• 

  distribution analysis

Sports Injury Risk Analytics Platform

This project is a sports injury risk analytics platform built for the AQX Sports Analytics Hackathon. It analyzes two complementary data sources:

• Biomechanical Movement: Joint angles, ground reaction force, fatigue, and postural instability
• Physiological Sensors: Heart rate, muscle activation, hydration, and sensor fatigue

Together, these datasets help identify patterns and signals associated with sports injury risk.

Key Components

• Full exploratory data analysis (EDA)
• Grouped histograms
• Separate ML models trained on each dataset using 5-fold cross-validation
• Model comparison across:
  • Random Forest
  • Gradient Boosting
  • XGBoost
  • CatBoost
• Interactive Streamlit dashboard presenting:
  • Key findings
  • Risk-range tables
  • Data visualizations
  • Model performance

The dashboard enables coaches and analysts to explore injury-related signals and model insights without having to work directly with the raw datasets.

GitHub Repository

Built With

• matplotlib
• pandas
• plotly
• python
• scikit-learn