First place solution for "medicine" topic in AI Challenge 2023
-
Updated
Sep 25, 2024 - Python
First place solution for "medicine" topic in AI Challenge 2023
AxionCura is a machine learning–based early stroke prediction system that analyzes patient health attributes to estimate stroke risk. It includes a trained pipeline model and a simple Python interface for running predictions locally.
Implementation of the study: "The Use of Deep Learning to Predict Stroke Patient Mortality" by Cheon et al.
Production-ready stroke risk assessment platform powered by Dense Stacking Ensemble (DSE) ML models. Achieves 95-97% accuracy with real-time predictions.
🚑 AI-powered Stroke Risk Prediction System built with Ensemble Learning (Random Forest, XGBoost, LightGBM, CatBoost, Gradient Boosting). Designed to assist early detection of stroke-prone individuals using interpretable ML insights — bridging data science and preventive healthcare.
An AI-powered web application that predicts the risk of stroke in patients based on their health and lifestyle data.
End-to-end ML pipeline for stroke risk prediction with XGBoost, FastAPI API, and Streamlit dashboard — designed to handle severe class imbalance and maximize recall.
Neural network to predict strokes.
Stroke analysis, dataset - https://www.kaggle.com/datasets/fedesoriano/stroke-prediction-dataset. For analysis i used: mlp classifier, k-means clustering, k-neighbors classifier. Libraries: tensorflow, scikit-learn.
🚀 Stroke Data Analytics Projects: Python-powered insights into cardiovascular risks using 172K patient records – from procedural querying to ML predictions! 💉📊
Data Visualization on Medical Database
End‑to‑end project predicting stroke risk using Logistic Regression on the Kaggle Stroke Prediction Dataset, with EDA, preprocessing, and a Streamlit web app
Production-ready stroke prediction engine for pre-hospital triage, combining structure-aware clustering, interpretable ML models, and Azure-deployed containerized inference.
Leakage-aware reproducible benchmark for tabular stroke-risk prediction
Machine Learning based Brain Stroke Risk Prediction System using Random Forest with Visualization Dashboard and PDF Report Generation.
Machine Learning-based stroke risk prediction system using Python, Scikit-learn, and XGBoost with 91% accuracy.
Machine Learning-based Stroke Risk Prediction System built with Python, Scikit-learn and Streamlit featuring an interactive healthcare dashboard.
ML pipeline for stroke risk prediction — K-Means clustering, classification, and imbalance handling on patient health data
Streamlit app predicting stroke risk with an SVM whose hyperparameters are tuned by the Firefly Algorithm instead of grid search.
To associate your repository with the stroke-prediction topic, visit your repo's landing page and select "manage topics."