def test_rsp_rrv(): rsp90 = nk.rsp_simulate(duration=60, sampling_rate=1000, respiratory_rate=90, random_state=42) rsp110 = nk.rsp_simulate(duration=60, sampling_rate=1000, respiratory_rate=110, random_state=42) cleaned90 = nk.rsp_clean(rsp90, sampling_rate=1000) _, peaks90 = nk.rsp_peaks(cleaned90) rsp_rate90 = nk.rsp_rate(peaks90, desired_length=len(rsp90)) cleaned110 = nk.rsp_clean(rsp110, sampling_rate=1000) _, peaks110 = nk.rsp_peaks(cleaned110) rsp_rate110 = nk.rsp_rate(peaks110, desired_length=len(rsp110)) rsp90_rrv = nk.rsp_rrv(rsp_rate90, peaks90) rsp110_rrv = nk.rsp_rrv(rsp_rate110, peaks110) assert np.array(rsp90_rrv["RRV_SDBB"]) < np.array(rsp110_rrv["RRV_SDBB"]) assert np.array(rsp90_rrv["RRV_RMSSD"]) < np.array(rsp110_rrv["RRV_RMSSD"]) assert np.array(rsp90_rrv["RRV_SDSD"]) < np.array(rsp110_rrv["RRV_SDSD"]) # assert np.array(rsp90_rrv["RRV_pNN50"]) == np.array(rsp110_rrv["RRV_pNN50"]) == np.array(rsp110_rrv["RRV_pNN20"]) == np.array(rsp90_rrv["RRV_pNN20"]) == 0 # assert np.array(rsp90_rrv["RRV_TINN"]) < np.array(rsp110_rrv["RRV_TINN"]) # assert np.array(rsp90_rrv["RRV_HTI"]) > np.array(rsp110_rrv["RRV_HTI"]) assert np.array(rsp90_rrv["RRV_HF"]) < np.array(rsp110_rrv["RRV_HF"]) assert np.array(rsp90_rrv["RRV_LF"]) < np.array(rsp110_rrv["RRV_LF"])
def test_signal_rate(): # since singal_rate wraps signal_period, the latter is tested as well # Test with array. duration = 10 sampling_rate = 1000 signal = nk.signal_simulate(duration=duration, sampling_rate=sampling_rate, frequency=1) info = nk.signal_findpeaks(signal) rate = nk.signal_rate(peaks=info["Peaks"], sampling_rate=1000, desired_length=len(signal)) assert rate.shape[0] == duration * sampling_rate # Test with dictionary.produced from signal_findpeaks. assert info[list(info.keys())[0]].shape == (info["Peaks"].shape[0],) # Test with DataFrame. duration = 120 sampling_rate = 1000 rsp = nk.rsp_simulate( duration=duration, sampling_rate=sampling_rate, respiratory_rate=15, method="sinuosoidal", noise=0 ) rsp_cleaned = nk.rsp_clean(rsp, sampling_rate=sampling_rate) signals, info = nk.rsp_peaks(rsp_cleaned) rate = nk.signal_rate(signals, sampling_rate=sampling_rate, desired_length=duration * sampling_rate) assert rate.shape == (signals.shape[0],) # Test with dictionary.produced from rsp_findpeaks. rate = nk.signal_rate(info, sampling_rate=sampling_rate, desired_length=duration * sampling_rate) assert rate.shape == (duration * sampling_rate,)
def test_signal_rate(): # Test with array. signal = nk.signal_simulate(duration=10, sampling_rate=1000, frequency=1) info = nk.signal_findpeaks(signal) rate = nk.signal_rate(peaks=info["Peaks"], sampling_rate=1000, desired_length=None) assert rate.shape[0] == len(info["Peaks"]) # Test with dictionary.produced from signal_findpeaks. assert info[list(info.keys())[0]].shape == (info["Peaks"].shape[0], ) # Test with DataFrame. rsp = nk.rsp_simulate(duration=120, sampling_rate=1000, respiratory_rate=15, method="sinuosoidal", noise=0) rsp_cleaned = nk.rsp_clean(rsp, sampling_rate=1000) signals, info = nk.rsp_peaks(rsp_cleaned) rate = nk.signal_rate(signals, sampling_rate=1000) assert rate.shape == (signals.shape[0], ) # Test with dictionary.produced from rsp_findpeaks. test_length = 30 rate = nk.signal_rate(info, sampling_rate=1000, desired_length=test_length) assert rate.shape == (test_length, )
