Heart rate estimation #85
There are no files selected for viewing
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -146,8 +146,6 @@ def __init__(self) -> None: | |
|
|
||
| self.window_length_s: int = 6 | ||
| self.window_step_size_s: int = 1 | ||
|
|
||
| # Domain feature extraction configs | ||
| class GaitFeatureExtractionConfig(GaitBaseConfig): | ||
|
|
@@ -375,9 +373,8 @@ def __init__(self, min_window_length: float = 10) -> None: | |
| super().__init__() | ||
|
|
||
| # Parameters for HR analysis | ||
| self.window_overlap_s = self.window_length_s - self.window_step_size_s | ||
| self.min_hr_samples = int(min_window_length * self.sampling_frequency) | ||
|
There was a problem hiding this comment. E.g., if sampling frequency equals 25 Hz, and window length equals 3.5, then we have There was a problem hiding this comment. Goed punt! En zal het aanpassen, heb je bezwaren tegen round i.p.v. np.round omdat je anders in de config numpy moet inladen als module @Erikpostt There was a problem hiding this comment. There are some minor differences between the two methods, and I don't think that having to import There was a problem hiding this comment. ALlright, but it doesn't seem that np.round is preferred over round() right? So than we can use just round() @Erikpostt There was a problem hiding this comment. No they seem to have different methods for rounding, but I expect that, in case the rounding is ever done differently, it should not affect the results much. |
||
| self.threshold_sqa = 0.5 | ||
|
|
||
| # Heart rate estimation parameters | ||
|
|
@@ -389,12 +386,18 @@ def __init__(self, min_window_length: float = 10) -> None: | |
| self.kern_type = 'sep' | ||
| win_type_doppler = 'hamm' | ||
| win_type_lag = 'hamm' | ||
| win_length_doppler = 8 | ||
| win_length_lag = 1 | ||
| doppler_samples = self.sampling_frequency * win_length_doppler | ||
| lag_samples = win_length_lag * self.sampling_frequency | ||
| self.kern_params = { | ||
| 'doppler': { | ||
| 'win_length': doppler_samples, | ||
| 'win_type': win_type_doppler, | ||
| }, | ||
| 'lag': { | ||
| 'win_length': lag_samples, | ||
| 'win_type': win_type_lag, | ||
| } | ||
| } | ||
|
|
||
| Original file line number | Diff line number | Diff line change | ||||
|---|---|---|---|---|---|---|
|
|
@@ -3,16 +3,18 @@ | |||||
|
|
||||||
| import pandas as pd | ||||||
| import os | ||||||
| import numpy as np | ||||||
|
|
||||||
| import tsdf | ||||||
| import tsdf.constants | ||||||
| from paradigma.heart_rate.heart_rate_analysis_config import SignalQualityFeatureExtractionConfig, SignalQualityClassificationConfig, HeartRateExtractionConfig | ||||||
| from paradigma.util import read_metadata | ||||||
| from paradigma.config import SignalQualityFeatureExtractionConfig, SignalQualityClassificationConfig, HeartRateExtractionConfig | ||||||
| from paradigma.segmenting import tabulate_windows_legacy | ||||||
| from paradigma.heart_rate.feature_extraction import extract_temporal_domain_features, extract_spectral_domain_features | ||||||
| from paradigma.heart_rate.heart_rate_estimation import assign_sqa_label, extract_hr_segments, extract_hr_from_segment | ||||||
| from paradigma.constants import DataColumns | ||||||
|
|
||||||
| def extract_signal_quality_features(df: pd.DataFrame, config: SignalQualityFeatureExtractionConfig) -> pd.DataFrame: | ||||||
| # Group sequences of timestamps into windows | ||||||
| df_windowed = tabulate_windows_legacy(config, df) | ||||||
|
|
@@ -31,7 +33,7 @@ def extract_signal_quality_features_io(input_path: Union[str, Path], output_path | |||||
| metadata_time, metadata_values = read_metadata(input_path, config.meta_filename, config.time_filename, config.values_filename) | ||||||
| df = tsdf.load_dataframe_from_binaries([metadata_time, metadata_values], tsdf.constants.ConcatenationType.columns) | ||||||
|
|
||||||
| # Extract signal quality features | ||||||
| df_windowed = extract_signal_quality_features(df, config) | ||||||
| return df_windowed | ||||||
|
|
||||||
|
|
@@ -74,7 +76,7 @@ def signal_quality_classification(df: pd.DataFrame, config: SignalQualityClassif | |||||
| # Make predictions for PPG signal quality assessment | ||||||
| df[DataColumns.PRED_SQA_PROBA] = lr_clf.predict_proba(X_normalized)[:, 0] | ||||||
| df.drop(columns = lr_clf.feature_names_in_, inplace=True) # Drop the features used for classification since they are no longer needed | ||||||
|
|
||||||
| return df | ||||||
|
|
||||||
|
|
||||||
|
|
@@ -85,4 +87,60 @@ def signal_quality_classification_io(input_path: Union[str, Path], output_path: | |||||
| metadata_time, metadata_values = read_metadata(input_path, config.meta_filename, config.time_filename, config.values_filename) | ||||||
| df = tsdf.load_dataframe_from_binaries([metadata_time, metadata_values], tsdf.constants.ConcatenationType.columns) | ||||||
|
|
||||||
| df = signal_quality_classification(df, config, path_to_classifier_input) | ||||||
|
|
||||||
|
|
||||||
| def estimate_heart_rate(df: pd.DataFrame, df_ppg_preprocessed: pd.DataFrame, config:HeartRateExtractionConfig) -> pd.DataFrame: | ||||||
|
There was a problem hiding this comment. To me it's a little ambiguous as to the meaning of There was a problem hiding this comment.
