Getting started¶

Install¶

From source¶

git clone git clone git@bitbucket.org:gcpds/pycollect.git
cd pycollect
python setup.py install

From PyPi (not available yet)¶

pip install pycollect

Collect data¶

There are two main modules, GEDevice that handle the connection and data recollecting, and GEDecode that parse and sort the received data strings.

from pycollect import GEDevice

Connection¶

device = GEDevice()
device.connect('/dev/ttyUSB0')

Send request¶

A request is an instruction that enable the data transmission from monitor.

Subrecords¶

There are three subrecord types for the actual measurement data:

device.DISPL for the displayed values.
device.TREND_10S for the 10 s trended values.
device.TREND_60S for the 60 s trended values.

When device.DISPL is used, there is an extra argument, interval, that indicate the transmission interval in seconds, this must be possitive and greater or equal to 5.

device.request(subtype=device.DISPL)

device.TREND_10S and device.TREND_60S are used for request trends.

device.request(subtype=device.TREND_10S)

device.request(subtype=device.TREND_60S)

No further transmission requests are needed after this.

Waveforms¶

Depending on the total number of samples per second the monitor sends a waveform packet every 1000 ms, 500 ms or 250 ms. Request up to 8 waveforms but total sample rate should be less than 600 samples/sec, sample rate for ECG is 300, INVP 100, EEG 100, PLETH 100, respiratory (CO2, O2, NO2 …) 25 each.

The waveform options can be consulted with:

from pycollect.modules import WAVEFORMS_DICT

for wave in WAVEFORMS_DICT:
    print("{label}: {samps}".format(**WAVEFORMS_DICT[wave]))

ECG1: 300
ECG2: 300
ECG3: 300
INVP1: 100
INVP2: 100
INVP3: 100
INVP4: 1
INVP5: 100
INVP6: 100
PLETH: 100
CO2: 25
NO2: 25
AA: 25
AWP: 25
FLOW: 1
VOL: 25
RESP: 25
EEG1: 100
EEG2: 100
EEG3: 1
EEG4: 100
TONO_PRESS: 1
SPI_LOOP_STATUS: 1
ENT_100: 1
EEG_BIS: 1

device.request(waveform_set=['PLETH', 'ECG1'])  #400 samp/s

The limit of sample rate can not be exceeded:

try:
    device.request(waveform_set=['ECG1', 'ECG2', 'ECG3'])  #900 samp/s
except Exception as error:
    print(error)

Sample rate must be less or equal to 600.

No further transmission requests are needed after this.

Mixings¶

A combination of Waveforms and Subrecords can be requested at the same time.

device.request(subtype=device.DISPL, waveform_set=['PLETH', 'ECG1'])

Or in a sequence of requests.

device.request(subtype=device.DISPL, interval=10)
device.request(waveform_set=['PLETH', 'ECG1'])

device.request(subtype=device.DISPL, interval=10)
device.request(subtype=device.TREND_60S)
device.request(waveform_set=['PLETH', 'ECG1'])

Read data¶

The transmitted data is recollected asynchronously with a background thread and appended to a BUFFER list.

To activate the data collecting.

device.collect(True)

To clear the buffer input:

device.clear_buffer()

To stop the data collecting:

device.collect(False)

Decode data¶

The GEDecode module is used for parse and sort the received data strings.

from pycollect import GEDecode

decoder = GEDecode(device.BUFFER)
decoder.process(True)

Clear buffers¶

There are a set of methods for clear correctly the stored buffer and recollected data.

device.clear_buffer()  # clear input buffer.
decoder.clear_buffer()  # clear decoded data, breaks the synchrony.
decoder.clear_data()  # clear recollected data.

The above instructions breaks the synchrony between the collector and decoder, in order to decode input data again GEDecode must be reinstantiated:

decoder = GEDecode(device.BUFFER)
decoder.process(True)

Save data¶

The data can be saved in two differents formats csv and edf, each type of data will be saved with their own sufix: .wave for waveforms, .trend10s and .trend60s for trends.

Save data as CSV¶

decoder.save_as_csv('data_out')

Save data as EDF+¶

The edf format need extra patient information.

decoder.set_edf_header(
    admincode = '',
    birthdate = date(1900, 1, 1), #datetime object
    equipment = '',
    gender = 0, #0 for male, 1 for female
    patientcode = '',
    patientname = '',
    patient_additional = '',
    recording_additional = '',
    technician = '',
)

decoder.save_as_edf('data_out')