Human movement modeling and understanding: biomedical, biometrical and neurocognitive applications.

点击翻译按钮获取中文摘要

1-Context : Human movement modeling is of great interest for the design of pattern recognition systems relying on the understanding of the fine motor control as well as for the development of intelligent systems involving in a way or another the processing of human movements. Among the models providing analytical representations of the trajectories, the Kinematic Theory of rapid human movements and its lognormal families of models have been used to explain most of the basic phenomena reported in classical studies on human motor control and to study basic factors involved in the fine motricity. Several software packages have been developed by our team over the years to extract the lognormal parameters from various pen tip velocity curves under different experimental data acquisition conditions and set-up. These different tools allow a researcher to study handwriting through the parametric lognormal description of the neuromuscular networks involved in a given task. 2-Goal: The present proposal aims at elaborating a fundamental and theoretical background for any handwriting processing applications as well as providing some basic knowledge that can be integrated in the development of many automatic systems in three fields of applications: Biomedical: We will focus on three main diseases that affect neuromotor control: Cerebrovascular accidents, Parkinson disease and Alzheimer disease. We will also study how children learning handwriting slowly move and master their motor control behavior. Biometrical: We will focus on signature verification and writer identification to point out which combinations of neuromuscular features are the most representative of a person writing and signing habits. Neurocognitive: We will focus on extending the theory to eye movements, incorporating visual feedback to the models, generalizing the theory to the modeling of 3D movements and putting the whole emergent approach in the context of a universal and global physical model. 3-Methodology and strategy. Our methodology is based on the exploitation of our parameter extraction algorithms to study the behavior of the neuromuscular system under various experimental conditions for different practical applications. Since there are numerous potential pathways that can be explored (which cannot be done by a single isolated team), we have signed recently several research cooperation agreements with various national and international partners. These agreements allows our collaborators to use our software in the context of non-profit research projects in which our team plays the key role of an active associate providing basic knowledge to successfully use our technology and direction in research to many students, resulting in the formation of highly qualified personnel. 4-Impact. A part from training numerous students and researchers, the whole program will lead to important theoretical breakthroughs in neuromotor modeling and set the knowledge background for the design of numerous automatic or interactive devices in three lucrative fields of pattern recognition. This can be illustrated by a typical actualization of the vision that supports our research. In years to come, the gesture and the handwriting habits of people using intelligent handheld devices will be monitored and analyzed by their own apparatuses, in the context of the our neuromuscular models. These systems will be able to identify their users, protect their equipment and sensitive data, to follow the evolution of their fine motor control to check for any departure form lognormality that might necessitate medical intervention. These new tools will facilitate man machine interactions in numerous applications where handwriting and drawing remains the most natural and ergonomic way to communicate.