Virtual Reality-Based Adapted Handball Serious Game for Upper Limb

Rehabilitation in Spinal Cord Injured Patients

J. Albusac

1 a

, V. Herrera

1 b

, Oscar Dominguez-Oca

, E. Angulo

1 c

A. de los Reyes-Guzm

2 d

and D. Vallejo

1 e

Department of Technologies and Information Systems, University of Castilla-La Mancha, Spain

Biomechanics and Technical Aids Unit. Hospital Nacional de Parapl

ejicos de Toledo, Spain

Keywords:

Upper Limb, Rehabilitation, Virtual Reality, Serious Games, Handball.

Abstract:

In this paper, we present a virtual reality-based serious game that simulates the training of a wheelchair hand-

ball goalkeeper. It is designed to complement traditional therapy for upper limb rehabilitation and trunk mo-

bility improvement in spinal cord injury patients. The proposal is underpinned by a multi-layered architecture

that provides a therapeutic environment that enhances patient motivation and satisfaction through gamiﬁcation

techniques. The architecture also provides precise kinematic recording during exercise performance, since the

recorded data is essential for therapists to objectively assess each patient’s progress. Particularly, the recorded

can be used to assess the extent of movement, how fast and smooth it is, the number of repetitions and their

consistency, as well as the accuracy and precision of movements, balance, and posture control. The seri-

ous game was tested in the Hospital Nacional de Parapl

ejicos de Toledo, involving patients and healthcare

professionals. The collected data are publicly available. This preliminary evaluation has been focused on

assessing its functionality and safety. Following the exercise sessions, all participants were asked to complete

a short questionnaire to measure their motivation, sense of achievement, satisfaction and overall comfort and

well-being in the virtual environment. Future plans include expanding the patient sample and monitoring the

long-term progress and impact of VR therapy on the recovery of mobility in the affected limbs.

1 INTRODUCTION

Individuals with spinal cord injury (SCI) confront

substantial physical challenges daily. Annually, be-

tween 250,000 and 500,000 people worldwide sus-

tain a spinal cord injury, predominantly due to pre-

ventable circumstances such as road trafﬁc accidents,

falls, or violence (WHO, 2021). A signiﬁcant num-

ber of individuals with SCI often rely on wheelchairs

for mobility, depending on the severity of their injury

and the resulting impairment in leg function, and they

may often experience limited mobility in their trunk,

arms, and hands. These limitations further complicate

daily activities, posing challenges for independent liv-

ing and overall quality of life.

In this sense, rehabilitation plays a crucial role in

enhancing the quality of life for individuals with SCI.

Through consistent and targeted rehabilitative exer-

https://orcid.org/0000-0003-1889-3065

https://orcid.org/0000-0002-6187-4794

https://orcid.org/0000-0002-2659-3129

https://orcid.org/0000-0003-2905-2405

https://orcid.org/0000-0002-6001-7192

cises, it is possible to regain some degree of mobility,

improve motor functions, and promote overall physi-

cal well-being (Spooren et al., 2011). Persistence in

rehabilitation is vital; a committed approach can facil-

itate more effective recovery, helping individuals nav-

igate their daily lives with increased conﬁdence and

capability. Traditional therapeutic approaches, while

fundamental, often involve repetitive and routine ex-

ercises, which can, over time, diminish a patient’s mo-

tivation and engagement to rehabilitation programs.

This drawback can adversely impact the consistency

of practice, which is essential for recovery and im-

provement in individuals with SCI (Shahmoradi et al.,

2021).

Recognizing this challenge, there has been a pro-

gressive introduction of technology in the ﬁeld of re-

habilitation in recent years. The integration of inno-

vative therapeutic strategies, such as adaptive sports

and technology-augmented exercises, can further en-

rich the rehabilitation process, making it more en-

gaging and sustainable for SCI patients (Herne et al.,

2022). Technological integration aims to mitigate

the monotony of traditional therapy, incorporating en-

gaging, varied, and personalized exercises to main-

Albusac, J., Herrera, V., Dominguez-Ocaña, O., Angulo, E., Reyes-Guzmán, A. and Vallejo, D.

Virtual Reality-Based Adapted Handball Serious Game for Upper Limb Rehabilitation in Spinal Cord Injured Patients.

DOI: 10.5220/0012554600003690

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 26th International Conference on Enterpr ise Information Systems (ICEIS 2024) - Volume 2, pages 337-347

ISBN: 978-989-758-692-7; ISSN: 2184-4992

337

tain patients’ interest and commitment. By enhanc-

ing the appeal and interactivity of rehabilitation ac-

tivities, technology facilitates a more patient-centered

approach, promoting improved consistency and out-

comes in the recovery process (Doumas et al., 2021).

Building on the technological advancements in

rehabilitation, serious games and exergames have

emerged as one of these innovative therapeutic strate-

gies (Proenc¸a et al., 2018). These applications use

gamiﬁcation principles, incorporating elements of

play into the rehabilitation process to create an im-

mersive and motivating environment for individuals

with SCI. They enrich the therapy experience, trans-

forming conventional exercises into interactive and

purposeful activities, and promote a sense of achieve-

ment and progress, encouraging patients to actively

participate in their recovery process (de Los Reyes-

Guzm

an et al., 2021). By transforming therapeutic

tasks into game-like experiences, these approaches fa-

cilitate higher levels of patient engagement and sat-

isfaction, which are essential for maintaining consis-

tency and enhancing the effectiveness of rehabilita-

tion programs.

