Affiliations: [a] Graduate Program in Health Sciences, UNIMONTES, Av. Rui Braga, sn, Vila Mauricéia, Montes Claros, Brazil
| [b] Federal Institute Northern Minas Gerais, Campus Montes Claros, Montes Claros, Brazil
| [c] Department of Computer Science, UNIMONTES, Av. Rui Braga, sn, Vila Mauricéia, Montes Claros, Brazil
| [d]
Pontifical Catholic University of Minas Gerais, Graduate Program in Electrical Engineering, Belo Horizonte, Brazil
Correspondence:
[*]
Corresponding author. Marcos Flávio Silveira Vasconcelos D’Angelo, Department of Computer Science, UNIMONTES, Av. Rui Braga, sn, Vila Maurict’eia, Montes Claros, Brazil. marcos.dangelo@unimontes.br
Abstract: Falls in the elderly are a public health problem because this population tends to have a longer recovery time and consequently longer hospital beds. Studies show that 84% of falls in hospital rooms occur near the bed, that led to strategies to prevent falls in the elderly population have been studied. In this context, this paper presents a schema for the detection and emission of bed exit alerts in the elderly. This schema uses signals derived from RFID sensors processed by a model based on Intelligent Swarm and Fuzzy Sets. The main contribution of this study is the use of a Membership Windows that reduces the effects of missclassification of other strategies. The proposed work evaluated a data set containing 14 elderly aged between 66 and 86 years divided into two rooms. The results show that the presented approach improves the precision and recall in environments with greater uncertainty of classification.
Keywords: Bed exit alarms, elderly care, intelligent swarm, fuzzy sets
