def test_emg_simulate():
emg1 = nk.emg_simulate(duration=20, length=5000, n_bursts=1)
assert len(emg1) == 5000
emg2 = nk.emg_simulate(duration=20, length=5000, n_bursts=15)
assert scipy.stats.median_absolute_deviation(
emg1) < scipy.stats.median_absolute_deviation(emg2)
def test_emg_simulate():
emg1 = nk.emg_simulate(duration=20, length=5000, n_bursts=1)
assert len(emg1) == 5000
emg2 = nk.emg_simulate(duration=20, length=5000, n_bursts=15)
assert scipy.stats.median_absolute_deviation(emg1) < scipy.stats.median_absolute_deviation(emg2)
emg3 = nk.emg_simulate(duration=20, length=5000, n_bursts=1, duration_bursts=2.0)
# pd.DataFrame({"EMG1":emg1, "EMG3": emg3}).plot()
assert len(nk.signal_findpeaks(emg3, height_min=1.0)["Peaks"]) > len(nk.signal_findpeaks(emg1, height_min=1.0)["Peaks"])
def test_emg_eventrelated():
emg = nk.emg_simulate(duration=20, sampling_rate=1000, burst_number=3)
emg_signals, info = nk.emg_process(emg, sampling_rate=1000)
epochs = nk.epochs_create(emg_signals,
events=[3000, 6000, 9000],
sampling_rate=1000,
epochs_start=-0.1,
epochs_end=1.9)
emg_eventrelated = nk.emg_eventrelated(epochs)
# Test amplitude features
no_activation = np.where(emg_eventrelated["EMG_Activation"] == 0)[0][0]
assert int(
pd.DataFrame(emg_eventrelated.values[no_activation]).isna().sum()) == 4
assert np.alltrue(
np.nansum(np.array(emg_eventrelated["EMG_Amplitude_Mean"])) <
np.nansum(np.array(emg_eventrelated["EMG_Amplitude_Max"])))
assert len(emg_eventrelated["Label"]) == 3
assert len(emg_eventrelated.columns) == 7
assert all(elem in [
"EMG_Activation", "EMG_Amplitude_Mean", "EMG_Amplitude_Max",
"EMG_Amplitude_Max_Time", "EMG_Bursts", "Label", "Event_Onset"
] for elem in np.array(emg_eventrelated.columns.values, dtype=str))
def test_emg_plot():
sampling_rate = 1000
emg = nk.emg_simulate(duration=10, sampling_rate=1000, burst_number=3)
emg_summary, _ = nk.emg_process(emg, sampling_rate=sampling_rate)
# Plot data over samples.
nk.emg_plot(emg_summary)
# This will identify the latest figure.
fig = plt.gcf()
assert len(fig.axes) == 2
titles = ["Raw and Cleaned Signal", "Muscle Activation"]
for (ax, title) in zip(fig.get_axes(), titles):
assert ax.get_title() == title
assert fig.get_axes()[1].get_xlabel() == "Samples"
np.testing.assert_array_equal(fig.axes[0].get_xticks(),
fig.axes[1].get_xticks())
plt.close(fig)
# Plot data over time.
nk.emg_plot(emg_summary, sampling_rate=sampling_rate)