The success of rehabilitation relies not only on

consistent practice, but also on performing exercises

accurately, which requires the therapist’s guidance.

In addition to enhancing motivation, incorporating

advanced technologies offers beneﬁts like precise

recording of the patient’s tracking. Hence, the ob-

tained data is highly valuable for the patient to per-

form the exercises accurately, and for the therapist to

objectively analyse the rehabilitation sessions and the

patient’s progress (Herrera et al., 2023a).

Within the diverse spectrum of technologies used

for serious games and exergames, Virtual Reality

(VR) holds a distinctive place, offering immersive

environments that can be tailored to meet the indi-

vidual needs and rehabilitation goals of persons with

SCI (Herrera et al., 2023b). VR creates dynamic,

customizable spaces, allowing therapists to adjust the

complexity and type of exercise, as well as accurate

recording of kinematics during execution. In partic-

ular, sport-based VR rehabilitation is a transforma-

tive approach to the rehabilitation of individuals with

SCI (de Los Reyes-Guzm

an et al., 2021). This ap-

proach combines the advantages of VR with the bene-

ﬁts of physical activity to create an engaging and per-

sonalized therapeutic environment. In recent years,

sport-based VR rehabilitation has incorporated vari-

ous sports, however, there has yet to be a development

in adapted handball exercises within this innovative

form of therapy. Adapted handball, particularly ori-

ented towards the training of goalkeepers, inherently

exercises the trunk, arms, and hands. Its unique na-

ture encourages enhanced trunk stability (especially

in patients with injuries to the C5, C6, and C7 cervi-

cal levels) and improved mobility in arms and hands,

making it an ideal component for upper limb rehabil-

itation.

Our proposal aims to ﬁll this gap by designing

and developing a VR-based serious game speciﬁcally

for the training of goalkeepers in adapted handball,

which also serves as an adjunct to traditional reha-

bilitation methods. The system allows the patient to

perform exercises without additional components at-

tached to hands and arms, and to interact with virtual

elements through their hands. The proposal encom-

passes accurate kinematic recordings, allowing ther-

apists to objectively analyze patient progress using

supportive data. This development has been incorpo-

rated into the Rehab-Immersive platform used at the

Hospital Nacional de Parapl

ejicos de Toledo (HNPT).

The platform hosts a collection of immersive serious

games that support the upper limb rehabilitation pro-

cess (Herrera et al., 2023a) (Herrera et al., 2023b).

Experiments conducted in a real-world setting,

such as the HNPT, focused on evaluating function-

ality with patients and therapists. The results show

the system’s ability to record movements made dur-

ing therapeutic exercises, offering the capability to in-

dividually analyze the mobility of the left and right

hand, arms, and head, thereby inferring trunk mobil-

ity. Additionally, a progressive improvement in per-

formance is observed among individuals as they be-

come more accustomed to the virtual environment.

The rest of the paper is structured as follow: Sec-

tion 2 reviews some of the relevant previous works

related to the main topic addressed in this paper. Sec-

tion 3 introduces the adapted handball solution based

on VR for upper limb rehabilitation. Section 4 de-

scribes the experimentation and results obtained. The

paper concludes with conclusions and future work in

Section 5.

2 PREVIOUS WORK

Recent advancements in game-based virtual reality

(VR) have signiﬁcantly impacted upper-limb rehabil-

itation post-stroke, combining gamiﬁcation elements

with therapeutic exercises. A systematic review and

meta-analysis of 20 clinical trials highlighted the ef-

fectiveness of game-based VR in improving motor

function and quality of life in stroke survivors, indi-

cating a promising direction for rehabilitation prac-

tices (Dom

ınguez-T

ellez P, 2020). Complementing

this, a study involving the Microsoft Xbox 360 Kinect

system integrated with conventional therapy showed

ICEIS 2024 - 26th International Conference on Enterprise Information Systems

338

notable improvements in upper limb motor functions

in subacute stroke patients, suggesting an added value

of interactive gaming in rehabilitation (Ikbali Afsar

et al., 2018).

Pereira et al.’s exploration into VR for hand reha-

bilitation physiotherapy further underlines the diver-

sity of approaches being explored in this ﬁeld, empha-

sizing on the adaptability of these systems to different

patient needs, based on a study with able-bodied par-

ticipants testing the hardware and software’s usabil-

ity (Pereira et al., 2020).

Meanwhile, the efﬁcacy of VR in sports training,

particularly in handball, has been explored by Vogt

et al. in the context of balance training and muscu-

loskeletal injury rehabilitation. Although VR showed

positive effects in healthy adults, it did not signiﬁ-

cantly outperform traditional balance training in reha-

bilitation settings, pointing to a nuanced understand-

ing of VR’s role in sports-focused rehabilitation (Vogt

et al., 2019).

