Note
Click here to download the full example code
Gait Analysis Pipeline#
This example illustrates how the gait analysis pipeline by the EarGait
can be applied to ear-worn accerlation data.
The used gait event detection method is based on the work of Diao et al. [1] with a few adaptations as described in DiaoAdaptedEventDetection
Getting some example data#
For this we take some example data that contains regular walking movements.
from eargait import EarGait
from eargait.event_detection import DiaoAdaptedEventDetection
from eargait.spatial_params import SpatialParamsExample
Loading the data#
Calibrated + aligned to gravity + body frame
from eargait.utils.example_data import get_example_data
data, target_sample_rate = get_example_data()
data
{'left_sensor': acc_pa acc_ml acc_si gyr_pa gyr_ml gyr_si
n_samples
3588 0.062110 -0.560160 -9.902774 -0.524762 0.910834 -0.250719
3589 0.054049 -0.541775 -9.891725 -1.643633 0.590841 -0.469453
3590 0.020290 -0.521113 -9.830480 -1.772921 -0.687454 -0.555743
3591 0.035146 -0.504830 -9.696257 -0.863035 -1.820974 -1.032123
3592 0.094157 -0.485468 -9.934264 -0.118817 -1.412612 -0.554654
... ... ... ... ... ... ...
4385 -1.291555 0.638652 -10.180196 1.278888 3.291048 -50.932310
4386 -1.141385 0.851419 -9.837720 0.324699 4.242024 -45.653053
4387 -1.084186 1.003016 -9.705787 -0.146751 6.534404 -39.831984
4388 -1.244444 1.031254 -9.520316 -1.176098 7.666734 -34.136159
4389 -1.320403 1.099410 -9.590169 -0.213508 7.253635 -30.441111
[802 rows x 6 columns], 'right_sensor': acc_pa acc_ml acc_si gyr_pa gyr_ml gyr_si
n_samples
3588 0.109796 0.514149 -9.853317 1.460804 0.476808 0.056860
3589 0.057601 0.472815 -9.724967 0.686210 -1.145863 0.370734
3590 0.108112 0.461451 -9.820133 0.446589 -1.240347 0.582822
3591 -0.008284 0.455469 -9.790361 0.057984 -0.906538 0.635999
3592 0.117946 0.414403 -9.746449 0.069601 -0.950669 0.239099
... ... ... ... ... ... ...
4385 -1.979446 -0.776602 -9.482163 1.124168 5.551943 39.798419
4386 -2.016508 -0.807854 -9.386433 2.282151 7.007394 34.884770
4387 -1.928316 -0.828154 -9.623480 2.228703 7.231336 31.170216
4388 -2.001359 -0.935844 -9.428640 1.708899 6.963127 27.820772
4389 -2.093137 -0.987707 -9.495858 2.090396 7.229204 24.002879
[802 rows x 6 columns]}
50
Initializing event detection algorithm#
Recommended parameters: apply filter = True <br /> sampling_rate_hz needs to correspond to target_sample_rate_hz <br /> window_length should be equal to sampling_rate_hz
event_detection_algorithm = DiaoAdaptedEventDetection(
sample_rate_hz=target_sample_rate, window_length=target_sample_rate
)
Initializing spatial parameter estimation method#
Note: SpatialParamsExample is a placeholder class. Needs to be implemented by user if spatial parameters want to be estimated.
spatial_method = SpatialParamsExample(target_sample_rate)
Initializing Gait Analysis Pipeline#
Recommended parameters: sampling_rate_hz needs to correspond to target_sample_rate_hz
ear_gait = EarGait(
sample_rate_hz=target_sample_rate,
event_detection_method=event_detection_algorithm,
spatial_params_method=spatial_method,
bool_use_event_list_consistent=True,
)
Detect gait events of gait sequence#
ear_gait.detect(data)
gait_events = ear_gait.event_list
Get all gait parameters#
gait_params = ear_gait.get_gait_parameters()
gait_params
/home/docs/checkouts/readthedocs.org/user_builds/eargait/checkouts/stable/eargait/spatial_params/spatial_params_example_class.py:29: UserWarning: Example class for spatial parameter estimation is used. No spatial parameters are calculated, step length and stride length are set to NaN.
