Project R-12784

Automated Functional Evaluation During Rehabilitation Exercises Using Wearable Sensors (Research)

Motor dysfunctions caused by neurological conditions represent the third most common cause of the global burden of disease (WHO, 2019), affecting more than half of the population in Europe. Despite great efforts, however, only a limited portion of participants in clinical studies experiences satisfactory improvements, highlighting the need to maximise rehabilitation outcomes: in other words, there is a necessity for personalised medicine and "precision rehabilitation" that would account for the unique characteristics of each individual, to then develop patient-specific interventions; in this sense, it is therefore fundamental to be able to track progress and predict treatment outcomes (Adans-Dester et al., 2020; Hulsen et al., 2019; Niederberger et al., 2019). Yet, one of the main challenges seen in outpatient rehabilitation settings is indeed the lack of information on how the progress observed in clinic translates into functional performance at home: as a matter of fact, nowadays healthcare providers have to rely mainly on patient-reported outcomes about their patients' progress outside the clinic, usually lacking longitudinal analysis on the evolution of treatment outcomes (Jones et al., 2020). In this context, gathering data to augment the information otherwise obtained solely in clinics could improve both the effectiveness and the efficiency of rehabilitation, by monitoring and predicting the trajectory of patient's recovery (Adans-Dester et al., 2020). In order to actualise the collection and analysis of "big data" on patients, though, there is a need for technological solutions that are easy to access and to use for both clinicians and patients, discreet and cost-effective: wearable sensors and smartphone-based health apps could in this sense represent the ideal solution, as the adoption of such technologies is rapidly accelerating and becoming increasingly available to people worldwide (Aitkin et al., 2017; Bonnechère & Sahakian, 2020). As a matter of fact, the use of wearable sensors has increased exponentially in recent years as they allow to monitor patients with a higher sensitivity compared to classical clinical assessment. Currently, wearable sensors are widely accepted for the assessment of motor dysfunctions, with sensor-based rehabilitation for neurological and musculoskeletal conditions showing an improved efficacy of the rehabilitation interventions, both in face-to-face and remote settings (Picerno et al., 2021). Further development of cost-effective technological solutions aiming to monitor and evaluate the quality of rehabilitative interventions and to remotely assess patients in telehealth scenarios, while keeping their motivation and adherence high, would make the future implementation of such technologies easier, allowing their use on a large scale. Such evaluation can also be used to determine if it is possible to predict the long-term prognosis of these patients based on the data obtained through the wearable sensors.

01 January 2022 - 31 December 2023