• Version
• Download 48
• File Size 758.18 KB
• File Count 1
• Create Date 18 June 2024
• Last Updated 18 June 2024

Intelligent Stroke Analysis: Utilizing Machine Learning Algorithms for Enhanced Clinical Decision Making

Geeta Bharti , Rajnandani Patil , Vedant Kulkarni , Pratik Kulkarni

Bachelors of Technology, Department of Computer Science and Engineering, SOET

DY PATIL UNIVERSITY, AMBI, PUNE.

e-mail: geetabhrti789@gmail.com, rajnandnipatil2@gmail.com, vedantk0501@gmail.com,

Abstract

Strokes happen when blood flow to the brain is cut off, causing serious damage. Predicting who might have a stroke beforehand is crucial for early treatment and better outcomes. This study explores a new way to predict stroke risk using a powerful type of artificial intelligence (AI) called XGBoost.

Imagine XGBoost as a super-sleuth that analyzes patient information like age, weight, medical history, and blood pressure. But before feeding this data to XGBoost, we clean and organize it for the best results. Unlike some AI, XGBoost is transparent, revealing the key factors that put each patient at higher risk of stroke.

Our research has two main goals:

1. Improved Stroke Prediction: By leveraging XGBoost's strengths, we aim to predict stroke risk even more accurately than previous methods.

2. Understanding Risk Factors: XGBoost's transparency allows us to see why someone might be at higher risk. This knowledge helps doctors create personalized plans to prevent strokes.

This research has the potential to revolutionize stroke prevention. By combining high accuracy with clear explanations, XGBoost can become a valuable tool for early intervention and customized prevention strategies.

Keywords: Stroke, AI, Prediction, Risk factors, Prevention