Keywords

1 Introduction

Childhood cerebral palsy (CP) is considered one of the most common causes of motor disability in children. CP refers to a group of non-progressive movement and posture alterations due to a brain injury that occurred during fetal brain development, childbirth or in the first years of child life [1]. Studies have shown that the prevalence of CP worldwide is 2.11 per 1000 live births [2]. Different types of CP within motor manifestations can be positive and negative symptoms [3, 4]. The positive symptoms include spasticity and involuntary movement and negative symptoms include weakness, impaired dexterity and selective motor control.

Physical therapy in children with mild motor impairment is one element in a childhood development program where therapists have to incorporate authentic efforts to achieve stimulating, varied and quality environment, so a child with cerebral palsy, like any child, needs new experiences and interaction with the outside world in order to learn new things [5, 6].

Children with CP are less involved in physical and recreational activities compared to children with the same age who have a typical development. [7] The low physical activity that presents the CP children could influence to develop chronic diseases in adulthood, while children who have developed physical activities tend to have happiness and better life quality [8].

Muscle strengthening exercise can improve muscle strength of the lower limbs in children with CP without increasing spasticity [9, 10]. Cycling can be effective to increase physical activity and improve the life quality of children with CP. Studies with stationary bicycles show improvements [11, 12] in terms of strength, endurance [13], muscle activation [14], life quality [15], and bone mineral density [16]. Clinical experience suggests that children with CPs that are within the GMFCS level I-II can learn how to ride a double bicycle accordance with the appropriate conditions [17].

Virtual environments have become an interactive tool that is used for immersion and motivation of patients [18,19,20]. The environment created with virtual reality present environments very close to reality which provide visual and auditory feedback. This features can be manipulated in a precise and systematic way and allow individualized training in motor learning [21, 22].

There are few studies related to the interaction of bicycles and virtual environments and especially dedicated to children [23,24,25]. Therefore, the purpose of this work presents a low cost system built with a bicycle and a 3D virtual environment. The system allows children to strengthen the motor control development while it is offering fun and training the lower limbs. The paper has the next sections: Introduction, System Implementation, Test and Results, Conclusions.

2 System Implementation

The System implementation has two stages. The first stage use an arduino for data acquisition. The second stage presents the creation of virtual environments. In Fig. 1 the block diagram of the implemented system is presented.

Fig. 1.
figure 1

Implemented system block diagram.

Full size image

2.1 Data Acquisition

A small conventional bicycle is used in children between 5–9 years old, which has been immobilized to work in a static way. Variables are taken from the bicycle as angular velocity (RPM), distance and turn of the handlebar (left right). In data acquisition is used an embedded system with Arduino mega, the variables obtained are presented below:

  1. (i)

    Angular Velocity. The angular velocity of a rotating body related to an axis is calculated in revolutions per minute (RPM). In the case of the bicycle, a Hall Effect sensor and a permanent magnet are used in the wheel. The equation used to obtain RPM is detailed below:

    $$ RPM = \frac{f*60}{\# \;of\;magnets} $$
    (1)

    • f = Sensor frequency (Hz)

    • # of magnets = According to the magnets located on the rotating shaft (bicycle wheel, for this application 1 magnet).

  2. (ii)

    Turn of the bicycle. An incremental encoder coupled to the bicycle handle obtains the bicycle rotation. The incremental encoder has a resolution of 36 ppr (pulses per revolution). Encoder generates a displacement of −9 ppr to the left and 9 ppr to the right according to 90° rotation on each side.

2.2 Virtual Environment Development

The virtual environment is designed in Unity. The scene has been mounted with 2 tracks. The idea of combining dexterity on the handle and user driving force. So in the slopes the user force increase on the pedal Fig. 2a, while the curves Fig. 2b, demand the user’s motor coordination.

Fig. 2.
figure 2

Track #1 hard difficult.

Full size image

2.3 3D Bicycle Development

To the original 3D model is added 4 Wheel Colliders that simulate the dynamics and wheels collisions. A rectangular block simulates the size and mass of the bicycle Fig. 3. This block is also used as a sensor to determine the slope in the ground where the mobile is located. The variable is send to the script “Bluetooth Controller” and it will be send to the hardware later. The driver dynamic by bicycle colliders is similar to carlike mobile with four traction wheels and two steering wheels, which allows obtaining the effects of skidding, dynamics and inertia like a truth bicycle.

