Grip Strength and Body Balance in Static and Dynamic Push Walkers Measurement Using Force Sensing Resistors and Kinect System

. Most parents use push walkers to encourage toddlers in gross and fine motor stimulation to develop basic walking skills. Push walkers are a choice for parents other than baby walkers to help toddlers under one year learn to walk. The push walker options for toddlers include static (the shaft moves in place) and dynamic (moving mobile). This study aimed to compare the effect of push walkers between static and dynamic types on motor stimulation of toddlers using biomechanics by measuring hand grip strength and body balance. This research contributes to selecting a push walker for parents to support gross and fine motor development after agreeing on the consideration of the toddler expert panel for the toddler walking task simulation using a push walker involving nine toddlers according to body mass index. Simulation of toddler walking in two experimental sessions using static and dynamic push walkers in measuring hand grip strength using force-sensing resistors and body balance using the Kinect system with Vitruvius software. The grip strength value was tested by paired t-test to determine the significance of the grip strength value. Body balance values were tested by paired t-test to determine the significance of leg compression force and leg moment. Dynamic push walker has a greater value of grip strength than static, and p-value = 0.001 (< 0.005), providing good motor stimulation to toddlers when pushing with more muscular grip strength, allowing the toddler's hand muscles to tighten up. Static push walkers have a greater compression force and foot moment than dynamic and p-value = 0.001 (< 0.005), providing gross motoric stimulation to toddlers while walking in a balanced and stable manner, allowing the toddler's leg muscles to tighten up.


Introduction
Toddler growth and development are essential for human growth and development in life.The growth and development of toddlers can experience developmental delays that substantially impact children's health and education.The Indonesian Pediatrician Association (IDAI) states that 5-10% of toddlers experience developmental delays [1].Internal and external factors influence Toddler growth and development [2].The influence of internal factors consists of race, age, genetics, and chromosomes.The result of external factors includes the state of the social environment, economy, nutrition, and stimulation [3].The stimulation makes body tissues grow up to 80% at the age of 3 years [4].Lack of stimulation causes muscle tissue to shrink and delay development.The development consists of motor, language/speech, and emotional development.
Motor development is divided into two groups: gross and fine motor.Gross motor skills relate to skills that involve large muscle movements, such as walking, sitting independently, reluctantly, or running [5].Fine motor skills include using small muscles such as grasping, manipulating objects, and drawing [6].Toddlers can be trained to learn to walk at 9-18 months [7].The toddler's age outside the normal limits of * lobesh@gmail.comwalking is 15-18 months, and motor disturbances and balance disorders appear.Therefore, it is necessary to provide intervention and stimulation from an early age [8].Toddler motor skills in hand grip strength and body balance are trained by providing a tool like a push walker.Using a push walker stimulates the ability to walk in an upright position and hand movements to be active [9].
The baby walker products on the market consist of sitins, moonwalkers, and push walkers.The advantage of the push walker is that it lets toddlers be active in interactive learning walking and does not require a harness as a seat.The benefits of push walker products are that they help stimulate the ability of the hands and foot balance to probably have no change in the center of gravity of the feet against the floor [10], [11].Toddler motoric development in push walkers is influenced by external factors that depend on hand skills and foot balance [12].There are two types of push walkers: static and dynamic.
Static push walkers move around a stationary rotary axis, while dynamic push walkers move mobile.Different types of motion are thought to influence users' gross and fine motor development.For this reason, research is needed to investigate the effect of pushing walkers on motor development.The results of this study can contribute to determining the form of stimulation exercises and considering the design of a push walker for motor development.The push walker's effect on motor skills was measured by measuring hand grip strength and body balance.
Hand grip strength and body balance measurements were carried out when toddlers used push walkers.Previous studies have researched measuring hand grip strength and body balance.The research conducted by Biju et al. [13] used a force-sensing resistor (FSR) to measure grip force applied in different parts of the palm to the gripped object.Meanwhile, the research conducted by Bijalwan et al. [14] used the Kinect system to measure body balance through the biomechanics of movement during the walking phase.This study will use the FSR to measure hand grip strength and the Kinect system to measure toddler body balance when using a push walker.
FSR characteristic in this study is a thick polymer film that exhibits a decrease in resistance when there is an increase in surface forces.Using FSR to measure hand grip strength provides adequate results for hand grip assessment [15].FSR has the advantage of being applied to every object surface and designed according to the user's needs, including measuring the strength of a toddler's grip.The Kinect system is an input device with an RGB camera, infrared projector, and 3D depth sensor for motion recognition and detection of the human body.The research conducted by Lim [16] measured several movement activities for evaluating body balance using the Kinect system with coherent results for clinical evaluation.This effectively measures body balance for toddlers when driving a push walker.
Based on the information presented, it is necessary to conduct research on toddlers using biomechanical parameters, including compressive force on the feet and distribution of pressure on the feet, to determine the effect of push walkers between static and dynamic types on toddler motor stimulation.This study aimed to compare the product of static and dynamic walkers on the motor stimulation of toddlers using biomechanics by measuring hand grip strength and body balance.

