Automatic UPDRS Evaluation in the Sit-to-Stand Task of Parkinsonians: Kinematic Analysis and Comparative Outlook on the Leg Agility Task

Abstract

Abstract-In this work, we first characterize the Sit-to-Stand (S2S) task, which contributes to the evaluation of the degree of severity of the Parkinson's Disease (PD), through kinematic features, which are then linked to the Unified Parkinson's Disease Rating Scale (UPDRS) scores. We propose to use a single body-worn wireless inertial node placed on the chest of a patient. The experimental investigation is carried out considering 24 PD patients, comparing the obtained results directly with the kinematic characterization of the Leg Agility (LA) task performed by the same set of patients. We show that (i) the S2S and LA tasks are rather unrelated and (ii) the UPDRS distributions (for both S2S and LA tasks) across the patients have a direct impact on the observed system performance.

On the correlation between UPDRS scoring in the leg agility, sit-to-stand, and gait tasks for parkinsonians

“…The characterization of single UPDRS tasks has been investigated and discussed in our previous works [4], [6], [8], [9]. As mentioned in Section I, across those works we have designed and implemented a unified approach to automatic UPDRS scoring for each of the considered tasks, using the same data processing and classification methods.…”

“…For this reason, the features chosen for its characterization are very intuitive and can be easily computed considering the inclination of the torso measured using the sensor placed on the chest [6]. The typical shape of the signal during the S2S exercise allows to identify the time instants corresponding to: the beginning of the bending forward phase; the transition to the bending backward phase; and the end of the latter.…”

“…In [4], [6], [9], exhaustive analyses have been carried out in order to find the system configurations (in terms of selected features and classification algorithms) which allow to maximize the Area under the Curve (AuC) of the Cumulative Distribution Function (CDF) of e in the LA, S2S, and G tasks. To this end, the three classifiers have been tested considering all the possible combinations of kinematic features in each task-when the kNN algorithm is used, different values of k (from 1 to 10) have been considered.…”

“…The measured variables can be used directly by neurologists to achieve a more objective evaluation of patients' symptoms or to train proper e-health systems, with the aim to automatically assess the clinical trials, trying to match as much as possible the medical evaluation criteria. For PD patients, motor tasks such as Leg Agility (LA) [1], [3], [4], Sit-to-Stand (S2S) [5], [6], Gait (G) [7]- [9], and tremors [10] have been evaluated. In the literature, most of the works in this field focus on specific aspects of single UPDRS tasks and, for the best of our knowledge, limited attention has been devoted to investigate the relationships between different motor tasks quantitatively characterized through the same motion capture technology.…”

“…Recently, a Body Sensor Network (BSN)-based unified approach for the evaluation of different UPDRS tasks (namely, the LA [4], S2S [6], and G tasks [8], [9]), based on a simple BSN formed by three IMUs, has been proposed. Every task is characterized by extracting kinematic features associated with its typical movement patterns, in both time and frequency domains, and an automatic UPDRS scoring system is developed.…”