Suggested change
|
||||||
| """ | ||||||
| Estimate the heart rate from the PPG signal using the time-frequency domain method. | ||||||
|
|
||||||
| Parameters | ||||||
| ---------- | ||||||
| df : pd.DataFrame | ||||||
| The DataFrame containing the PPG signal. | ||||||
| df_ppg_preprocessed : pd.DataFrame | ||||||
| The DataFrame containing the preprocessed PPG signal. | ||||||
| config : HeartRateExtractionConfig | ||||||
| The configuration for the heart rate estimation. | ||||||
|
|
||||||
| Returns | ||||||
| ------- | ||||||
| pd.DataFrame | ||||||
| The DataFrame containing the heart rate estimations. | ||||||
| """ | ||||||
|
|
||||||
| # Assign window-level probabilities to individual samples | ||||||
| ppg_post_prob = df.loc[:, DataColumns.PRED_SQA_PROBA].to_numpy() | ||||||
| #acc_label = df.loc[:, DataColumns.ACCELEROMETER_LABEL].to_numpy() # Adjust later in data columns to get the correct label, should be first intergrated in feature extraction and classification | ||||||
|
There was a problem hiding this comment. bedoel je dat de acc_label uitgecomment is? Want dat is intentional nu ja @Erikpostt There was a problem hiding this comment. Yes, I was just wondering. |
||||||
|
|
||||||
| sqa_label = assign_sqa_label(ppg_post_prob, config) | ||||||
| v_start_idx, v_end_idx = extract_hr_segments(sqa_label, config.min_hr_samples) | ||||||
|
Comment on lines
+116
to
+117
There was a problem hiding this comment. Perhaps add brief comments (#) here to explain what happens, e.g., |
||||||
|
|
||||||
| fs = config.sampling_frequency | ||||||
|
|
||||||
| v_hr_rel = np.array([]) | ||||||
| t_hr_rel = np.array([]) | ||||||
|
|
||||||
| for start_idx, end_idx in zip(v_start_idx, v_end_idx): | ||||||
| # Skip if the epoch cannot be extended by 2s on both sides | ||||||
| if start_idx < 2 * fs or end_idx > len(df_ppg_preprocessed) - 2 * fs: | ||||||
KarsVeldkamp marked this conversation as resolved.
Show resolved
Hide resolved
|
||||||
| continue | ||||||
|
|
||||||
| # Extract the extended PPG segment for HR estimation | ||||||
| extended_ppg_segment = df_ppg_preprocessed[DataColumns.PPG][start_idx - 2 * fs : end_idx + 2 * fs] | ||||||
|
|
||||||
| # Perform HR estimation | ||||||
| hr_est = extract_hr_from_segment(extended_ppg_segment, config.tfd_length, fs, config.kern_type, config.kern_params) | ||||||
|
|
||||||
| # Generate HR estimation time array | ||||||
| rel_segment_time = df_ppg_preprocessed.time[start_idx:end_idx].values | ||||||
|
There was a problem hiding this comment. Is There was a problem hiding this comment. nee, goed punt |
||||||
| n_full_segments = len(rel_segment_time) // config.hr_est_samples | ||||||
| hr_time = rel_segment_time[:n_full_segments * config.hr_est_samples : config.hr_est_samples] # relative time in seconds after the start of the segment | ||||||
|
There was a problem hiding this comment. Perhaps you could clarify this process a little using comments (#), I'm not sure I understand it as it is now. |
||||||
|
|
||||||
| # Concatenate HR estimations and times | ||||||
| v_hr_rel = np.concatenate((v_hr_rel, hr_est)) | ||||||
| t_hr_rel = np.concatenate((t_hr_rel, hr_time)) | ||||||
|
|
||||||
| df_hr = pd.DataFrame({"rel_time": t_hr_rel, "heart_rate": v_hr_rel}) | ||||||
|
|
||||||
| return df_hr | ||||||
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Je hebt hier en daar nog execution counts, wat op zich niet een probleem is omdat je misschien juist deze output wilt laten zien aan een gebruiker zonder dat een gebruiker de code hoeft te draaien. Even een check of dat je intentie was.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ehm uiteindelijk wel, maar niet voor nu eig dus excuses