WADA's anti-doping controls involve the collection of blood and urine samples. This project addresses the critical need for enhancing anti-doping practices in sports through the application of advanced machine learning tools, focusing on the detection of erythropoietin (EPO) abuse. Doping undermines the integrity of competitive sports, and with the increasing complexity of doping techniques, traditional methods struggle to keep pace. The motivation for this project stems from the vast yet underutilized longitudinal data collected from athletes, which holds immense potential for uncovering patterns indicative of doping. By leveraging state-of-the-art algorithms, this initiative aims to analyze hematological profiles over time, minimizing noise and variability while identifying subtle biological signatures that might indicate EPO abuse. The project seeks to ensure fairness and objectivity in anti-doping decision-making by reducing reliance on manual evaluations, which can be prone to error and subjectivity. Building on prior successes in detecting recombinant human EPO using machine learning, this project goes a step further by integrating longitudinal analysis and exploring novel data-driven approaches. The ultimate goal is to develop a robust, transparent, and scientifically sound tool that aligns with WADA’s mission to promote clean sports. This tool will not only improve detection accuracy but also bolster confidence in anti-doping measures among athletes, stakeholders, and the global sporting community, safeguarding the principles of fairness and integrity in sports.
Partners
Deutsches Forschungszentrum für Künstliche Intelligenz (DFKI) GmbH, World Anti-Doping Agency (WADA)