Publication
Generative AI in Anti-Doping Analysis in Sports
Maxx Richard Rahman; Wolfgang Maaß
In: Carlo Dindorf; Eva Bartaguiz; Freya Gassmann; Michael Fröhlich. Artificial Intelligence in Sports, Movement, and Health. Chapter 6, Pages 81-95, Springer, 2024.
Abstract
Doping in sports involves the abuse of prohibited substances to enhance performance in the sporting event. Blood doping, a prevalent method, allows the increase in red blood cell count to improve aerobic capacity, often through blood transfusions or synthetic Erythropoietin (rhEPO). Current indirect detection methods require a large amount of data for performing analysis. In this paper, we study the use of generative modelling for generating synthetic blood sample data to improve anti-doping analysis in sports. We performed experiment on the blood samples collected during the clinical trial. The dataset comprised haematological parameters from real blood samples, which were analyzed to understand the baseline characteristics. The Generative Adversarial Network (GAN) is used to understand the complexity and variability of real blood sample data. Results demonstrated that the model could successfully generate synthetic samples that closely resembled real samples, indi-cating its potential for augmenting datasets used in doping detection. This approach not only enhances the robustness of indirect methods of doping detection by providing a larger dataset for analysis but also addresses ethical concerns related to privacy and consent in using athletes’ biological data.
Projects
- 3S Project - Steroid Sample Swapping
- EPOPredictII - Erythropoietin Prediction II
- MARVIN - Development of AI-based Screening Tool to detect Identical Urine Samples within the Athlete Biological Passport
- 3S-II - AI-based detection of steroid sample swapping in the anti-doping area and analysis of the associated decision-making processes