# This will identify the latest figure.
fig = plt.gcf()
assert fig.get_axes()[1].get_xlabel() == "Time (seconds)"
def test_bio_process():
sampling_rate = 1000
# Create data
ecg = nk.ecg_simulate(duration=30, sampling_rate=sampling_rate)
rsp = nk.rsp_simulate(duration=30, sampling_rate=sampling_rate)
eda = nk.eda_simulate(duration=30,
sampling_rate=sampling_rate,
scr_number=3)
emg = nk.emg_simulate(duration=30,
sampling_rate=sampling_rate,
burst_number=3)
bio_df, bio_info = nk.bio_process(ecg=ecg,
rsp=rsp,
eda=eda,
emg=emg,
sampling_rate=sampling_rate)
# SCR components
scr = [val for key, val in bio_info.items() if "SCR" in key]
assert all(len(elem) == len(scr[0]) for elem in scr)
assert all(bio_info["SCR_Onsets"] < bio_info["SCR_Peaks"])
assert all(bio_info["SCR_Peaks"] < bio_info["SCR_Recovery"])
# RSP
assert all(bio_info["RSP_Peaks"] > bio_info["RSP_Troughs"])
assert len(bio_info["RSP_Peaks"]) == len(bio_info["RSP_Troughs"])
# EMG
assert all(bio_info["EMG_Offsets"] > bio_info["EMG_Onsets"])
assert len(bio_info["EMG_Offsets"] == len(bio_info["EMG_Onsets"]))
def generate_emg(time, rate):
emg = emg_simulate(duration=time,
sampling_rate=rate,
random_state=1,
burst_number=3)
return emg
def generate(time: int, rate: int):
ecg1 = nk.ecg_simulate(duration=time, sampling_rate= rate, method = "ecgsyn") # duration - ile sekund #sampling rate - ile punktów na sekunde
ecg2 = nk.ecg_simulate(duration=time, sampling_rate= rate, method = "ecgsyn")
ecg3 = nk.ecg_simulate(duration=time, sampling_rate= rate, method="ecgsyn")
ecg4 = nk.ecg_simulate(duration=time, sampling_rate= rate, method="daubechies")
emg1 = nk.emg_simulate(duration=time, sampling_rate= rate, random_state=1, burst_number=3)
emg2 = nk.emg_simulate(duration=time, sampling_rate= rate, random_state=2, burst_number=3)
emg3 = nk.emg_simulate(duration=time, sampling_rate= rate, random_state=3, burst_number=3)
emg4 = nk.emg_simulate(duration=time, sampling_rate= rate, random_state=4, burst_number=3)
# sa jeszcze analogicznie fajne wykresy dla ppg, rsp i eda
nk.signal_plot(ecg1, sampling_rate=rate)
nk.signal_plot(ecg2, sampling_rate=rate)
data = pd.DataFrame({'ecg1': ecg1,
'ecg2': ecg2,
'emg1': emg1,
})
return data
def test_emg_activation():
emg = nk.emg_simulate(duration=10, burst_number=3)
cleaned = nk.emg_clean(emg)
emg_amplitude = nk.emg_amplitude(cleaned)
activity_signal, info = nk.emg_activation(emg_amplitude)
assert set(activity_signal.columns.to_list()) == set(list(info.keys()))
assert len(info['EMG_Onsets']) == len(info['EMG_Offsets'])
for i, j in zip(info['EMG_Onsets'], info['EMG_Offsets']):
assert i < j
def test_bio_process():
sampling_rate = 1000
# Create data
ecg = nk.ecg_simulate(duration=30, sampling_rate=sampling_rate)
rsp = nk.rsp_simulate(duration=30, sampling_rate=sampling_rate)
eda = nk.eda_simulate(duration=30,
sampling_rate=sampling_rate,
scr_number=3)
emg = nk.emg_simulate(duration=30,
sampling_rate=sampling_rate,
burst_number=3)
bio_df, bio_info = nk.bio_process(ecg=ecg,
rsp=rsp,
eda=eda,
emg=emg,
sampling_rate=sampling_rate)
# SCR components
scr = [val for key, val in bio_info.items() if "SCR" in key]
assert all(len(elem) == len(scr[0]) for elem in scr)
assert all(bio_info["SCR_Onsets"] < bio_info["SCR_Peaks"])
assert all(bio_info["SCR_Peaks"] < bio_info["SCR_Recovery"])
# RSP
assert all(bio_info["RSP_Peaks"] > bio_info["RSP_Troughs"])
