred form a person’s physical state and medical report data. We have collected a good number of entries from the hospitals and use them to solve our problem. The classification result shows that the result is satisfactory and can be used in real time medical report. We believe that machine learning algorithms can help better understanding of diseases and can be a good healthcare companion. Index Terms—Stroke, machine learning, WEKA, Naive Bayes, J48, k-NN, Random Forest.
EXISTING SYSTEM :
A stroke occurs due to poor blood flow to the brain which results in cell death. Two main types of stroke are ischemic stroke and hemorrhagic stroke. Ischemic stroke occurs due to lack of blood flow and hemorrhagic stroke occurs due to bleeding [1]. Another type of stroke is transient ischemic attack. Ischemic stroke has two categories- embolic stroke and thrombotic stroke. An embolic stroke occurs by forming a clot in any part of the body and moves toward the brain and blocks blood flow. A thrombotic stroke caused by a clot that weakens blood flow in an artery. Hemorrhagic stroke is classified into two types- subarachnoid hemorrhage and intracerebral hemorrhage. Transient ischemic attack is also known as ”ministroke”.
PROPOSED SYSTEM :
used Artificial Neural Networks (ANN), Support Vector Machine (SVM), Decision Tree, Logistic Regression, and ensemble methods (Bagging and Boosting) to classify the stroke disease [2]. They have collected the data from Sugam Multispeciality Hospital, India which contains information about 507 stroke patients ranging from 35 to 90 years of age. The novelty of their work is in the data processing phase, where an algorithm called novel stemmer was used to attain the dataset. In their collected dataset, 91.52% of patients were affected by ischemic stroke and only 8.48% of patients were affected by haemorrhagic stroke. Among the mentioned algorithms, Artificial Neural Networks with stochastic gradient descent learning algorithm have the highest accuracy with 95.3% for classifying stroke.
SYSTEM REQUIREMENTS
SOFTWARE REQUIREMENTS:
• Programming Language : Python
• Font End Technologies : TKInter/Web(HTML,CSS,JS)
• IDE : Jupyter/Spyder/VS Code
• Operating System : Windows 08/10
HARDWARE REQUIREMENTS:
Processor : Core I3
RAM Capacity : 2 GB
Hard Disk : 250 GB
Monitor : 15″ Color
Mouse : 2 or 3 Button Mouse
Key Board : Windows 08/10