warnings.warn(
/home/docs/checkouts/readthedocs.org/user_builds/eargait/checkouts/stable/eargait/spatial_params/spatial_params_example_class.py:29: UserWarning: Example class for spatial parameter estimation is used. No spatial parameters are calculated, step length and stride length are set to NaN.
warnings.warn(
{'left_sensor': stride_time 1.176000
stance_time 0.677895
swing_time 0.429000
step_length NaN
stride_length NaN
stride_time_asymmetry 0.130101
stance_time_asymmetry 0.008667
swing_time_asymmetry 0.006000
step_length_asymmetry NaN
stride_length_asymmetry NaN
stride_time_asymmetry_percent 0.110630
stance_time_asymmetry_percent 0.012785
swing_time_asymmetry_percent 0.013986
step_length_asymmetry_percent NaN
stride_length_asymmetry_percent NaN
stride_time_si 11.124547
stance_time_si 1.278898
swing_time_si 1.398601
step_length_si NaN
stride_length_si NaN
stride_time_std 0.312450
stance_time_std 0.024850
swing_time_std 0.010208
step_length_std NaN
stride_length_std NaN
stride_time_cv 0.265689
stance_time_cv 0.036658
swing_time_cv 0.023796
step_length_cv NaN
stride_length_cv NaN
number_of_steps 25.000000
cadence 1.765537
cadence_dom_freq 1.851852
gait_velocity NaN
gait_velocity_dom_freq NaN
dtype: float64, 'right_sensor': stride_time 1.175000
stance_time 0.676842
swing_time 0.429000
step_length NaN
stride_length NaN
stride_time_asymmetry 0.128283
stance_time_asymmetry 0.010889
swing_time_asymmetry 0.010000
step_length_asymmetry NaN
stride_length_asymmetry NaN
stride_time_asymmetry_percent 0.109177
stance_time_asymmetry_percent 0.016088
swing_time_asymmetry_percent 0.023310
step_length_asymmetry_percent NaN
stride_length_asymmetry_percent NaN
stride_time_si 10.977613
stance_time_si 1.609460
swing_time_si 2.331002
step_length_si NaN
stride_length_si NaN
stride_time_std 0.312873
stance_time_std 0.026885
swing_time_std 0.012096
step_length_std NaN
stride_length_std NaN
stride_time_cv 0.266275
stance_time_cv 0.039721
swing_time_cv 0.028196
step_length_cv NaN
stride_length_cv NaN
number_of_steps 25.000000
cadence 1.768034
cadence_dom_freq 1.851852
gait_velocity NaN
gait_velocity_dom_freq NaN
dtype: float64}
Get temporal gait parameters of gait sequence#
temporal_params = ear_gait.temporal_params
temporal_params
{'left_sensor': stride_time stance_time swing_time side
s_id
0 1.16 0.72 0.44 ipsilateral
1 1.14 0.72 0.42 contralateral
2 1.12 0.70 0.42 ipsilateral
3 1.12 0.70 0.42 contralateral
4 1.10 0.68 0.42 ipsilateral
5 1.10 0.66 0.44 contralateral
6 1.10 0.66 0.44 ipsilateral
7 1.08 0.66 0.42 contralateral
8 1.10 0.66 0.44 ipsilateral
9 1.12 0.70 0.42 contralateral
10 NaN NaN NaN ipsilateral
11 2.50 NaN NaN contralateral
12 NaN NaN NaN ipsilateral
13 NaN NaN 0.44 ipsilateral
14 1.14 0.72 0.42 contralateral
15 1.10 0.68 0.42 ipsilateral
16 1.08 0.66 0.42 contralateral
17 1.08 0.66 0.42 ipsilateral
18 1.08 0.66 0.42 contralateral
19 1.08 0.64 0.44 ipsilateral
20 1.10 0.66 0.44 contralateral
21 1.10 0.66 0.44 ipsilateral
22 1.12 0.68 0.44 contralateral
23 NaN NaN NaN ipsilateral
24 NaN NaN NaN contralateral, 'right_sensor': stride_time stance_time swing_time side
s_id
0 1.16 0.74 0.42 contralateral
1 1.16 0.72 0.44 ipsilateral
2 1.14 0.70 0.44 contralateral
3 1.10 0.68 0.42 ipsilateral
4 1.10 0.66 0.44 contralateral
5 1.10 0.68 0.42 ipsilateral