The broader applicability of VR and gaming in-

terventions in enhancing upper extremity function

post-stroke is evident in studies that have reported a

28.5% average improvement in motor functions, with

a 10.8% greater beneﬁt observed when gaming ele-

ments were incorporated compared to visual feedback

alone (Karamians et al., 2020). This is supported by

another study involving VR training with upper ex-

tremity tasks using the HTC Vive HMD, where pa-

tients demonstrated signiﬁcant functional improve-

ments and high satisfaction, afﬁrming the feasibility

and effectiveness of fully immersive VR rehabilita-

tion (Lee et al., 2020).

Feedback from physiotherapists and game design-

ers has been instrumental in shaping VR game con-

tent, as seen in a study involving chronic stroke pa-

tients. This study reported signiﬁcant improvements

in shoulder, wrist, and elbow movements, though not

uniformly across all tested ranges, suggesting the po-

tential of VR games in enhancing speciﬁc motor func-

tions (Shahmoradi et al., 2021).

The role of serious games in upper limb rehabili-

tation is further emphasized by Provenc¸a et al., who

analyzed 38 studies on technological gaming plat-

forms for patients with neuromotor disorders. Al-

though only a fraction of these studies reported im-

provements, this emerging paradigm in rehabilitation

signiﬁes a shift towards integrating serious gaming

technologies into therapeutic practices (Proenc¸a et al.,

2018). This is corroborated by another review that

found serious games to be more effective than conven-

tional treatments in upper limb recovery post-stroke,

highlighting their long-term effectiveness in main-

taining improvements (Doumas et al., 2021).

Exploring serious games further, a randomized

controlled trial using the Leap Motion Controller

for patients with multiple sclerosis demonstrated sig-

niﬁcant improvements in coordination and dexter-

ity, illustrating the potential of these technologies

in diverse patient populations (Cuesta-G

omez et al.,

2020). In a similar vein, Herne et al.’s assessment

of stroke survivors’ engagement with a VR-based

serious game underscored the importance of feed-

back and personalized experiences to optimize en-

gagement (Herne et al., 2022), while Baluz et al.’s

development of the Rehabilite Game received posi-

tive feedback from both physiotherapists and patients,

pointing to its suitability as a complementary therapy

tool (Baluz et al., 2022).

Innovative approaches like the wearable multi-

modal serious games for hand movement training in

stroke patients, which combine various sensor tech-

nologies for movement classiﬁcation, have shown an

81.0% accuracy rate and heightened patient enthusi-

asm compared to conventional methods, further indi-

cating the growing potential of such tools in rehabil-

itation (Song et al., 2022). Finally, a study on a VR

prototype using hand gestures for interaction in up-

per limb rehabilitation revealed a preference for hand-

based interaction among participants, especially those

with motor problems, suggesting its potential in en-

hancing motivation and facilitating home-based exer-

cise therapy (Juan et al., 2023).

3 ADAPTED HANDBALL

SERIOUS GAME FOR UPPER

LIMB REHABILITATION

3.1 Architecture

A schematic overview of the developed system archi-

tecture for upper limb sports practice and rehabilita-

tion is presented in Figure 1. It features a multi-layer

architecture, each layer being independent, serving a

clear function, and interconnected within the overall

system.

The system involves two main actors: patients and

therapists who supervise the rehabilitation process.

The top layer, or presentation layer, is where patients

directly interact. Patients can immerse themselves in

a virtual environment using a VR headset, speciﬁcally

the Oculus Quest 2 model. This device allows patients

to experience a realistic recreation of a sports hall and

a visual interface to conﬁgure training that includes a

series of exercises. The layer also facilitates train-

ing calibration by allowing patients to stretch their

Virtual Reality-Based Adapted Handball Serious Game for Upper Limb Rehabilitation in Spinal Cord Injured Patients

339

Application

Layer

Game

Controller

Hand Interaction

Controller

Kinematics

Data Logging

Achievements

User Interaction Reception

Data Analysis

Exercise Result Viewer

Overview of the

Patient’s Evolution

EXECUTION

Presentation & Sensor Layers

VR - Headset

Visual

Interface

Game

Configuration

Calibration

Unity

Framework

PATIENT

BIO

Hand L/R

Type of injury

THERAPIST

Data Layer

Presentation Layer

Hand, head & trunk

tracking

Persistence Layer

Medical

History

Figure 1: General architecture of the developed system.

arms fully, marking the maximum horizontal and ver-

tical reach, with the system subsequently adapting

ball throws to the patient’s limits. Virtual worlds’ sig-

niﬁcant advantage is the ease of creating accessible

spaces tailored to each patient’s unique needs.

On the other hand, the application layer manages

the control logic, comprising various controllers: a)

The Game Controller oversees the game dynamics

and overall training, controlling the virtual environ-

ment elements and interacting with the physics en-

gine. b) The Hand Tracking module accurately tracks

the patients’ hands. Most VR headsets have associ-

ated controllers equipped with accelerometers and gy-

roscopes, enabling precise hand movement registra-

tion. However, for patients with spinal cord injuries,

using these controllers is not feasible due to motor

limitations and grip difﬁculties. Therefore, hands-

free interaction and accurate hand tracking during ex-

ercises are essential. Moreover, the system continu-

ously monitors head position, determinable through

the VR headset sensors, and infers the trunk position

through inverse kinematics. c) The Hand Interaction

Controller is crucial for modeling the real hand’s in-

teraction with virtual elements, like the thrown balls,

necessitating the determination of intersections be-

tween hands and balls, representing a save to be

counted.

