Study design and setting

The present study was a multicenter, observational, cross-sectional pilot study conducted from May to September 2022. The study was coordinated by the Pediatric Physical Medicine and Rehabilitation Service, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Milan, Italy in collaboration with IRCCS E. Medea, Bosisio Parini, Italy. Partners FifthIngenium and the National Research Council of Italy handled product calibration and technically supervised the video game application. The study was conducted in a hospital setting (i.e., the two research centers involved). The study was approved by the Ethics Committee of Milano Area B (No 817; date of approval: 20/07/2021) and by the IRCCS Medea Ethics Committee (Prot. N. 89/21-CE; date of approval: 12/11/2021). Participation in the present study was proposed to parents/legal guardians during routine follow-up visits. Written informed consent was provided by parents/legal guardians of the children enrolled.

Study population and initial evaluations

We enrolled school-aged children with CP, among those regularly followed at the participating centers.

The initial assessment aimed to evaluate the participants’ overall functioning and their ability to engage in play activities. Trained examiners from the two centers conducted the evaluation using the International Classification of Functioning, Disability, and Health (ICF) framework. This comprehensive assessment considered not only body structures and functions, but also levels of activity and participation, focusing on Chaps. 1 (mental and cognitive functions) and 7 (musculoskeletal and motor functions) of the Body Functions domain of the ICF.

Regarding the motor (Chap. 7 of the ICF) and cognitive (Chap. 1 of the ICF) functioning, children classified as 0 (no impairment), 1 (mild impairment), 2 (moderate impairment) and 3 (severe impairment) were included, while those classified as 4 (complete impairment) were excluded due to severe motor, gestural and cognitive impairment hindering task execution and understanding.

The initial evaluation also included:

Gross Motor Function Classification System (GMFCS) [17]: a widely employed framework to assess the gross motor function of children and youth with CP. It comprises five levels (I to V), each denoting a distinct level of motor impairment, progressing in severity.

Manual Ability Classification System (MACS) [18]: a system designed to classify the manual abilities of children with CP. It consists of five levels, ranging from Level I (limited limitations) to Level V (no effective hand use).

Finally, parents or legal guardians were asked to answer the Assistance to Participate Scale (APS) [19] questionnaire to assess the child’s autonomy and participation levels in play and leisure activities both at home and in social contexts. Questions were scored using a five-point Likert scale (1 = Unable to participate; 2 = Participates with my assistance at all stages of the activity; 3 = Participates after I have set him/her up and help at times during the activity; 4 = Participates with my supervision only; 5 = Participates independently). The APS total score, calculated according to Bourke-Taylor et al. [19], ranges from 0 to 100, where higher scores indicate less assistance required for the child to participate in the activity.

Instrument

GiocAbile is a sensory-motor PC game specifically tailored for children with neuromotor disabilities like CP. The name “GiocAbile” is a blend of the Italian words “gioco” (play) and “abile” (able). The combination of these terms conveys the idea of a video game that is both enjoyable and accessible, particularly for individuals with CP. Indeed, GiocAbile addresses the need for both rehabilitation and integration of children with CP, offering an enabling and inclusive gaming experience. GiocAbile can be played while sitting on a chair, in a wheelchair or standing, depending on preference and ability. The GiocAbile system comprises both software (the video game) and hardware components (PhiCube and PlayCuff).

Video game

The software features a collection of game activities stimulating various motor and cognitive abilities, with the possibility of a multiplayer mode focused on cooperative, non-competitive, gameplay. The video game features an island structure (Fig. 1A) consisting of a series of platform games where the player must navigate through different levels of increasing difficulty, each set in unique environments, such as the Jungle or Lands of ice (Fig. 1B and C). These platform games are interspersed with minigames grouped according to specific body functions, which vary in difficulty levels, ranging from simple segmental movements to more complex articulated motions (Fig. 2). The final challenge is the assembly of a crystal, where the participant must piece together crystal fragments like a puzzle (Fig. 2G). The main character of the video game is Mako the monkey (Figs. 1 and 2).

Fig. 1

The video game: (A) island structure of the video game (B) Jungle platform-game environment and (C) Lands of ice platform-game environment

Fig. 2

Minigames: (A) Canoeing; (B) Drill; (C) Fishing; (D) Boat; (E) Flight; (F) Climbing; (G) Crystal assembly

PhiCube

PhiCube (Fig. 3) is a modular and portable robotic bilateral haptic interface, characterized by high flexibility of use [20]. It allows children to perform different bilateral motor gestures required for different game scenarios.