6 1.08 0.66 0.42 contralateral
7 1.10 0.66 0.44 ipsilateral
8 1.10 0.66 0.44 contralateral
9 1.10 0.68 0.42 ipsilateral
10 NaN NaN NaN contralateral
11 2.50 NaN NaN ipsilateral
12 NaN NaN NaN contralateral
13 NaN NaN 0.44 contralateral
14 1.14 0.72 0.42 ipsilateral
15 1.10 0.68 0.42 contralateral
16 1.08 0.66 0.42 ipsilateral
17 1.08 0.66 0.42 contralateral
18 1.10 0.68 0.42 ipsilateral
19 1.10 0.64 0.46 contralateral
20 1.08 0.66 0.42 ipsilateral
21 1.08 0.64 0.44 contralateral
22 1.10 0.68 0.42 ipsilateral
23 NaN NaN NaN contralateral
24 NaN NaN NaN ipsilateral}
Get average temporal gait parameters#
{'left_sensor': stride_time stance_time swing_time
mean 1.17600 0.677895 0.429000
std 0.31245 0.024850 0.010208, 'right_sensor': stride_time stance_time swing_time
mean 1.175000 0.676842 0.429000
std 0.312873 0.026885 0.012096}
Get spatial parameter for walking bout#
spatial_params = ear_gait.spatial_params
spatial_params
{'left_sensor': step_length stride_length side
s_id
0 NaN NaN ipsilateral
1 NaN NaN contralateral
2 NaN NaN ipsilateral
3 NaN NaN contralateral
4 NaN NaN ipsilateral
5 NaN NaN contralateral
6 NaN NaN ipsilateral
7 NaN NaN contralateral
8 NaN NaN ipsilateral
9 NaN NaN contralateral
10 NaN NaN ipsilateral
11 NaN NaN contralateral
12 NaN NaN ipsilateral
13 NaN NaN ipsilateral
14 NaN NaN contralateral
15 NaN NaN ipsilateral
16 NaN NaN contralateral
17 NaN NaN ipsilateral
18 NaN NaN contralateral
19 NaN NaN ipsilateral
20 NaN NaN contralateral
21 NaN NaN ipsilateral
22 NaN NaN contralateral
23 NaN NaN ipsilateral
24 NaN NaN contralateral, 'right_sensor': step_length stride_length side
s_id
0 NaN NaN contralateral
1 NaN NaN ipsilateral
2 NaN NaN contralateral
3 NaN NaN ipsilateral
4 NaN NaN contralateral
5 NaN NaN ipsilateral
6 NaN NaN contralateral
7 NaN NaN ipsilateral
8 NaN NaN contralateral
9 NaN NaN ipsilateral
10 NaN NaN contralateral
11 NaN NaN ipsilateral
12 NaN NaN contralateral
13 NaN NaN contralateral
14 NaN NaN ipsilateral
15 NaN NaN contralateral
16 NaN NaN ipsilateral
17 NaN NaN contralateral
18 NaN NaN ipsilateral
19 NaN NaN contralateral
20 NaN NaN ipsilateral
21 NaN NaN contralateral
22 NaN NaN ipsilateral
23 NaN NaN contralateral
24 NaN NaN ipsilateral}
Get average spatial parameter over walking bout#
spatial_params_average = ear_gait.average_spatial_params
spatial_params_average
{'left_sensor': step_length stride_length
mean NaN NaN
std NaN NaN, 'right_sensor': step_length stride_length
mean NaN NaN
std NaN NaN}
Get cadence (num steps/duration)#
{'left_sensor': 1.7655367231638417, 'right_sensor': 1.768033946251768}
Get cadence based on the dominant frequency#
{'left_sensor': 1.8518518518518516, 'right_sensor': 1.8518518518518516}
Get asymmetry, symetry index or variability#
symmetry_index = ear_gait.get_symmetry_index()
symmetry_index
# same for ear_gait.get_variability(), ear_gait.get_asymmetry()
{'left_sensor': stride_time_si 11.124547
stance_time_si 1.278898
swing_time_si 1.398601
step_length_si NaN
stride_length_si NaN
dtype: float64, 'right_sensor': stride_time_si 10.977613
stance_time_si 1.609460
swing_time_si 2.331002
step_length_si NaN
stride_length_si NaN
dtype: float64}
Plotting gait events#
ear_gait.plot()
k = 1
Total running time of the script: ( 0 minutes 9.690 seconds)
Estimated memory usage: 59 MB