def test_rsp_rrv(): rsp90 = nk.rsp_simulate(duration=60, sampling_rate=1000, respiratory_rate=90, random_state=42) rsp110 = nk.rsp_simulate(duration=60, sampling_rate=1000, respiratory_rate=110, random_state=42) cleaned90 = nk.rsp_clean(rsp90, sampling_rate=1000) _, peaks90 = nk.rsp_peaks(cleaned90) rsp_rate90 = nk.signal_rate(peaks90, desired_length=len(rsp90)) cleaned110 = nk.rsp_clean(rsp110, sampling_rate=1000) _, peaks110 = nk.rsp_peaks(cleaned110) rsp_rate110 = nk.signal_rate(peaks110, desired_length=len(rsp110)) rsp90_rrv = nk.rsp_rrv(rsp_rate90, peaks90) rsp110_rrv = nk.rsp_rrv(rsp_rate110, peaks110) assert np.array(rsp90_rrv["RRV_SDBB"]) < np.array(rsp110_rrv["RRV_SDBB"]) assert np.array(rsp90_rrv["RRV_RMSSD"]) < np.array(rsp110_rrv["RRV_RMSSD"]) assert np.array(rsp90_rrv["RRV_SDSD"]) < np.array(rsp110_rrv["RRV_SDSD"]) # assert np.array(rsp90_rrv["RRV_pNN50"]) == np.array(rsp110_rrv["RRV_pNN50"]) == np.array(rsp110_rrv["RRV_pNN20"]) == np.array(rsp90_rrv["RRV_pNN20"]) == 0 # assert np.array(rsp90_rrv["RRV_TINN"]) < np.array(rsp110_rrv["RRV_TINN"]) # assert np.array(rsp90_rrv["RRV_HTI"]) > np.array(rsp110_rrv["RRV_HTI"]) assert np.array(rsp90_rrv["RRV_HF"]) < np.array(rsp110_rrv["RRV_HF"]) assert np.isnan(rsp90_rrv["RRV_LF"][0]) assert np.isnan(rsp110_rrv["RRV_LF"][0]) # Test warning on too short duration with pytest.warns(nk.misc.NeuroKitWarning, match=r"The duration of recording is too short.*"): short_rsp90 = nk.rsp_simulate(duration=10, sampling_rate=1000, respiratory_rate=90, random_state=42) short_cleaned90 = nk.rsp_clean(short_rsp90, sampling_rate=1000) _, short_peaks90 = nk.rsp_peaks(short_cleaned90) short_rsp_rate90 = nk.signal_rate(short_peaks90, desired_length=len(short_rsp90)) nk.rsp_rrv(short_rsp_rate90, short_peaks90)
def test_rsp_peaks(): rsp = nk.rsp_simulate(duration=120, sampling_rate=1000, respiratory_rate=15, random_state=42) rsp_cleaned = nk.rsp_clean(rsp, sampling_rate=1000) signals, info = nk.rsp_peaks(rsp_cleaned) assert signals.shape == (120000, 2) assert signals["RSP_Peaks"].sum() == 28 assert signals["RSP_Troughs"].sum() == 28 assert info["RSP_Peaks"].shape[0] == 28 assert info["RSP_Troughs"].shape[0] == 28 assert np.allclose(info["RSP_Peaks"].sum(), 1643817) assert np.allclose(info["RSP_Troughs"].sum(), 1586588) # Assert that extrema start with a trough and end with a peak. assert info["RSP_Peaks"][0] > info["RSP_Troughs"][0] assert info["RSP_Peaks"][-1] > info["RSP_Troughs"][-1]
def test_rsp_amplitude(): rsp = nk.rsp_simulate(duration=120, sampling_rate=1000, respiratory_rate=15, method="sinusoidal", noise=0) rsp_cleaned = nk.rsp_clean(rsp, sampling_rate=1000) signals, info = nk.rsp_peaks(rsp_cleaned) # Test with dictionary. amplitude = nk.rsp_amplitude(rsp, signals) assert amplitude.shape == (rsp.size, ) assert np.abs(amplitude.mean() - 1) < 0.01 # Test with DataFrame. amplitude = nk.rsp_amplitude(rsp, info) assert amplitude.shape == (rsp.size, ) assert np.abs(amplitude.mean() - 1) < 0.01