Fig. 3.
figure 3

Virtual bicycle development.

Full size image

The bicycle control script assigns the maximum value of rotation angle and torque to make the movement and bend to the model respectively. To perform the brake action isn’t enough just to assign the torque variable to zero, because the inertia will keep moving the object. It is necessary to assign a value of 1000 to the variable Brake so that the bicycle stops unexpectedly. The RPM and twist values come from the Bluetooth Controller script that receives data from the serial frame from the hardware.

The spin data comes from the hardware and has a range of [−9.9]. The data less than zero imply to turn to the left and the greater than zero indicate to turn to the right. The value zero indicates the central position of the handlebar. To plot the movement the original values are scaled to a range of [−90, 90]. The minimum movement resolution is 10°, Fig. 4.

Fig. 4.
figure 4

Handlebar data script

Full size image

The real direction data is transmitted to 3D model handlebar by an array in the model hierarchy. It uses the pivot position in local coordinates of the parent object, which has the handlebar elements, Fig. 5.

Fig. 5.
figure 5

Handlebar 3D model

Full size image

2.4 Bluetooth Controller

The script manages the Bluetooth connection. It has the next functions:

  1. (i)

    Connect the hardware Bluetooth device using the fixed device MAC address.

  2. (ii)

    Receive data from the hardware. It analyze the frame and isolate the RPM and handlebar rotation values.

  3. (iii)

    Send the ground slope data to the hardware.

  4. (iv)

    Calculate the velocity and distance values from the RPM data.

The Status Connection, RPM, Speed and distance values are used by the HUD Script to display them in the user’s Head’s Up Display, Fig. 6. This data is necessary for the user to know in real time the data of movement that realizes in the virtual environment. In addition, data is stored by each session like activity history and will be available for statistical analysis.

Fig. 6.
figure 6

Virtual system in operation.

Full size image

3 Test and Results

3.1 Test

Operation tests are performed in the implemented system. The participants receive an explanation about the virtual environment operation especially about the curves indicated in the interface. Four healthy children between 5 and 9 years old used the system. They work with the system like a traditional bicycle and steer the bicycle according to the virtual environment movement.

To evaluate usability in virtual systems several scientific works authors validate the virtual tools for rehabilitation through usability tests [26,27,28,29]. The studies allow determining the user acceptance, this information is of great importance to determine the security, sensation, discomfort when a virtual system is used. The test for usability evaluation is SEQ developed by Gil-Gómez [27]. Has 14 questions, where 13 questions have a score of 1 to 5 points. After the game is finished, the children are asked to perform SEQ usability surveys to determine the system’s acceptance.

3.2 Results

Implemented Interface Operation

Verifies the operation of the system in each one of the stages in the implemented levels. Figures 7 and 8 show the sequences performed by the users where the interaction between the children and the environment is verified. The variables obtained from the bicycle allow exercising the lower limbs. The exercise in arms and hands is performing with the movement in the curves.

Fig. 7.
figure 7

The system in operation presents interaction of the interface according to the movements made by the user. (Game level 1)

Full size image
Fig. 8.
figure 8

The system operation presents the interface interaction according to user movements. (Game Level 2)

Full size image

The implemented system allows to obtain angular velocity, time and distancing data. This information required to determine the specialist therapies time and the user evolution. The data obtained from several tests are presented in Table 1.

Table 1. Operation test data
Full size table

System Usability Result

The result of system usability is presented. Four children between 5 and 9 years old completed the SEQ survey. The outcomes (54.5 ± 0.34) in Table 2 are according to [27]. If SEQ result is within a range of 40–65, the application is considered for rehabilitation. This result indicates that the virtual system has acceptance to be used in rehabilitation.

Table 2. SEQ mean and standard deviation
Full size table

4 Conclusions

An immersive virtual system was developed which helps to strengthen legs in children. Two realistic scenes was created which based on competence tracks, with different difficulty levels (curves and slopes). Bluetooth makes the user interaction with the system. This is an advantage because is a wireless system.

The virtual system allows obtaining angular velocity, traveled distance and data execution time to plan the routines of exercises performed by the rehabilitators. Thus, the system operation and usability based on SEQ (54.5 ± 0.34) determined that users feel immersed and enjoy the game. Users don’t have inconveniences or difficulty while they were using it.