assert len(bio_info["RSP_Peaks"]) == len(bio_info["RSP_Troughs"])
# EMG
assert all(bio_info["EMG_Offsets"] > bio_info["EMG_Onsets"])
assert len(bio_info["EMG_Offsets"] == len(bio_info["EMG_Onsets"]))
assert all(elem in [
'ECG_Raw', 'ECG_Clean', 'ECG_Rate', 'ECG_Quality', 'ECG_R_Peaks',
"ECG_P_Peaks", "ECG_Q_Peaks", "ECG_S_Peaks", "ECG_T_Peaks",
"ECG_P_Onsets", "ECG_T_Offsets", "ECG_Atrial_Phase",
"ECG_Ventricular_Phase", "ECG_Atrial_PhaseCompletion",
"ECG_Ventricular_PhaseCompletion", 'RSP_Raw', 'RSP_Clean',
'RSP_Amplitude', 'RSP_Rate', 'RSP_Phase', 'RSP_PhaseCompletion',
'RSP_Peaks', 'RSP_Troughs', 'EDA_Raw', 'EDA_Clean', 'EDA_Tonic',
'EDA_Phasic', 'SCR_Onsets', 'SCR_Peaks', 'SCR_Height', 'SCR_Amplitude',
'SCR_RiseTime', 'SCR_Recovery', 'SCR_RecoveryTime', 'EMG_Raw',
'EMG_Clean', 'EMG_Amplitude', 'EMG_Activity', 'EMG_Onsets',
'EMG_Offsets', 'RSA_P2T'
] for elem in np.array(bio_df.columns.values, dtype=str))
def test_emg_clean():
sampling_rate=1000
emg = nk.emg_simulate(duration=20, sampling_rate=sampling_rate)
emg_cleaned = nk.emg_clean(emg, sampling_rate=sampling_rate)
assert emg.size == emg_cleaned.size
# Comparison to biosppy (https://github.com/PIA-Group/BioSPPy/blob/e65da30f6379852ecb98f8e2e0c9b4b5175416c3/biosppy/signals/emg.py)
original, _, _ = biosppy.tools.filter_signal(signal=emg,
ftype='butter',
band='highpass',
order=4,
frequency=100,
sampling_rate=sampling_rate)
emg_cleaned_biosppy = nk.signal_detrend(original, order=0)
assert np.allclose((emg_cleaned - emg_cleaned_biosppy).mean(), 0, atol=1e-6)
def test_emg_eventrelated():
emg = nk.emg_simulate(duration=20, sampling_rate=1000, burst_number=3)
emg_signals, info = nk.emg_process(emg, sampling_rate=1000)
epochs = nk.epochs_create(emg_signals,
events=[3000, 6000, 9000],
sampling_rate=1000,
epochs_start=-0.1,
epochs_end=1.9)
emg_eventrelated = nk.emg_eventrelated(epochs)
# Test amplitude features
no_activation = np.where(emg_eventrelated["EMG_Activation"] == 0)[0][0]
assert int(
pd.DataFrame(emg_eventrelated.values[no_activation]).isna().sum()) == 4
assert np.alltrue(
np.nansum(np.array(emg_eventrelated["EMG_Amplitude_Mean"])) <
np.nansum(np.array(emg_eventrelated["EMG_Amplitude_Max"])))
assert len(emg_eventrelated["Label"]) == 3
# Test warning on missing columns
with pytest.warns(nk.misc.NeuroKitWarning,
match=r".*does not have an `EMG_Onsets`.*"):
first_epoch_key = list(epochs.keys())[0]
first_epoch_copy = epochs[first_epoch_key].copy()
del first_epoch_copy["EMG_Onsets"]
nk.emg_eventrelated({**epochs, first_epoch_key: first_epoch_copy})
with pytest.warns(nk.misc.NeuroKitWarning,
match=r".*does not have an `EMG_Activity`.*"):
first_epoch_key = list(epochs.keys())[0]
first_epoch_copy = epochs[first_epoch_key].copy()
del first_epoch_copy["EMG_Activity"]
nk.emg_eventrelated({**epochs, first_epoch_key: first_epoch_copy})
with pytest.warns(nk.misc.NeuroKitWarning,
match=r".*does not have an.*`EMG_Amplitude`.*"):
first_epoch_key = list(epochs.keys())[0]
first_epoch_copy = epochs[first_epoch_key].copy()
del first_epoch_copy["EMG_Amplitude"]
nk.emg_eventrelated({**epochs, first_epoch_key: first_epoch_copy})
def test_emg_intervalrelated():
emg = nk.emg_simulate(duration=40, sampling_rate=1000, burst_number=3)
emg_signals, info = nk.emg_process(emg, sampling_rate=1000)
columns = ['EMG_Activation_N', 'EMG_Amplitude_Mean']