This layer communicates with the data layer,

which receives information from the tracking module

and virtual element interactions, recording kinematics

during exercises and achievements made during train-

ing. All this information becomes part of the patient’s

clinical history in the persistence layer.

Lastly, in a separate application, the therapist can

analyze all recorded data, objectively determining the

patient’s progress in the rehabilitation process. This

analysis involves evaluating performance based on

achievements and a detailed examination of the three-

dimensional space kinematics of hands, head, and

trunk.

3.2 Immersive Environment for

Exercise Execution

In the design of the virtual training space, achieving

an appropriate level of realism is crucial to ensure

that the patient experiences an almost full sense of

immersion. This aims to give the patient the feeling

of being transported to a typical environment where

handball sports practice takes place. In such a set-

ting, lighting is essential, especially considering that

ball throws occur and the objective is to intercept as

many as possible. Thus, the illumination of moving

objects and shadow projections should be optimized

to provide clear visibility for the participant undergo-

ing the exercise. Adequate visibility empowers the

user to precisely ascertain the position and distance

of the target.

A handball court was designed in accordance with

the ofﬁcial dimensions speciﬁed by the International

Handball Federation (IHF), measuring 40 meters in

length and 20 meters in width (see Figure 2). Tra-

ditional handball goals have dimensions of 2 meters

in height and 3 meters in width. However, the hand-

ball goal featured in the virtual environment adheres

to adapted handball standards, measuring 1.60 meters

in height and 3 meters in width, catering to users who

engage in the sport from a wheelchair (see Figure 3).

Regarding illumination, Baked Lightmaps have

been implemented in Unity. Baked Lightmaps is a

rendering technique used to enhance realism and ef-

ICEIS 2024 - 26th International Conference on Enterprise Information Systems

340

Figure 2: Design of the immersive training environment.

Figure 3: Comparison of handball and adapted handball

goals.

ﬁciency in representing lighting in 3D scenes. These

lightmaps store precomputed lighting information and

apply it to the scene. The primary advantage of this

lighting technique is its ability to reduce the com-

putational load on the CPU and GPU in real-time.

As the lighting is precomputed and stored within the

lightmaps, there is no need to perform intricate light-

ing calculations for every frame, thus enhancing game

performance.

Additionally, Dynamic Lights have been incorpo-

rated, which are computed for every frame and re-

spond in real-time to changes in the scene. A series

of Spotlights have been introduced and emit light in

a speciﬁc direction, forming a cone of light. These

lights are ideal for simulating light sources with a de-

ﬁned direction, such as the spotlighting in the scene.

The use of these types of lights has been minimized

to prevent excessive resource consumption that could

lead to a decline in application performance. An un-

wanted decline in system performance may lead to

observable disruptions, such as dropped frames and

interrupted playback. These technical inconsistencies

not only hamper the smooth operation of the appli-

cation but also induce physical discomfort in users,

manifesting as sensations of dizziness or disorien-

tation. Such adversities compromise the immersive

quality of the user experience, detracting from its

overall enjoyment and appeal.

3.3 Throwing and Interception of

Virtual Balls

The developed system features two game and train-

ing modes for physical exercise practice: a) Maxi-

mum Number of Saves; in this mode, players are chal-

Figure 4: Ball and cannon prefabs designed in the applica-

tion.

lenged to make as many saves as possible within a

ﬁxed time limit. The patient’s skill is gauged based

on the quantity of saves achieved during this period,

providing a measure of their performance. b) Maxi-

mum Time; this mode sets a limit on the number of

goals that can be conceded. The focus here is on

the length of time the patient can keep playing be-

fore reaching this limit. The duration of the exercise

session extends with each successful save, turning the

number of saves into a determinant of the overall ex-

ercise time.

In all these training modes, there are two funda-

mental aspects for them to take place: i) the throw-

ing of balls and ii) the interaction between the pa-

tient’s real hands and the virtual balls, an event trans-

lated as a goalkeeper’s save. For these two aspects,

there are three key components: a) cannons responsi-

ble for throwing the balls, b) the virtual balls, and c)

the user’s hands performing the exercise.

The system represents several moving cannons

positioned on the handball court in the classic posi-

tions where real players usually stand. At this point, it

is worth introducing the concept of a prefab. A prefab

is a type of resource that allows creating, conﬁguring,

and reusing objects in a Unity scene. Using prefabs is

ideal when the same object needs to be used multiple

times. Prefabs ensure that objects sharing the same

prefab have the same properties and conﬁgurations.

Handball balls and the cannons that will propel these

balls are managed using prefabs. Figure 4 displays

the designed prefabs.

The ball throws are determined by the trajectory

drawn between a cannon, whose position determines

the starting point, and an endpoint calculated from

one of the shooting areas. As illustrated in Figure

5, the upper half of the goal is divided into an array

of regions or zones, spanning from the base of the

player’s trunk to their head. These zones are labeled

from A1 to A5 and B1 to B5. The regions located

at the extremes: A1, A5, B1, and B5, require greater

trunk mobility. Recording saves and goals in each of

these zones allows the therapist to identify strengths

and weaknesses in the patient’s limb movements, dif-

ferentiating between the left and right hand and arm.

