*4.3. Data Acquisition Procedure*

During the experimental campaign, the HC and PD subjects were equipped with the set of reflective markers shown in Figure 4, and were instructed to perform the UPDRS tasks. Their performances were supervised and assessed by two neurologists and simultaneously acquired by the two systems. The neurologist' scores, the kinematic parameters and the videos of each task performance were recorded for the subsequent analysis.

The PD and HC subjects performed all the tests facing our system, that is, with the depth axis of the Kinect device perpendicular to the subject frontal plane. The proper position of the subject was verified both by the neurologists and by the system software, which checks the availability of the whole Kinect skeleton and its correct positioning. The device was placed 1.2 m height and at a distance of about 2 m from the subjects. No other object apart the chair, and limited to the AC and LA tasks, was allowed in the working volume. At the beginning of each test, the subject elevated the right arm three times, this to allow the synchronization between the two systems. The synchronization was performed by time-shifting the signals of the arm elevation angles measured by the systems such that their cross-correlation was maximized. The PD and HC subjects were told to perform the LA, AC and Po tasks as indicated by the UPDRS guidelines. Furthermore, the PD subjects performed also the Gait and the PSretrop tasks, in order to evaluate the PIGD sub-score. For the LA task both legs were assessed independently. Two acquisition sessions were planned separated by an interval of thirty minutes to allow subjects to rest.

In the first session a total of five tasks were assessed by the neurologists for the fourteen PD subjects, and a total of three tasks were supervised for the twelve HC subjects. In the second session the same tasks of the first one were repeated in random order. The body sway of the CoM movements was measured during the Po task: in the first phase (indicated as Phase1) each subject was told to stand up straight for ten seconds. Then, during the second phase (indicated as Phase2) each subject was told to try to improve and maintain a more straight posture for other ten seconds: this can be considered a sort of secondary motor task, that potentially can highlight differences between PD and HC subjects [38].