Methods
Nine toddlers were involved in the study at the Happy Healthcare Center in Surakarta, Indonesia.Toddlers were selected according to the criteria according to research by Kuo et al. [17].They are 9-15 months old and have an average BMI according to WHO (14.5-18.3).The inclusion criteria for toddlers were born at a sufficient gestational age (9 months and 10 days) and had no physical or mental disabilities.The baby was breastfed, could not stand up or walk independently, could sit independently, and had no congenital abnormalities or past medical history, including musculoskeletal disorders, movement disorders, and experience of severe pain.The inclusion criteria for mothers were not more than 32 years of age and having one child.In this study, mothers were chosen to determine the number of samples because the population studied, especially mothers directly responsible for caring for toddlers.The data in this experiment are described in Table 1.

Experimental setup
The measurement test is carried out in several steps to measure the toddlers' strength, grip, and body balance when using the push walker.The experiment begins with setting up the measurement equipment.The process measurement test can be explained in Figure 1.The laptop specs have a quad-core Intel Core i5-8265U, 8GB LPDDR3 RAM, and a dedicated USB 3.0 port.The FSR series is built with Arduino UNO to detect grip strength.PLX-DAQ Software records the input of the FSR sensor and is connected to the laptop by USB Cable.The FSR is placed in the right hand palm by a glove.The Microsoft Kinect V2 has a highresolution 1080p RGB camera, an infrared projector, and a 3D depth sensor to measure body movements.The Kinect system is placed in the position center of the track; the Kinect system distance on the way is 170 cm; mounted in a sagittal position facing the subject; the length of the straight path is 200 cm.Kinect height of 40 cm mounted on a tripod.The subject's movement begins to be recorded when it moves on the track in the capture range of the Kinect system.Experimental setups were made to provide the best data quality by ensuring that all toddler bodies were within the Kinect operating range as the manufacturer recommended.

The protocol
Before the experiments on measuring toddlers' grip strength and body balance began, toddlers' parents were instructed to use static and dynamic push walkers.Toddlers are directed to push the walker while taking measurements.The test environment in this experiment has a flat base, is free of distractions, does not endanger the subject, and has sufficient room to move around [18].Data recording of toddler's grip strength using the FSR-402 sensor.The data is generated by the FSR sensor placed on the toddler's glove with three points on the toddler's right hand.The FSR circuit module is connected to the laptop and records the input.The grip strength data collection was done for five minutes with five repetitions.Toddlers rest for two minutes between trials.Data recording of toddlers walking for five minutes randomly with five repetitions of the test recording body posture on video images using Kinect.The Kinect system is connected to a laptop and is run with the Vitruvius software.