To draw the ball throwing trajectory, a subrou-

tine in Unity has been implemented, utilizing the Lerp

Virtual Reality-Based Adapted Handball Serious Game for Upper Limb Rehabilitation in Spinal Cord Injured Patients

341

Figure 5: Target points where the ball throws are directed.

function. The Lerp function (meaning linear interpo-

lation) is a useful tool for interpolating between two

values over a period of time. The algorithm 1 shows

a simpliﬁed implementation of the Lerp function for

trajectory animation. In the Lerp function, each argu-

ment represents:

• Starting Point: represents the cannon’s position

from which the throw is made.

• End Point: represents the throw’s ﬁnal position

at the goal.

• t: represents the interpolation value. A value of

0 represents the Starting Point, a value of 1 rep-

resents the End Point, and any value between 0

and 1 produces an interpolation between the two

values.

Data: Starting point, End point, Movement

speed, Total distance

Result: New position after interpolation

Initialization: Set initial value for

distanceTraveled;

while Update is called do

distanceTraveled += movementSpeed *

Time.deltaTime;

ﬂoat t = distanceTraveled / totalDistance;

Vector3 newPosition =

Vector3.Lerp(startingPoint, endPoint, t);

transform.position = newPosition;

end

Algorithm 1: Simpliﬁed use of the Lerp function.

On the other hand, the handball ball’s prefab con-

tains a vital component for game logic. This element

is the Collider. A Collider deﬁnes a Game Object’s

collision shape, in this case, the ball. It is used to de-

tect collisions and trigger events when other objects

come into contact with the Collider. The handball

used in this project implements a Sphere Collider, at-

taching a spherical collision shape to the ball.

Finally, the user’s hands in the handball scene are

represented by the Custom Hand component. This

component, in turn, consists of a Collider that will

be set up to manage interaction with the balls. In the

Collider, the isTrigger property is activated so that the

OnTriggerEnter function always runs when there is a

collision with another Collider. It is necessary to en-

sure in the OnTriggerEnter function that the colliding

Collider is the handball. To address this issue, Unity

tags are used to distinguish between virtual compo-

nents. Thus, only the intersections between the virtual

hands (which match the real ones in position and ro-

tation) and the virtual ball are counted as a new save.

If there is an intersection with an object that has a dif-

ferent tag, it does not trigger the event, and therefore,

it does not result in the recording of a new save

The user interacts at all times through their hands,

without the need to hold joysticks or external con-

trollers. This forms a natural mode of interaction,

closely mimicking real sports practice. This detail is

crucial for patients with spinal cord injuries who have

motor limitations and difﬁculty in executing grips cor-

rectly. Some patients even have serious difﬁculties

in lifting weights, making hand-free interaction a top

priority requirement.

3.4 Recording of Kinematics

In the realm of rehabilitation and therapeutic exer-

cises, leveraging advanced technologies such as com-

puter vision algorithms from the Meta API and Ocu-

lus Quest 2 has revolutionized the capture of detailed

kinematic data. This sophisticated approach employs

deep neural networks to predict and locate the posi-

tions of a person’s hands and other signiﬁcant land-

marks like the joints of the hands. The resultant data

is a comprehensive 3D model that encapsulates the

nuanced conﬁguration and surface geometry of the

hands and ﬁngers, offering a 26 degree-of-freedom

pose reconstruction.

This tool facilitates a recording of movement and

orientation of various body parts, providing therapists

with a profound window to analyze the exercises ex-

ecuted by patients. Such precise data aids in deliv-

ering an objective evaluation of a patient’s progress,

bridging the gap between subjective observations and

quantiﬁable metrics.

In our proposal, the dataset generated is rich and

multifaceted, encompassing several relevant compo-

nents that contribute to a comprehensive understand-

ing of the user’s movements. Fundamental variables

such as Frame and Time establish a foundational con-

text, anchoring each data point within a coherent se-

quence.

ICEIS 2024 - 26th International Conference on Enterprise Information Systems

342

Figure 6: Patient and therapist testing the developed soft-

ware during the experimentation session.

For capturing the intricacies of head movements,

variables like HeadPosition x, HeadPosition y, Head-

Position z, HeadRotation x, HeadRotation y, and

HeadRotation z are employed. They detail both the

spatial position and the angular orientation of the

user’s head during the interaction.

On the other hand, the intricate movements and

orientations of the user’s hands are captured using a

series of variables. For the right hand, aspects such

as its detection, position in space, orientation, veloc-

ity, and wrist rotation are detailed by HandDetect-

edR, HighConﬁdenceR, RHandPosition x, RHandRo-

tation x, and RWristTwist, among others. The left

hand’s kinematics mirror this structure, with variables

like HandDetectedL, LHandPosition x, and LWrist-

Twist playing analogous roles.

4 EXPERIMENTATION

The experimental session conducted at the HNPT in-

volved a total of 9 participants, consisting of 3 SCI pa-

tients and 6 healthy individuals (HNPT staff).All the

patients had a chronic cervical SCI (more than one

year after the injury). Additionally, the patient with

ID:1 regularly engages in adaptive sports, ensuring a

commendable overall mobility.

