In the context of clinical activities of the Research, Innovation and Dissemination Center for Neuromathematics (RIDC NeuroMat), the Goalkeeper's Game appears as a potential resource for the diagnosis of Parkinson's Disease. A degenerative disorder of the nervous system, this disease affects 1% of the world population of over 65 years, according to the Association Brazil Parkinson. This work is part of the NeuroMat-led network AMPARO.
Current protocols for Parkinson's Disease tend to require time-extensive application and assessment for diagnosis. Strategies for diagnosing and treating Parkinson's Disease in Brazil are regulated by Clinical Protocols and Therapeutic Guidelines issued by the Ministry of Health. These guidelines stem from broad technical-scientific consensus and are formulated within parameters of quality, precision, indication and posology. NeuroMat's aim is to provide a more efficient resource for diagnosis, relying on the Goalkeeper's Game, in accordance with criteria associated to these protocols and guidelines.
Maria Elisa Pimentel Piemonte, a professor at the Institute of Psychology at the University of São Paulo and an associate investigator at NeuroMat, points out that one of the objectives of the Goalkeeper's Game is to be able to anticipate the identification of Parkinson's Disease. According to Piemonte, some consequences of the disease appear 10 to 20 years before the diagnosis is possible today, and therefore it is unlikely that changes in gait, for instance, are associated to the disease. Furthermore, changes in gait automaticity, related to levels of attention to routine tasks such as walking, also occur years before the diagnosis is possible. It has been hypothesized that changes in automaticity are associated with changes in implicit learning. According to Piemonte, the Goalkeeper's Game has been experimentally used in patients associated to the AMPARO network in order to further investigate this association.
Early diagnosis has been seen as a means of slowing down the evolution of the disease, which can eventually lead to a significant decrease on the costs associated to treating the disease. According to the Brazilian Ministry of Health, Parkinson's Disease is progressive and typically leads to severe disability after 10 to 15 years; social and financial costs associated to advanced steps of the disease are high, particularly for the elderly. It was estimated that the annual worldwide cost of drugs used in the context of the treatment of the disease cost around US$ 11 billion, and that this cost is three to four times higher in the case of patients at advanced stages of the disease.
Professor and researcher Rafael Bassi Stern, from the Department of Statistics of the Federal University of São Carlos (UFSCar), and his team have worked to establish parameters for diagnosis based on the Goalkeeper's Game. The challenge is to retrieve from game strategies information regularities to be able to establish groups or types of players and identify relevant characteristics that might be related to what one would get from a more complicated clinical experiment. According to Stern, in each phase of the game, the automated kicker's behavior is driven by a probabilistic tree and one of the objectives is to know if certain trees are harder to learn than others. In a similar way, it would be important to know what the components in a tree make it more difficult to learn. In the context of the research in conjunction with the AMPARO Network, the goal is to understand, in a general way, how participants strategized and behaved during the game.
Stern and his team have begun to model basic components that influence the likelihood of a player to have a successful hit. Components such as the "learning rate", the probability of increasing success throughout the game, and "learning limit", the maximum likelihood of success that is achievable for a player, were chosen. These parameters are adjusted given each participant and each phase of the goalkeeper's game. This is seen as an initial stage in the development, since as Stern put it: "Once we understand the behavior of these participants and develop mathematical models that are suitable to model theur behavior, we can reach the general clinical objectives of the game."
This piece is part of NeuroMat's Newsletter #44. Read more hereShare on Twitter Share on Facebook
Featuring this week:
Stay informed on our latest news!
|Follow Us on Facebook|