Data acquisition and processing
Measurement of toddler grip strength based on using FSR.The grip strength measurement is designed considering the application holding a push walker handle or a cylinder (a) (b) grasp.According to research conducted by Lee et al. [19], the area of the middle of the thumb (thumb proximal phalanx), the middle of the middle finger (middle proximal phalanx), and the palm of the little finger (fifth metacarpal) is the area that has the greatest force distribution when gripping.The FSR placement can be explained in Figure 3. Joint-based biomechanical calculations were applied to determine the sagittal segment of the treated thigh, calf, and ankle according to biomechanics theory [20].Measurement of force and moment on a toddler's body as a biomechanical model when walking using a push walker consists of the palms, forearms, upper arms, shoulders, thighs, shank, and ankles.Kinematic measurements on toddler body segments consist of palm sagittal flexion/extension, forearm sagittal flexion/extension, upper arm sagittal flexion/extension, shoulder sagittal flexion/extension, thigh sagittal flexion/ extension, sagittal thigh abduction/adduction, thigh sagittal flexion/extension sagittal calf, sagittal dorsiflexion/ plantarflexion of the foot.The joint angle is calculated from the relative angle between two adjacent links.Table 2 shows the coding to determine the joint angle in the vector.

FSR-402 sensor for measuring grip strengths
Measuring is needed to calculate the grip strength of a toddler's hand automatically so that the value can be known and analysis can be carried out.The FSR sensor measures a toddler's grip strength using a push walker.
The FSR sensor is a tool that can measure force pressure by touching the surface [15].The research previously conducted by Biju et al. [21] uses gloves with an FSR sensor to determine the user's weak or strong grip strength in real time.
This study uses the FSR sensor to detect toddler grip strength when using a push walker in real-time with biomechanical analysis.The FSR sensor can provide accurate information to record grip strength and convenient ability [22].The series of FSR sensors is used to detect grip strength and must be suitable for the toddler.The FSR sensor series is designed and developed according to the anthropometry of toddler's hands that consists of a Microcontroller Unit Arduino UNO, three units of FSR-402, a 10kΩ resistor, a jumper cable, a USB cable, and a laptop.The FSR sensor has a positive pole as a source of 5 volts and a negative pole as analog input and the ground.The connection series of sensor FSR is described in Figure 4.The FSR sensor is attached to the glove according to the middle of the thumb (thumb proximal phalanx), the middle of the middle finger (middle proximal phalanx), and the palm of the little finger (fifth metacarpal) position because of the natural anatomy of the hand as the most pressure during gripping [19], [23].The gloves make it easier to install when testing the device on a toddler-the grip strength measuring is shown in Figure 5.The Kinect system is a device for detecting body position, motion detection, and skeletal tracking processes [24].Kinect has input devices like an RGB camera, infrared projector, and 3D depth sensor.An RGB camera provides information about an object's shape, texture, or surface.The depth camera functions to find out the object's depth from the camera by capturing the infrared emitted by the IR Transmitter and reflected by the object [25].This study uses the Kinect system to detect 3D motion and skeleton tracking for toddlers using a push walker.Skeleton tracking on the Kinect system can provide accurate information on the position of the toddler's body joint points [14].The Kinect system runs on Vitruvius and Microsoft Visual Studio to read the angles on the detected skeletal joints.
Vitruvius is a Kinect framework that can calculate the skeletal joint angle received by the Kinect system [26].
Vitruvius is a Kinect framework that can calculate the skeletal joint angle received by the Kinect system.Microsoft Visual Studio is software for applications and running Vitruvius software.Microsoft Visual Studio becomes the Vitruvius software tool to calculate the required angle.The joint angle calculation process will be read directly on the laptop using Vitruvius software.The program interface from Microsoft Visual Studio can be explained in Figure 8.The output results from the Vitruvius software to capture the static and the dynamic push walker can be explained in Figure 9 and Figure 10.

