Falls in the elderly are a known problem, leading to hospitalization, impaired life quality, and social costs. Falls are associated to multiple risk factors, related to the subject’s health, lifestyle, and living environment. Living alone makes it difficult to detect a patient’s decline, which increases the fall risk. In this paper, we present NONCADO, a project funded by the Lombardy Region (Italy), aimed at developing a system for preventing falls in the elderly living alone, by integrating data from a network of sensors (both wearable and environmental). The collected data are analyzed by a decision support system (DSS) that exploits advanced temporal data analysis techniques to detect behaviors known to increase the individual risk (e.g. moving within the house with inadequate lighting, or performing not enough physical activity). A daily report listing the detected risky behaviors is produced and delivered through a mobile app. Since we address long-term monitoring, it’s important to detect as well the changes in a subject’s habits that may increase fall risk. Such changes are summarized in a weekly report. A preliminary feasibility evaluation of the system was performed during a 2-weeks pilot study involving 16 patients treated at the Casa di Cura Privata del Policlinico hospital, in Milan, Italy. Patients were asked to perform 5 activities, and the system’s ability to correctly detect them was assessed. The study results were encouraging, as the system reached an overall accuracy of 90%.

Salvi, E., Panzarasa, S., Bagarotti, R., Picardi, M., Boninsegna, R., Sterpi, I., et al. (2019). NONCADO: A system to prevent falls by encouraging healthy habits in elderly people. In Artificial Intelligence in Medicine 17th Conference on Artificial Intelligence in Medicine, AIME 2019, Poznan, Poland, June 26–29, 2019, Proceedings (pp.227-232). Springer Verlag [10.1007/978-3-030-21642-9_28].

NONCADO: A system to prevent falls by encouraging healthy habits in elderly people

Picardi M.;
2019

Abstract

Falls in the elderly are a known problem, leading to hospitalization, impaired life quality, and social costs. Falls are associated to multiple risk factors, related to the subject’s health, lifestyle, and living environment. Living alone makes it difficult to detect a patient’s decline, which increases the fall risk. In this paper, we present NONCADO, a project funded by the Lombardy Region (Italy), aimed at developing a system for preventing falls in the elderly living alone, by integrating data from a network of sensors (both wearable and environmental). The collected data are analyzed by a decision support system (DSS) that exploits advanced temporal data analysis techniques to detect behaviors known to increase the individual risk (e.g. moving within the house with inadequate lighting, or performing not enough physical activity). A daily report listing the detected risky behaviors is produced and delivered through a mobile app. Since we address long-term monitoring, it’s important to detect as well the changes in a subject’s habits that may increase fall risk. Such changes are summarized in a weekly report. A preliminary feasibility evaluation of the system was performed during a 2-weeks pilot study involving 16 patients treated at the Casa di Cura Privata del Policlinico hospital, in Milan, Italy. Patients were asked to perform 5 activities, and the system’s ability to correctly detect them was assessed. The study results were encouraging, as the system reached an overall accuracy of 90%.
abstract + poster
Fall risk; Home monitoring; Temporal data analysis;
English
17th Conference on Artificial Intelligence in Medicine, AIME 2019 - 26 June 2019 through 29 June 2019
2019
Artificial Intelligence in Medicine 17th Conference on Artificial Intelligence in Medicine, AIME 2019, Poznan, Poland, June 26–29, 2019, Proceedings
9783030216412
2019
11526 LNCS
227
232
none
Salvi, E., Panzarasa, S., Bagarotti, R., Picardi, M., Boninsegna, R., Sterpi, I., et al. (2019). NONCADO: A system to prevent falls by encouraging healthy habits in elderly people. In Artificial Intelligence in Medicine 17th Conference on Artificial Intelligence in Medicine, AIME 2019, Poznan, Poland, June 26–29, 2019, Proceedings (pp.227-232). Springer Verlag [10.1007/978-3-030-21642-9_28].
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10281/468568
Citazioni
  • Scopus 0
  • ???jsp.display-item.citation.isi??? 0
Social impact