Electrocardiogram (ecg)
Following the guidelines from BIDS v1.10.0, we propose the following additional requirements for ECG data.
Modality specific data SHOULD be split into modality specific files carrying the modality-label. Modality label for electrocardiogram data is ecg and ECG specific data files would be organized as follows
physio/
├─ sub-01_ses-1_task-TaskName_recording-ecg_physio.json
├─ sub-01_ses-1_task-TaskName_recording-ecg_physio.tsv.gz
Column names in the ecg data files MUST be ecg for experiments involving single recordings, or ecg1, ecg2, ecg3, … for experiments involving simultaneous parallel ECG recordings from different locations on the body.
These fields MUST be included in the JSON sidecar file:
<matches>_recording-ecg_physio.json
Additional Metadata Fields
Key |
Description |
Data Type |
Possible Values |
|---|---|---|---|
|
Configuration of the recorded ECG leads |
Text |
|
Example TSV Data File
ecg
-0.140686035156250
-0.131378173828125
-0.121612548828125
-0.111236572265625
-0.0994873046875000
-0.0865173339843750
-0.0727844238281250
. . .
In case of compressed tabular files (like `.tsv.gz`) column names MUST NOT be included.
Example JSON Sidecar
{
"Columns": [
"ecg",
],
"Manufacturer": "<Manufacturer-Name>",
"ManufacturersModelName": "<Manufacturer-Model-Name>",
"DeviceSerialNumber": "<Device-Serial-Number>",
"SamplingFrequency": 2000,
"SoftwareVersion": "<Software-version>",
"StartTime": 0,
"PhysioType": "specified",
"ecg": {
"Description": "ECG Recording",
"LeadConfiguration": "II",
"Units": "mV"
}
}