Feature Set: FS-W-P-5
Overview
Section titled “Overview”This feature set is targeted for sleep apnea classification based on sensor data available from wrist location. The generator computes various features over temporal windows (e.g. 30 seconds) captured from PPG sensors.
Target Location/Sensors
Section titled “Target Location/Sensors”The target location for this feature set is the wrist. From this location, the features are compute from the following raw sensors:
- PPG: Dual photoplethysmography (PPG) sensor data is used to compute SpO2 and heart rate variability (HRV) features.
Dataset Support
Section titled “Dataset Support”- MESA: This dataset does not directly provide dual PPG, however, the dataset does provide SpO2 and single channel PPG which is sufficient.
Features
Section titled “Features”This feature set includes the following 5 features:
| Feature Name | Description | Sensor |
|---|---|---|
| spo2 | Blood oxygen saturation | PPG |
| piav | Peak-to-trough amplitude delta of PPG | PPG |
| piiv | Peak-to-peak amplitude delta of PPG | PPG |
| pifv | Peak-to-peak interval delta | PPG |
| qos | PPG signal quality | PPG |
Output
Section titled “Output”The feature set is stored as HDF5 files (.h5) with one file per subject with path: {save_path}/{dataset}/{subject_id}.h5. Each HDF5 file includes the following entries:
-
/features: Time x Feature tensor (fp32). Features are computed over windows of sensor data. -
/mask: Time x Mask tensor (bool). Mask indicates valid feature values. -
/stage_labels: Time x Label (int). Labels are sleep stages. -
/apnea_labels: Time x Label (int). Labels are apnea events.