Before starting the experimentation, participants

were thoroughly briefed on the nature of the exer-

cises. Subsequently, with the VR headset on, they

underwent an initial immersion into the virtual world

to get acquainted with it, without any data being

recorded yet.

Each participant, seated in a traditional chair or

wheelchair, underwent individual assessments and

participated in exercises designed to measure their

progress. The primary objective was to prevent a

speciﬁed number of shots from becoming goals. The

goalkeeper, whether a patient or a therapist, aimed to

maximize their successful saves. The system tracked

the duration the goalkeeper remained active. The ex-

ercise persisted until the predetermined number of

goals were conceded by the goalkeeper (15 in par-

ticular). As the goalkeeper’s duration in the exer-

cise extended, due to successfully preventing goals,

the difﬁculty level progressively heightened with balls

launched at faster speeds and greater frequencies.

On the other hand, the healthy individuals partic-

ipating in the study are experienced in rehabilitating

SCI patients. They are professionals specializing in

ﬁelds such as biomedical engineering, occupational

therapy, and physical therapy. The gameplay mechan-

ics and objectives for these individuals were identical

to those applied to the patients, maintaining a consis-

tent approach across all participants in the study.

4.1 Sample Description

This section outlines the data pertaining to the partic-

ipants, including their age, patient type, and type of

injury. Table 1 summarizes the general characteristics

of each participant.

Table 1: Participant data in the HNPT study.

ID Age Type Injury

1 47 Patient SCI

2 31 Healthy -

3 24 Healthy -

4 20 Patient SCI

5 45 Healthy -

6 44 Healthy -

7 29 Healthy -

8 23 Healthy -

9 22 Patient SCI

Two groups of subjects are distinguished: 88.9%

of the participants had occasionally used VR de-

vices previously. Furthermore, 11.1% used them fre-

quently. 100% of the subjects had used a VR device

at some point. It is also noteworthy that 33.3% of

the study participants were HNPT patients, and the

remaining 66.6% were healthy individuals.

44.4% of the participants were aged between 20

and 25 years, 44.4% were between 25 and 30 years

Virtual Reality-Based Adapted Handball Serious Game for Upper Limb Rehabilitation in Spinal Cord Injured Patients

343

old, 11.1% were between 30 and 40 years old, and

the remaining 33.3% were above 40 years old.

4.2 Results

After conducting the study, the results obtained are

displayed in Table 2. This table presents the dura-

tion times of each participant in each of the two series

conducted, along with the number of saves made in

each series. The longer the duration time, the better

the performance. The last column displays the per-

centage of improvement or worsening that the partic-

ipants achieved in the number of saves made. Finally,

the last row presents the arithmetic mean of all the

measurements recorded.

As can be seen in the table, the degree of improve-

ment is positive in 8 out of 10 cases, with an over-

all average improvement of 29%. Only two subjects,

5 and 8 (both healthy subjects), showed worse per-

formance and results after the second session. The

main cause was fatigue after a short time elapsed be-

tween sessions S1 and S2. In the case of SCI patients,

subjects 1, 4 and 9, the degree of improvement was

42.8%, 31.2% and 33.3% respectively, the arithmetic

mean being 35.7%.

On the other hand, it was observed that the aver-

age time taken in S1 was 80 seconds, whereas in S2, it

averaged at 90 seconds. This indicates an overall in-

crease in time, showing a progression from S1 to S2

and, in consequence, a better performance. Among

the patients, particularly subjects 1, 4, and 9, there

was a variance in the degree of time growth noticed.

The patients exhibited a diverse range of time incre-

ments, with a mean growth of approximately 31.84%.

Speciﬁcally, P1 showed an increase of 27.27%, P4

had an increase of 53.97%, and P9 recorded a growth

of 14.29%.

In Figure 7, a mosaic layout presents the kine-

matic recordings for each of the subjects who partici-

pated in the study. The red dots represent movements

of the right hand, the blue dots denote movements of

the left hand, and the yellow dots depict head move-

ments. Among the subjects, those identiﬁed by IDs

1, 4, and 9 are patients with SCI, while the rest are

healthy participants. This visual representation serves

to compare and contrast the motion patterns and ten-

dencies between the two groups.

In general, it appears that patients with spinal cord

injuries tend to demonstrate a higher concentration of

movements in the center of the chart for both hands,

potentially reﬂecting a limitation in the range of mo-

tion. Additionally, their head movements display a

notable restriction in breadth, focusing their attention

on speciﬁc areas without encompassing the full vi-

sual ﬁeld. On the other hand, healthy participants

exhibit a broader dispersion of movements with both

hands, suggesting an enhanced capability to respond

to stimuli from various directions. Their more dis-

persed head movements indicate an aptitude for ac-

tively tracking the trajectory of the ball across diverse

directions and elevations.

All data collected during the experiment are

publicly available in the GitHub repository:

https://github.com/AIR-Research-Group-UCLM/

Rehab-HandballVR. The repository is organized into

nine folders, each corresponding to an individual

participant in the experiment. Each folder contains a

Historical.CSV ﬁle detailing the number of throws,

saves, and goals detected, as well as a TrackingData

subfolder, which includes a CSV ﬁles with the

kinematics recorded for each exercise performed.