Static and dynamic push walker
Static push walkers move around a static, rotary axis.The static push walker in this study is made with wood on the handle and stainless steel on the frame.Toddlers use a static push walker by gripping the handle and pushing it.The handle will hold the toddler's body steady.The dimension is 42 cm, the diameter handle is 2.5 cm, the handle is 35 cm, and the mat length is 100x100 cm.This push walker is equipped with a friction level setting on the axis.This push walker has rotating speed settings, such as in Figure 11.The dynamic push walker moves the mobile by pushing it forward.Toddlers grip the push walker handles and balance to practice walking well.This requires toddlers to use their gross and fine motor skills when practicing push walkers.The dimensions are height 42 cm, width 38 cm. and diameter handle 2.5 cm.The push walker has wheels and weights that can be adjusted according to the needs of toddlers, such as in Figure 12.

Statistical analysis
Statistical data testing to find differences in results measuring body biomechanics parameters, including testing the grip strength, moment foot, and compression force of the foot when toddlers used push walker static and dynamic.The paired sample t-test was tested to find the significant differences in the values.
The Kolmogorov-Smirnov normality test initially tested the normality of all biomechanical parameters (p > 0.05).The statistical tests use a significance level α = 0.05.Differences using static and dynamic push walkers with a p-value of 0.0001 (p < 0.05) using paired sample t-tests, including the grip strength, foot segment moments, and foot compression forces.

Grip Strength
The use of push walkers for toddlers aims to stimulate their motor development so they can walk and have muscle strength.Measurement of grip strength was carried out to determine motor strength.Grip strength is the recommended assessment for measuring muscle strength for assessing muscle function in clinical practice [27].A grip strength measuring tool is developed to measure grip strength in toddlers using the FSR sensor as the grip strength data input.
The FSR sensor series is designed according to the anthropometry of toddlers' hands.Hand-held strength measuring tools include Arduino Uno, FSR, resistors, cables, and laptops.FSR becomes an input sensor for a toddler's hand grip strength when using a push walker.The number of FSR to measure grip strength is three pieces.The positioning of the FSR sensor on the glove makes it easy to attach to a toddler's hand during testing.However, using gloves reduces accuracy when placing a predetermined position.The results of grip strength are a mean of 3 maximum values according to the grip strength testing protocol by ASHT (American Society of Hand Therapists).The results of measuring the grip strength of toddlers when using a static and dynamic push walker can be explained in Table 3 and Table 4.The average grip strength when pushing a static walker is 0.794 kg; the grip strength on sensor 1 is 0.875 kg; the grip strength on sensor 2 is 0.924 kg, and the grip strength on sensor 3 is 0.584 kg.The average grip strength when pushing a dynamic walker is 0.903 kg.The grip strength in sensor 1 is 0.995 kg; the grip strength in sensor 2 is 1.044 kg; and the grip strength in sensor 3 is 0.671 kg.The comparison graphic of grip strength can be presented in Figure 13.The mean grip strength of toddlers using the static walker is lower than that of the dynamic walker.This is because toddlers using a dynamic walker tend to support the handle so that the FSR sensor detects a greater value than when using a static push walker.These results suggest that using a dynamic push walker helps toddlers gain more control over walking and improves gross and fine motor development by changing the hand grip strength when pushing or turning the walker.

Body Balance
Toddlers, when using a push walker, will be captured by the Kinect system.The Kinect system will detect the position of the toddler's body.The toddler's body position data is then processed by Vitruvius software, which can display the body angles formed using the push walker.This body angle will be used to evaluate the body using biomechanics [20].Using static and dynamic push walkers, angle readings from Vitruvius software are calculated for body balance using biomechanics.Biomechanics can explain the forces and moments that occur in every link and joint through mechanical analysis when carrying out motion activities [28,29].
Calculate moments and forces on each human body segment to get the resultant force on leg compression.The leg compression force is balanced by the muscle force on the spinal erector and abdominal force as the effect of abdominal pressure, which functions to help offset the body due to the influence of moments and forces.Large force and pressure values of the abdomen indicate good body balance.Lower leg compression forces suggest that the movement is considered good because the risk of fall and injury is minimal [30].The results of toddler biomechanics using a static and dynamic push walker are in Table 5 and Table 6.The mean value of foot moment on the static push walker is 47.114 N, and on the dynamic push walker is 45.743 N. The mean value of foot moment is bigger on the static push walker than the dynamic push walker, meaning that the body balance value on the push walker is better in static.The mean value of the compression force on the static push walker is 246.281N, and on the dynamic push walker is 264.644N. The mean value of the compression force is smaller on the static push walker than on the dynamic push walker, meaning that the body balance value on the push walker is better in static [31].In addition, the risk of falls and injury is lower.