PhiCube consists of a motorized unit, a control unit and a set of manipulanda (i.e., steering wheel, levers, cranks) (Supplementary Figs. 1 and 2). The core of PhiCube is a robotized central unit equipped with two motorized rotary axes. The axes are equipped with appropriate mechanical interfaces, placed on the two opposite sides of the central body, and with a quick connection for different manipulanda. The Control Unit allows to select the manipulanda and the corresponding control strategies. Manipulanda are made of plastic by additive manufacturing and have handle grips certified for use on bicycle handlebars. With PhiCube, the user can interact with the game in a bilateral way and, thanks to the chosen set of manipulanda, can execute various motor gestures, involving shoulder, elbow and wrist joints in different ways.

Fig. 3

PhiCube with the manipulanda: levers (left), steering wheel (center), cranks (right)

PlayCuff

PlayCuff is a wireless sensorized orthosis that can be worn like a fingerless glove (Suppl. Figure 3). It consists of a double layer of soft neoprene fabric and a separate hand piece. It can be worn by both right-handed and left-handed people and is available in three different sizes to fit children of different ages. PlayCuff implements an orthotic function thanks to two NiTi pseudoelastic alloy elements sewn in between the fabric layers to control wrist posture. The characteristics of this material allow proprioceptive feedback and dynamic joint extension, thereby improving motor control, especially in children with dyskinetic traits [21, 22]. PlayCuff is also equipped with two battery-powered inertial sensors (IMU) and a microcontroller that classifies forearm and wrist gestures in real-time, achieving a frequency of 20 Hz, with very high accuracy (94% and 99.5%, respectively). The recognizable movements include waving the limb up, down, right and left, pronating and supinating the forearm, extending and flexing the wrist or deviating it laterally. Additionally, static positions like pointing upwards, horizontally or downwards can be classified. Based on such gesture recognition capacity, the device can be used to play video games (Fig. 4), controlling the avatar’s actions and making choices within the game menus.

Fig. 4

Children playing GiocAbile using PlayCuff

Testing session and outcome measures

During the testing phase, participants had the opportunity to experience the video game in single-player mode for a total test duration of up to 1 h, as suggested by the American Academy of Pediatrics [23]. Short breaks (one minute) were scheduled every 20 min as advised by the American Optometric Association [24] and the American Academy of Ophthalmology [25].

Children engaged with GiocAbile for approximately 30 min using PhiCube and another 30 min using PlayCuff. Following their gameplay session, children’s gaming experience was assessed through the System Usability Scale (SUS) [26], the Technology Acceptance Model 3 (TAM3) [27] and an ad hoc questionnaire developed by the authors. This phase lasted from 30 to 60 min.

The SUS, a five-point Likert scale questionnaire, evaluates system usability, with a total score ranging from 0 to 100.

The TAM3 evaluates the acceptability of the system according to 14 constructs. Considering the type of device and the assessors (i.e., children) we reduced the assessed dimensions to 8 (i.e., perceived usefulness, perceived ease of use, self-efficacy, perception of external control, playfulness, anxiety, enjoyment and behavioral intention), each rated on a scale of 1 (totally disagree) to 7 (totally agree) [27].

The ad hoc questionnaire (Suppl. Table 1), specifically designed to assess GiocAbile’s software and hardware features, used a five-point Likert scale to evaluate children’s appreciation of the story, character, settings, and sounds proposed.

Language revisions were applied to all questionnaires to enhance understanding for children. Additionally, graphical representations with emoticons were included alongside the Likert scales for improved clarity.

Statistical analysis

Median and interquartile values (i.e., the difference between the 75th percentile and the 25th percentile) were calculated for GMFCS, MACS, ICF, APS and questionnaire scores. SUS total score was computed in Matlab 2022 according to Brooke et al. [26]. TAM3 mean score and constructs were computed in Matlab 2022 according to Venkatesh et al. [27]. For the ad hoc questionnaire, answers referring to the evaluation of PhiCube manipulanda (i.e., steering wheel, single and double crank, single and double lever) were aggregated. The questions regarding PlayCuff and PhiCube were divided in 5 constructs as follows: I managed to use the controllers (Usability); I managed to get the controllers to do what I wanted (Control); Using the controllers was fun (Enjoyment); Using the controllers was disappointing (Disappointment); Using the controllers was complex (Complexity).

A non-parametric correlation analysis was conducted to examine the relationships between GMFCS, MACS, ICF scores, and questionnaire results, aiming to identify influences of children’s motor and cognitive abilities on the declared usability and acceptability of GiocAbile.

Statistical analysis was performed using SPSS (Statistical Package for the Social Sciences, IBM SPSS Statistics for Windows, Version 21.0. Armonk, NY: IBM Corp.).

For all tests, a p value < 0.05 was considered statistically significant.