# Test with signal dataframe
features_df = nk.emg_intervalrelated(emg_signals)
assert all(elem in columns
for elem in np.array(features_df.columns.values, dtype=str))
assert features_df.shape[0] == 1 # Number of rows
# Test with dict
epochs = nk.epochs_create(emg_signals,
events=[0, 20000],
sampling_rate=1000,
epochs_end=20)
features_dict = nk.emg_intervalrelated(epochs)
assert all(elem in columns
for elem in np.array(features_dict.columns.values, dtype=str))
assert features_dict.shape[0] == 2 # Number of rows
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import neurokit2 as nk
# =============================================================================
# Simulate physiological signals
# =============================================================================
# Generate synthetic signals
ecg = nk.ecg_simulate(duration=10, heart_rate=70)
ppg = nk.ppg_simulate(duration=10, heart_rate=70)
rsp = nk.rsp_simulate(duration=10, respiratory_rate=15)
eda = nk.eda_simulate(duration=10, scr_number=3)
emg = nk.emg_simulate(duration=10, burst_number=2)
# Visualise biosignals
data = pd.DataFrame({
"ECG": ecg,
"PPG": ppg,
"RSP": rsp,
"EDA": eda,
"EMG": emg
})
nk.signal_plot(data, subplots=True)
# Save it
data = pd.DataFrame({
"ECG":
nk.ecg_simulate(duration=10, heart_rate=70, noise=0),
"PPG":
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import neurokit2 as nk
# =============================================================================
# Simulate physiological signals
# =============================================================================
# Generate synthetic signals
ecg = nk.ecg_simulate(duration=10, heart_rate=70)
rsp = nk.rsp_simulate(duration=10, respiratory_rate=15)
eda = nk.eda_simulate(duration=10, n_scr=3)
emg = nk.emg_simulate(duration=10, n_bursts=2)
# Visualise biosignals
data = pd.DataFrame({"ECG": ecg, "RSP": rsp, "EDA": eda, "EMG": emg})
nk.signal_plot(data, subplots=True)
# Save it
data = pd.DataFrame({
"ECG":
nk.ecg_simulate(duration=10, heart_rate=70, noise=0),
"RSP":
nk.rsp_simulate(duration=10, respiratory_rate=15, noise=0),
"EDA":
nk.eda_simulate(duration=10, n_scr=3, noise=0),
"EMG":
nk.emg_simulate(duration=10, n_bursts=2, noise=0)
})
plot = data.plot(subplots=True,
import numpy as np
import pandas as pd
import neurokit2 as nk
# =============================================================================
# Simulate physiological signals
# =============================================================================
# Generate synthetic signals
ecg = nk.ecg_simulate(duration=10, heart_rate=70)
rsp = nk.rsp_simulate(duration=10, respiratory_rate=15)
eda = nk.eda_simulate(duration=10, n_scr=3)
emg = nk.emg_simulate(duration=10, n_bursts=2)
# Visualise biosignals
data = pd.DataFrame({"ECG": ecg, "RSP": rsp, "EDA": eda, "EMG": emg})
data.plot(subplots=True, layout=(4, 1))
# Save it
plot = data.plot(subplots=True, layout=(4, 1))
plot[0][0].get_figure().savefig("README_simulation.png", dpi=300)
# =============================================================================
# Respiration (RSP) processing
# =============================================================================
# Generate one minute of respiratory signal
rsp = nk.rsp_simulate(duration=60, respiratory_rate=15)
# Process it
signals, info = nk.rsp_process(rsp)
