Subscribe to Our Newsletter and Save 10 %
Knowledge Base > Features and Functions



The AI DIAGNOSTICS method combines knowledge of mathematical relationships from biochemistry with methods from data science and machine learning. We embed findings from sports science studies and biochemical research into data-driven methods for identifying non-linear systems and use them in the development of artificial neural networks. Together, these form the key building blocks of our methodology.

Physiological laboratory parameters and performance data from completed field tests can be viewed as a system in which there are (non-linear) relationships between several key figures. Some of these relationships can be explained by sports science theories, others seem to be unrelated at first glance.

With a broad data base from laboratory measurements and field tests, we can thus confirm established theories and optimise used parameters in a data-driven way. In the case of more complex and in part still unexplored relationships, we use this data to identify the mathematical description of the underlying systems or to train artificial neural networks.

Taken together, this hybrid approach allows for field test-based performance diagnostics with accuracies that are within the spectrum of laboratory tests.