4.3 Participant Feedback

At the end of the exercise sessions, each participant

was asked to complete a simple questionnaire consist-

ing of three questions designed to measure their mo-

tivation, sense of achievement and satisfaction, and

comfort/well-being in the virtual space.

Q1 - ”After playing, I feel motivated to continue

playing or to return to the game at another time.”

Participants could respond with a value between 1-5,

where 1 signiﬁes ”strongly disagree” and 5 the oppo-

site.

Q2 - ”I have experienced a sense of accomplish-

ment and satisfaction upon completing the exercises.”

The response format is the same as Q1, with a value

between 1-5.

Q3 is designed to measure the user’s degree of

comfort and well-being: ”I have experienced dizzi-

ness, disorientation, or balance difﬁculties while play-

ing.” Possible answers are: a) I have not experi-

enced any symptoms at all, b) I have experienced mild

symptoms, c) I have experienced moderate symp-

toms, d) I have experienced signiﬁcant symptoms, e)

I have experienced very intense symptoms.

Figure 8 graphically summarizes the responses to

the above questions. Regarding motivation, 7 out of

9 participants gave the highest score for motivation

level, while 2 rated it 4, still above the average. In

response to Q2, 5 of the 9 participants experienced the

highest level of achievement and satisfaction, 3 rated

it 4, and only one rated it 3. Finally, the results for

Q3 indicate that the immersive experience was highly

enjoyable, with only one participant experiencing a

moderate sensation of dizziness during the exercises.

ICEIS 2024 - 26th International Conference on Enterprise Information Systems

344

Table 2: Summary of the evaluation session at HNPT.

Identiﬁer Time S1 (seconds) Saves S1 Time S2 (seconds) Saves S2 Degree of Improvement

1 99 49 126 70 42.8%

2 100 38 102 49 28.9%

3 96 45 108 54 20%

4 63 16 97 37 31.2%

5 96 51 65 20 -31.4%

6 77 28 122 56 100%

7 64 19 68 23 21%

8 74 25 66 21 -16%

9 56 12 64 16 33.3%

Average 80 31 90 40 29%

Figure 7: 2D representation of participants’ hand and head positions during the second training session.

5 CONCLUSIONS

This paper presents a VR-based serious game de-

signed for upper limb rehabilitation, replicating hand-

ball goalkeeper training for wheelchair users. It acts

as a complement to traditional therapy methods, pro-

viding patients with a more appealing and motiva-

tional environment for exercise.

A key strength of this work is the accurate record-

ing of kinematics during exercise performance. The

data gathered are crucial for therapists to objectively

assess patient progress. Parameters such as range

of motion (ROM), movement speed and smoothness,

repetition count and consistency, as well as the ac-

Virtual Reality-Based Adapted Handball Serious Game for Upper Limb Rehabilitation in Spinal Cord Injured Patients

345

Figure 8: Graphical representation of the responses

provided by participants regarding motivation, sense of

achievement and satisfaction, and comfort/well-being.

curacy and precision of movements, balance, or pos-

tural control can be calculated from this data. For

patients, this information is invaluable as it informs

them whether they are performing the movements

correctly, a vital aspect since the success of rehabil-

itation partly relies on exercise precision.

The system has been tested in a real-world setting

at the Hospital Nacional de Parapl

ejicos de Toledo

with SCI patients and has also been validated by ther-

apists. The conducted experimentation has some lim-

itations, such as a small participant sample and a sin-

gle exercise session in one day. However, the initial

goal of this preliminary work was mainly to analyze

the correct functionality of the system and ensure it

provides a safe testing environment for patients.

After veriﬁcation by the HNPT staff regarding the

system’s safety for patients and the reliability of the

data collected, future sessions are planned to involve

larger patient samples and regular repetitions to eval-

uate the effectiveness and impact of the proposal on

mobility recovery.

ACKNOWLEDGEMENTS

This work has been founded by the Spanish Ministry

of Science, Innovation and Universities under the Re-

search Project: Platform for Upper Extremity Reha-

bilitation based on Immersive Virtual Reality (Rehab-

Immersive), PID2020-117361RB-C21 and PID2020-

117361RB-C22.

REFERENCES

Baluz, R., Teles, A., Fontenele, J. E., Moreira, R., Fialho,

R., Azevedo, P., Sousa, D., Santos, F., Bastos, V. H.,

and Teixeira, S. (2022). Motor rehabilitation of upper

limbs using a gesture-based serious game: Evaluation

of usability and user experience. Games for Health

Journal, 11(3):177–185.

Cuesta-G

omez, A., S

anchez-Herrera-Baeza, P., O

na-

Simba

na, E. D., Mart

ınez-Medina, A., Ortiz-Comino,

C., Balaguer-Bernaldo-de Quir

os, C., Jard

on-Huete,

A., and Cano-de-la Cuerda, R. (2020). Effects of vir-

tual reality associated with serious games for upper

limb rehabilitation in patients with multiple sclerosis:

Randomized controlled trial. Journal of neuroengi-

neering and rehabilitation, 17:1–10.

de Los Reyes-Guzm

an, A., Lozano-Berrio, V., Alvarez-

Rodriguez, M., Lopez-Dolado, E., Ceruelo-Abajo, S.,

Talavera-Diaz, F., and Gil-Agudo, A. (2021). Rehab-

hand: Oriented-tasks serious games for upper limb re-

habilitation by using leap motion controller and target

population in spinal cord injury. NeuroRehabilitation,

48(3):365–373.

