Gallicchio C., Micheli A., Fortunati L., Vozzi F., Parodi O.
Reservoir computing Echo state network Learning with temporal data
A relevant aspect in the field of health monitoring is repre- sented by the evaluation of balance stability in the elderly. The Berg Bal- ance Scale (BBS) represents a golden standard test for clinical assessment of balance stability. Recently, the Wii Balance Board has been success- fully validated as an effective tool for the analysis of static balance-related features such as the duration or the speed of assessment of patient's center of pressure. In this paper we propose an innovative unobtrusive approach for automatic evaluation of balance assessment, by analyzing the whole temporal information generated by the balance board. In par- ticular, using Recurrent Neural Networks implemented according to the Reservoir Computing paradigm, we propose to estimate the BBS score of a patient from the temporal data gathered during the execution on the balance board of one simple BBS exercise. The experimental assessment of the proposed approach on real-world data shows promising results.
Source: ISTI Technical reports, 2015
@techreport{oai:it.cnr:prodotti:345042, title = {A reservoir computing approach for balance assessment}, author = {Gallicchio C. and Micheli A. and Fortunati L. and Vozzi F. and Parodi O.}, institution = {ISTI Technical reports, 2015}, year = {2015} }
DOREMI
Decrease of cOgnitive decline, malnutRition and sedEntariness by elderly empowerment in lifestyle Management and social Inclusion