Statistical analysis
Statistical test calculations using SPSS software version 25.The results of the paired t-test between static and dynamic push walkers are shown in Table 7.The grip strength has a probability value (sig) of 0.001, so there is a significant difference.A dynamic push walker's grip strength is greater than a static one's, so a dynamic push walker can provide better hand stimulation for toddlers.When using a dynamic push walker, this toddler tends to support the handle so that the FSR sensor detects more than a static push walker.The foot moment has a probability value (sig) of 0.001, so there is a significant difference.The foot moment of the static push walker is more significant than the dynamic, so the static push walker can provide a muscle reaction in the foot and stimulate gross motor skills better.Leg compression force has a probability value (sig) of 0.001, so there is a significant difference in value.The difference in the value of the compression force on the static and dynamic walker push shows a different body balance value.A static push walker has a lower leg compression force value, meaning the static type of push walker walks with better movement.Static push walkers make toddlers tend to be upright because the rotating movement requires body balance so they don't fall.Dynamic walker push makes toddlers tend to support the baby walker's grip so that the body tilts forward more.
The human biomechanical system is very complex and requires much understanding.The biomechanical provides many simplifications and basic assumptions for studying how different properties interact and coordinate to produce motor activity in the body.The real force in the body is not measured directly.
Biomechanical only helps to rationally interpolate and extrapolate limited data on the force and moment to provide results quickly [32].The Kinect system's biomechanical results may have an actual body balance value error.Body balance assessment can be measured using plantar pressure and optoelectronic measurements, providing more accurate motor measurements.

Discussion
The push walker stimulates gross and fine motor development to acquire basic toddler skills such as balanced walking and active hand movements.Two types of push walkers can be used for walking training in toddlers: static and dynamic.Static and dynamic exercises have an impact on different bodies.A study by Poon et al. [8] stated that the relative increases in stiffness of the tendon-aponeurosis complex and the patella tendon tended to be greater after static training than after dynamic activity, and the blood volume of the patella tendon increased significantly after dynamic.The dynamic push walker has greater grip strength than static, allowing the toddler's hand muscles to be tighter [27].It is based on measuring the grip strength in the palms of toddlers.The explanation that can be given is that toddlers tend to support the dynamic push walker when practicing.Static push walkers have a greater compression force and leg moment than dynamic ones, allowing the toddler's leg muscles to be tighter.The explanation that can be given is that toddlers tend to rest on their feet and balance their bodies so they are balanced because of the motion.
Measurement of grip strength is the recommended assessment for measuring muscle strength for assessing muscle function in clinical practice [27].Adults have had their toddler hand grip strength measured using the FSR sensor, and the results match those of more conventional grip strength testing instruments like dynamometers [13], [34].The FSR sensor series is designed according to the anthropometry of toddlers' hand grip strength measuring devices, including Arduino Uno, FSR, resistors, cables, and laptops.FSR becomes an input sensor for a toddler's hand grip strength when using a push walker.The during calibration is 97.55%, with an error value of 2.45%.FSR has several drawbacks and is susceptible to how the sensor surface is pressed.It must be about 50% in contact to capture the signal [15].FSR has a non-linear pressure response that varies with time, temperature, and humidity.Positioning the FSR sensor on the glove makes it easy to attach to a toddler's hand during testing.However, using gloves reduces accuracy when placing a predetermined position.It will affect the reading results that will be received.
Toddlers' push-walking motion helps them learn how to balance their bodies.Good body balance can have a maximum impact on motor development.Assessment of body balance is carried out using biomechanics.Biomechanics considers the forces and moments in each body connection by considering the effect of gravity's acceleration on the load segment's mass to get a better way of working or movement.Previous research from Dajime et al. [25] used the Kinect V2 sensor as input for movement assessment with kinematic features and Kinect position data by collecting sagittal position data from 31 subjects when performing movements.In this study, when pushing a baby walker, the toddler is recorded with a sagittal position using the Kinect system to detect body movement and position.The Kinect system and its supporting software produce the resulting movement angle values.Angle values are included in biomechanical calculations.The resulting accuracy during calibration is 96.44%, with an error value of 3.56%.The error value is caused by the calibration method used.The Vitruvius software failed several times in calculating and reading angles, resulting in error values.It can be considered in interpreting the results of this study against some limitations during the experiment.However, the study could not control for other plantar pressure parameters, nerves, and sensory, providing more comprehensive and clinically demonstrable results [35], [36].

