Cassiopée Applied Analytical Systems

Data and their Collection

Recently and thanks to the new digital technologies, training monitoring has become increasingly popular in the world of sports. A lot of parameters are described as useful to track in order to achieve an efficient athlete- and performance-oriented training monitoring. The choice of the adequate parameters depends on the sport-specific context, the goal of the monitoring program as well as on the available means and resources. We analyze both objectibe and subjetive data to acheive these goals.

Artificial Neural Networks Analyses

We implement an innovative Artificial Intelligence-based monitoring system of several training and recovery parameters assessed regularly during certain number of weeks. The result is a complex snapshot of performance, imprevement and injuri avoidence individualzed to each athlete.

Cassiopée Sport Platform

The Geometric Athletic Performnace Index (GAPI). A combination of external (distance) and internal training load is the most appropriate one to monitor training and recovery monitoring among athletes. Moreover, this training monitoring system is able to predict at any week during the whole season if an athlete is located in the positive or in the negative predictive area of performance. Last but not least, GAPI indicates athletes and coaches which parameters can be modified in order to reach the positive predictive performance area.

Coach View
We use an innovative Artificial Intelligence-based monitoring system of several training and recovery parameters assessed regularly during a number of weeks. Athletes provide both objective and subjective in a web platform. We consider different parameters combinations for which we use Artificial Neural Networks geometric optimization to construct separation hyperplanes representing a positive and a negative predictive performance area. To better understand the outcome of the complex graphs obtained, we introduce the Geometric Activity Performance Index (GAPI) defined as the ratio of the positive and the negative predictive performance. We correlate GAPI with objective performance measures to check the relevance of the selected combinations. Separation hyperplanes accurately discriminate improvement versus non-improvement. A combination of external (distance) and internal training load seems to be the most appropriate one to monitor training and recovery monitoring among swimmers. Moreover, this training monitoring system should be able to predict at any week during the whole season if an athlete is located in the positive or in the negative predictive area of performance. Our approach indicates to athletes and coaches which parameters can be modified in order to reach the positive predictive performance area.
support@cassiopee.org | Phone: +41 079 317 7889 | Cassiopee Applied Analytical Systems, Sàrl, Lausanne, Switzerland