A preliminary study on the employment of inertial sensors for wheelchair basketball classification: an investigation into sensor positioning

Karina Santos Guedes de Sá; José Irineu Gorla; Marília Passos Magno e Silva; Givago da Silva Souza; Anselmo de Athayde Costa e Silva

doi:10.5114/areh.2025.152011

Figures & Tables

Linear acceleration and angular velocity data of low-class athletes in the 20-meter speed test AP- Anteroposterior direction, ML- mediolateral direction, Vertical- Vertical direction

Linear acceleration and angular velocity data of high-class athletes in the 20-meter speed test AP- Anteroposterior direction, ML- mediolateral direction, Vertical- Vertical direction

Linear acceleration and angular velocity data of low-class athletes in the Illinois agility test AP- Anteroposterior direction, ML- mediolateral direction, Vertical- Vertical direction

Linear acceleration and angular velocity data of high-class athletes in the Illinois agility test AP- Anteroposterior direction, ML- mediolateral direction, Vertical- Vertical direction

Parameters extracted from the accelerometer and gyroscope signal

Parameter	Definition	Calculation	R function
Range	Difference between the maximum value and the minimum value found in the signal of linear acceleration (accelerometer) or angular velocity (gyroscope).	Range=Max(i=1\|N)−Min(i=1\|N) Range = Max\left({i=1}\|{N}\right) - Min\left({i=1}\|{N}\right)	which.min (x) - which.max (x)
Root Mean Square (RMS)	Root mean square is a measure related to the variability of the signal linear acceleration (accelerometer) or angular velocity (gyroscope). It is understood as an amplitude.	RMS=1N∑i=1n(x(i))2 RMS = \sqrt{\frac{1}{N} \sum\limits_{i=1}^{n} \left( x(i) \right)^2}	sqrt(sum(x^2)/length(x))
Mean	The mean value of linear acceleration (accelerometer) or angular velocity (gyroscope).	M=1N∑i=1nx(i) M = \frac{1}{N} \sum\limits_{i=1}^{n} x(i)	mean(x)
Standard Deviation (SD)	The standard deviation of linear acceleration signal (accelerometer) or angular velocity (gyroscope).	sd=1N−1∑i=1n(x(i)−mean(x))2 sd = \sqrt{\frac{1}{N-1} \sum\limits_{i=1}^{n} \left( x(i) - mean(x) \right)^2}	sd(x)
Max	The maximum value of linear acceleration (accelerometer) or angular velocity (gyroscope).	Max=(i=1\|N) Max=\left({i=1}\|{N}\right)	which.max(x)
Median	The median value of the linear acceleration signal (accelerometer) or angular velocity (gyroscope).	Median=li+⟨(N2−facant)fi⟩.h Median = li + \left\langle \frac{\left(\frac{N}{2} - f_{ac_{ant}}\right)}{fi} \right\rangle \cdot h	Median (x)
Kurtosis	It is a measure indicating the peak or flatness of the curves present in a central value signal of the linear acceleration signal (accelerometer) or angular velocity (gyroscope).	β=1N∑i=1n((x(i)−mean(x))sd(x))4−3 \beta = \frac{1}{N}\sum\limits_{i=1}^{n} \left( \frac{\left(x(i)-mean(x)\right)}{sd(x)}\right)^4 - 3	kurtosis(x) Need the package: PerformanceAnalytics
Skewness	A measurement of the symmetry of the linear acceleration signal (accelerometer) or angular velocity (gyroscope).	γ=1N∑i=1n((x(i)−mean(x))sd(x))3 \gamma = \frac{1}{N}\sum\limits_{i=1}^{n} \left( \frac{\left(x(i)-mean(x)\right)}{sd(x)} \right)^3	skewness(x) Need the package: PerformanceAnalytics

Correlations with sports class

Variables	R	p - value
20-meter speed test
Maximum linear acceleration in the anteroposterior direction	− 0.70	0.02 ^*
Wrist linear acceleration amplitude in the anteroposterior direction	− 0.71	0.02 ^*
Median linear acceleration wrist in the anteroposterior direction	0.81	0.00 ^*
kurtosis of the angular velocity of the wheelchair in the anteroposterior direction	0.67	0.03 ^*
Illinois agility test
Median linear acceleration of the wheelchair in the mediolateral direction	− 0.86	0.00 ^*
Kurtosis of the linear acceleration of the wrist in the anteroposterior direction	0.84	0.00 ^*
Wheelchair kurtosis of the linear acceleration in the anteroposterior direction	0.65	0.04 ^*
Asymmetry of trunk linear acceleration in the anteroposterior direction	0.68	0.03 ^*
Mean wheelchair angular velocity in the anteroposterior direction	0.70	0.02 ^*
Median wheelchair angular velocity in the resultant	− 0.65	0.04 ^*
Maximum wrist angular velocity in the mediolateral direction	0.65	0.04 ^*
Maximum angular velocity of the trunk in the resultant	− 0.77	0.00 ^*
Median angular velocity of the wheelchair in the anteroposterior direction	0.70	0.02 ^*
Asymmetry of the angular velocity of the trunk in the anteroposterior direction	− 0.73	0.01 ^*
Mean angular velocity of the wrist in the vertical direction	− 0.64	0.04 ^*

20-meter speed test and Illinois agility test results

20-meter speed test	n	Mean ± SD Time (s)	p - value
Total average	10	5.60 ± 0.96	0.38
High-class athletes (4.0 and 4.5)	3	5.25 ± 0.57
Middle-class athletes (2.0 - 3.5)	4	5.33 ± 0.95
Low-class athletes (1.0 and 1.5)	3	6.33 ± 1.15
Illinois agility test
Total average	10	30.80 ± 4.07	0.87
High-class athletes (4.0 and 4.5)	3	29.00 ± 4.08
Middle-class athletes (2.0 - 3.5)	4	30.67 ± 4.16
Low-class athletes (1.0 and 1.5)	3	33.33 ± 4.04

Description of the participants

Characteristics	Mean ± SD
Age (Years)	36.6 ± 9.1
Sex	N
Female	2
Male	8
N total	10
Health condition
Spinal cord injury	2
Congenital malformation of lower limbs	3
Polio sequelae	3
Amputation	1
Arthrogryposis	1
N total	10
Sports class
1.0	2
1.5	1
2.0	1
2.5	2
3.5	1
4.0	1
4.5	2
N total	10

A preliminary study on the employment of inertial sensors for wheelchair basketball classification: an investigation into sensor positioning

Figures & Tables

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Parameters extracted from the accelerometer and gyroscope signal

Correlations with sports class

20-meter speed test and Illinois agility test results

Description of the participants

Paradigm

My account