Dom

ınguez-T

ellez P, Moral-Mu

noz JA, S. A. C.-F. E. L.-

A. D. (2020). Game-based virtual reality interventions

to improve upper limb motor function and quality of

life after stroke: Systematic review and meta-analysis.

Games for Health Journal, 9(1):1–10.

Doumas, I., Everard, G., Dehem, S., and Lejeune, T.

(2021). Serious games for upper limb rehabilitation

after stroke: a meta-analysis. Journal of neuroengi-

neering and rehabilitation, 18:1–16.

Herne, R., Shiratuddin, M. F., Rai, S., Blacker, D., and

Laga, H. (2022). Improving engagement of stroke

survivors using desktop virtual reality-based serious

games for upper limb rehabilitation: A multiple case

study. IEEE Access, 10:46354–46371.

ICEIS 2024 - 26th International Conference on Enterprise Information Systems

346

Herrera, V., Reyes-Guzm

an, A., Vallejo, D., Castro-Schez,

J. J., Monekosso, D. N., Gonz

alez-Morcillo, C., and

Albusac, J. (2023a). Performance analysis for upper

limb rehabilitation in non-immersive and immersive

scenarios. In International Conference on Enterprise

Information Systems, ICEIS, pages 231–242.

Herrera, V., Vallejo, D., Castro-Schez, J. J., Monekosso,

D. N., de los Reyes, A., Glez-Morcillo, C., and Al-

busac, J. (2023b). Rehab-immersive: A framework to

support the development of virtual reality applications

in upper limb rehabilitation. SoftwareX, 23:101412.

Ikbali Afsar, S., Mirzayev, I., Umit Yemisci, O., and Cosar

Saracgil, S. N. (2018). Virtual reality in upper extrem-

ity rehabilitation of stroke patients: A randomized

controlled trial. Journal of Stroke and Cerebrovascu-

lar Diseases, 27(12):3473–3478.

Juan, M.-C., Elexpuru, J., Dias, P., Santos, B. S., and

Amorim, P. (2023). Immersive virtual reality for upper

limb rehabilitation: comparing hand and controller in-

teraction. Virtual Reality, 27(2):1157–1171.

Karamians, R., Profﬁtt, R., Kline, D., and Gauthier, L. V.

(2020). Effectiveness of virtual reality- and gaming-

based interventions for upper extremity rehabilitation

poststroke: A meta-analysis. Archives of Physical

Medicine and Rehabilitation, 101(5):885–896.

Lee, S. H., Jung, H.-Y., Yun, S. J., Oh, B.-M., and Seo, H. G.

(2020). Upper extremity rehabilitation using fully im-

mersive virtual reality games with a head mount dis-

play: a feasibility study. Pm&r, 12(3):257–262.

Pereira, M. F., Prahm, C., Kolbenschlag, J., Oliveira, E.,

and Rodrigues, N. F. (2020). A virtual reality serious

game for hand rehabilitation therapy. In 2020 IEEE

8th International Conference on Serious Games and

Applications for Health (SeGAH), pages 1–7. ISSN:

2573-3060.

Proenc¸a, J. P., Quaresma, C., and Vieira, P. (2018). Seri-

ous games for upper limb rehabilitation: a systematic

review. Disability and Rehabilitation: Assistive Tech-

nology, 13(1):95–100.

Shahmoradi, L., Almasi, S., Ahmadi, H., Bashiri, A., Azadi,

T., Mirbagherie, A., Ansari, N. N., and Honarpishe, R.

(2021). Virtual reality games for rehabilitation of up-

per extremities in stroke patients. Journal of bodywork

and movement therapies, 26:113–122.

Song, X., van de Ven, S. S., Chen, S., Kang, P., Gao, Q.,

Jia, J., and Shull, P. B. (2022). Proposal of a wearable

multimodal sensing-based serious games approach for

hand movement training after stroke. Frontiers in

Physiology, 13:811950.

Spooren, A. I. F., Janssen-Potten, Y. J. M., Kerckhofs, E.,

Bongers, H. M. H., and Seelen, H. A. M. (2011). Eval-

uation of a task-oriented client-centered upper extrem-

ity skilled performance training module in persons

with tetraplegia. Spinal Cord, 49(10):1049–1054.

Vogt, S., Skjæret-Maroni, N., Neuhaus, D., and Baumeister,

J. (2019). Virtual reality interventions for balance pre-

vention and rehabilitation after musculoskeletal lower

limb impairments in young up to middle-aged adults:

A comprehensive review on used technology, balance

outcome measures and observed effects. International

Journal of Medical Informatics, 126:46–58.

WHO (2021). Spinal cord injury. Accessed: 17-Oct-2023.

Virtual Reality-Based Adapted Handball Serious Game for Upper Limb Rehabilitation in Spinal Cord Injured Patients

347