Conclusion
Dynamic push walkers have greater grip strength than static, and they can provide fine motor stimulation to toddlers while pushing with stronger grip strength, allowing toddlers' hand muscles to be firmer.Static push walkers have a greater compression force and foot moment than dynamic ones, providing gross motoric stimulation to toddlers while walking in a balanced and stable manner, allowing the toddler's leg muscles to tighten up.Future research can use electromyography and optoelectronic movement to add variables like plantar pressure and nerve and muscle function.

Fig. 1 .
Fig. 1.Stage of measurement test for toddlers walking with a push walker Figure 2 (a) and (b) explain the testing protocol for static and dynamic push walkers.

Fig. 3 .
Fig. 3. Placement of the FSR sensor on the glove against the position of the palm Sensor input values in bit units are converted to grip strength (kg).The FSR sensor output data results are carried out by data cleaning.Data cleaning is identifying and correcting data by repairing or deleting it to obtain quality data.The resulting value of the grip strength used is the mean of 3 maximum values read by the grip strength measurement under the grip strength testing protocol by ASHT (American Society of Hand Therapists).Joint-based biomechanical calculations were applied to determine the sagittal segment of the treated thigh, calf, and ankle according to biomechanics theory[20].Measurement of force and moment on a toddler's body as a biomechanical model when walking using a push walker consists of the palms, forearms, upper arms, shoulders, thighs, shank, and ankles.Kinematic measurements on toddler body segments consist of palm sagittal flexion/extension, forearm sagittal flexion/extension, upper arm sagittal flexion/extension, shoulder sagittal flexion/extension, thigh sagittal flexion/ extension, sagittal thigh abduction/adduction, thigh sagittal flexion/extension sagittal calf, sagittal dorsiflexion/ plantarflexion of the foot.The joint angle is calculated from the relative angle between two adjacent links.Table2shows the coding to determine the joint angle in the vector.

Fig. 4 .
Fig. 4. Series FSR sensor for measurement of grip strength

Fig. 5 .
Fig. 5.The grip strength devices in the glove Two software programs were used for the measurement: Arduino IDE version 1.8.20 and PLX-DAQ version V.11.The Arduino IDE operates the Arduino UNO circuit system with the C++ programming language, such as adjusting the circuit's electric current, reading the FSR sensor and receiving analog data.Programming: Coding the Arduino IDE to read the data automatically to collaborate with the PLX-DAQ software-the code program for a grip strength measuring instrument is explained in Figure 6.

Fig. 6 .Fig. 7 .
Fig. 6.Coding to Arduino IDE PLX-DAQ software is add-on software that is connected to Microsoft Excel software.PLX-DAQ software will receive files from Arduino UNO in realtime on a laptop and record them in Microsoft Excel.This aims to accurately record real-time data from Arduino Uno and further data processing.The interface of the PLX-DAQ software can be explained in Figure 7.

Table 1 .
Participants in the experiment

Table 2 .
Coding in Vitruvius in Toddler Body Segments

Table 3 .
The grip strength on static push walker (kg)

Table 4 .
The grip strength on dynamic push walker (kg)

Table 5 .
Biomechanic of static push walker

Table 6 .
Biomechanic of dynamic push walker

Table 7 .
The paired t-test between static and dynamic push walkers