Using MATLAB to catch athletes who cheat
The difference between a gold and a silver medal in many Olympic races is only a fraction of a second. This pushes athletes, and the networks that support them, to look for ways to get even the slightest edge, often through unethical approaches.
On May 17th, The Washington Post reported that 31 Olympic athletes recently tested positive in doping reanalysis. The International Olympic Committee specifically retested samples from 454 athletes from the Beijing 2008 Olympics who are considered likely to compete in the upcoming Rio 2016 Olympics.
The 31 athletes who failed the reanalysis hail from 12 different countries and competed in six sports. Last week, 14 athletes from Russia were identified, including ten who had medaled in the Beijing Olympics. These athletes could be stripped of their medals in addition to being banned from future Olympic competitions.
Using MATLAB to catch athletes who cheat
To avoid discovery, athletes have moved away from heavy steroid use. Now, synthetic hormones and erythropoietin (EPO) are more common. EPO allows the blood to have a greater carrying capacity for oxygen. Some athletes take small amounts—known as “microdosing”—to evade detection and still get the benefits.
EPO is nearly impossible to detect in normal blood and urine tests, and synthetic hormones masquerade as regular hormones in athletes. In the case of microdosing, detection of minuscule amounts may not be enough to prove the use of doping in athletes, since “normal” varies on a person-by-person basis.
Researchers from World Anti-Doping Agency (WADA) have turned to Bayesian statistics to find changes that deviate from the “normal” for any given athlete.
“They soon realized that, even if they couldn’t detect the drugs themselves, they could often detect the effects of the drugs—changes in the levels and ratios of young and mature red blood cells, for example,” says Alex Hutchinson for The New Yorker. “Even better, this approach could flag athletes who were turning to so-called autologous blood doping, extracting and later reinjecting their own blood without actually using any illicit substances.”
A Bayesian solution looks at previous results to predict normal results going forward, and can flag those results that fall outside the expected range. In order to do this, testing centers create an athlete biological passport (ABP): a record in which results of doping tests are collected over a period of time and used to calculate thresholds of suspicious results for each individual athlete.
- Researchers from the Swiss Laboratory for Doping Analyses and WADA used a Bayesian approach to highlight the validity of a blood passport for elite athletes:
“In the fight against blood doping, the interpretation of the measured levels of blood markers is based on either population-derived reference ranges or the previous test history of the individual under scrutiny. In this report, we demonstrate how an empirical hierarchical Bayesian model can be used to unify both approaches,” excerpt from the paper A forensic approach to the interpretation of blood doping markers.
The Bayesian network utilized in this research was implemented in MATLAB with Statistics Toolbox. The researchers also created a standalone application with MATLAB Compiler, in order to provide access to the tool to other researchers.
- The Swiss Laboratory for Doping Analyses examined the ABPs of athletes from different countries, a process detailed in the paper Prevalence of Blood Doping in Samples Collected from Elite Track and Field Athletes. The paper outlines how researchers performed rigorous statistical analysis using MATLAB with Statistics Toolbox. In addition to the expected difference between endurance and non-endurance athletes, this study found nationality to be the major factor in the prevalence of expected doping.
What will result from recent testing revelations?
Will individual athletes be banned from the upcoming Olympics or entire teams? Will medals be stripped from athletes? While the outcome of the recent testing revelations is uncertain, we can be certain that as new methods evolve to defeat the testing procedures, new tests will be designed to catch athletes who cheat.
To leave a comment, please click here to sign in to your MathWorks Account or create a new one.