Automated Evaluation of Frailty Index in Elderly
According to the World Bank, Italy has the second-highest share of population aged over 65 worldwide, i.e., 22% in 2014, and statistics related to G20 countries are becoming increasingly similar. Our region Liguria is among the highest in this ranking worldwide. Aging brings, in general, the reduction of the individual’s potential, bringing him into a state of vulnerability and instability of the clinical condition. Over the last decade, in the medical literature, it was introduced the definition of frail elderly to highlight this condition, associated with an elevated risk of complications that may result in loss of functional autonomy or death. The accurate estimation of the frailty of an elderly is therefore an important objective to assess the overall well-being and to predict the risk of mortality. Moreover, in an elderly person the hospital stay, especially if prolonged, can lead to develop new disability and dramatically worsen the risk of mortality. Therefore, it is of particular clinical interest, the ability to quantitatively and automatically estimate the patient frailty during hospitalization. Current strategies are based on MPI (Multidimensional Prognostic Index), based on evaluation of five functionality domains (or dimensions): clinical, cognitive, functional, nutritional, and social, defined starting from the International Classification of Functioning, Disability, and Health (ICF). The assessment is mostly done through questionnaires and the estimate of scores manually by the medical staff, to establish the level of the cumulative deficit. Therefore, the index is computed episodically and not continuously for an extended time interval.
We aim to perform a continuous comprehensive geriatric assessment of the assisted persons frailty index, and coach them in keeping an active and healthy lifestyle, informing medical staff in case of medical assistance is required. Our study it taking place in a proof-of-concept model of protected discharge, in which the patient is hosted for few days (one week), after being discharged from the hospital. The facility is located within Ospedale Galliera (Genova) and equipped as a comfortable apartment, where the patients are monitored by a system of sensors (video-cameras, RGB-D, localization sensors, presence infra-red sensors) while physicians and nurses have the opportunity of monitoring the patient remotely.
We are currently developing algorithms for the automatic estimation of motility indices, analysing patients activities throughout the day.
- L. P. Fried et al “Untangling the concepts of disability, frailty, and comorbidity: implications for improved targeting and care,” The Journals of Gerontology Series A: Biological Sciences and Medical Sciences, vol. 59, no. 3, pp. M255–M263, 2004.
- A. Pilotto et al. “Development and validation of a multidimensional prognostic index for one-year mortality from comprehensive geriatric assessment in hospitalized older patients,” Rejuvenation research, vol. 11, no. 1, pp. 151–161, 2008.