def test_rsp_rate(): rsp = nk.rsp_simulate(duration=120, sampling_rate=1000, respiratory_rate=15, method="sinusoidal", noise=0) rsp_cleaned = nk.rsp_clean(rsp, sampling_rate=1000) signals, info = nk.rsp_peaks(rsp_cleaned) # Test with dictionary. test_length = 30 rate = nk.rsp_rate(peaks=info, sampling_rate=1000, desired_length=test_length) assert rate.shape == (test_length, ) assert np.abs(rate.mean() - 15) < 0.2 # Test with DataFrame. rate = nk.rsp_rate(signals, sampling_rate=1000) assert rate.shape == (signals.shape[0], ) assert np.abs(rate.mean() - 15) < 0.2
def respiration_sequence_features(data, events, column='PZT', known_sequences=None): # nk.rsp_peaks(pzt_signal['PZT'].values, sampling_rate=sampling_rate, method="BioSPPy") sampling_rate = estimate_rate(data) # Extract peak using neurokit BioSPPy method _index = data.index _, info = nk.rsp_peaks(data[column], sampling_rate=sampling_rate, method="BioSPPy") # Truncate so that first and last events are troughs RSP_Troughs = info['RSP_Troughs'] RSP_Peaks = info['RSP_Peaks'] # at this point, check that there are some peaks/trough if RSP_Troughs.size == 0 or RSP_Peaks.size == 0: raise NoRespirationPeaks( 'No peaks/trough could be detected in PZT signal. ') RSP_Peaks = RSP_Peaks[(RSP_Peaks > RSP_Troughs[0]) & (RSP_Peaks <= RSP_Troughs[-1])] # Estimate Inspiration and Expiration durations, cycle (I+E) durations and amplitude I_duration = (RSP_Peaks - RSP_Troughs[:-1]) / sampling_rate E_duration = (RSP_Troughs[1:] - RSP_Peaks) / sampling_rate IE_duration = np.ediff1d(RSP_Troughs) / sampling_rate amplitude = data.iloc[RSP_Peaks].values - data.iloc[ RSP_Troughs[:-1]].values IE_ratio = I_duration / E_duration # Back to Dataframe format to allow extracting feature between events I_duration_df = pd.DataFrame(index=_index[RSP_Peaks], columns=['I_duration'], data=I_duration) E_duration_df = pd.DataFrame(index=_index[RSP_Troughs[1:]], columns=['E_duration'], data=E_duration) IE_duration_df = pd.DataFrame(index=_index[RSP_Troughs[1:]], columns=['IE_duration'], data=IE_duration) amplitude_df = pd.DataFrame(index=_index[RSP_Peaks], columns=['IE_amplitude'], data=amplitude) IE_ratio_df = pd.DataFrame(index=_index[RSP_Peaks], columns=['IE_ratio'], data=IE_ratio) cycles_df = pd.concat([ I_duration_df, E_duration_df, IE_duration_df, amplitude_df, IE_ratio_df ], axis=1) known_sequences = known_sequences or VALID_SEQUENCE_KEYS features = [] # for name, row in events.T.iterrows(): # transpose due to https://github.com/OpenMindInnovation/iguazu/issues/54 for index, row in events.iterrows(): logger.debug('Processing sequence %s at %s', row.id, index) if row.id not in known_sequences: continue begin = row.begin end = row.end cycles_sequence = cycles_df.loc[begin:end].copy() # extract features on sequence sequence_features = dataclass_to_dataframe( respiration_features(cycles_sequence)).rename_axis( index='id').reset_index() sequence_features.insert(0, 'reference', row.id) features.append(sequence_features) if len(features) > 0: features = pd.concat(features, axis='index', ignore_index=True, sort=False) logger.info('Generated a feature dataframe of shape %s', features.shape) else: logger.info('No features were generated') features = pd.DataFrame(columns=['id', 'reference', 'value']) return features