The purpose of this project is to try to predict the occurrence of injuries based on player's in-game statistics.

Predicting NFL injuries and analyzing circumstances in which injuries occur from a play-by-play perspective

impelementation to my IJETI, 2026 paper "Real-Time Video-Based Measurement of Back Angles Using YOLOv8 and Edge Detection for Strength Training"

A model to predict NYC motor vehicle collisions and a heatmap to display where collisions are occuring.

Multi-axis taxonomy of Australian coronial recommendations 1998-2026 — code, codebook, and pre-registered analysis outputs. Pre-registered at OSF DOI 10.17605/OSF.IO/NEX85 under CC-BY 4.0.

Real-time monitoring system using ML to analyze lifting stability, detect anomalies, and provide injury risk feedback.

Comprehensive machine learning analysis on player injury prevention for the San Diego Padres baseball team (simulated dataset) conducted in MIS 401: Business Intelligence and Analytics at SDSU. Built in RapidMiner Studio (v10.2) using logistic regression, deep learning, and decision tree models.

Evidence-based sports injury prevention: load management, neuromuscular programs, ACL/hamstring/running injury protocols, and return-to-sport criteria. Peer-reviewed RCT evidence.

安全跑者的身体底盘知识库：用费曼学习法沉淀跑步安全、力量重建、伤病复盘、ACE-CPT 学习与长期训练系统。

An end-to-peer sports analytics system that utilizes the Acute:Chronic Workload Ratio (ACWR) and Machine Learning (Random Forest vs. Logistic Regression) to predict injury risk and provide clinical decision support for tennis coaches.

A cross-platform application designed for athletes to document and track minor sports injuries through structured observations, including pain levels, mobility, and photo documentation.

Analysis of Injury types within professional basketball players, alongside statistical analysis of type of rehabilitational programs used.

Three-class football player fatigue prediction from PAMAP2 wearable IoT data. Karvonen heart rate labeling, SMOTE balancing, LOSO cross-validation, personalized Random Forest. 97.96% LOSO accuracy + coach substitution-alert dashboard.