|
Main /
ErrorMessagesPreparations before each measurement session 1. Before entering the MSR, make sure that the subject has no metallic particles in his/her clothes or body (hair pins, eye glasses, cell phone, watch, jewels, ear rings, other piercings, bra, magnetic cards, coins, etc. It may be necessary that the subject brushes his/her hair in order to remove potential ferromagnetic particles.)
2. Check that the light on the right side of the wall behind the MEG is green before installing the subject to the chair
3. Plug the HPI coils and electrodes to the panels of MEG device etc.
4. Close the MSR door
5. Check the quality of the signals by browsing through all channels
6. If overall noise level is high, load the newest tuning file (from Squiddler -> Load settings).
7. If data seem to contain many bad channels, turn SSP vectors off (from raw data display -> Selection -> SSP) to see if the problem is on one or few channels in reality and SSP vectors just spread the noise. Handle individual bad channels as described below.
8. For individual bad channels:
* Try to heat the sensor (Squiddler -> choose the channel -> Commands -> Heat the channel + Commands -> reset all)
* If you have not been trained on how to tune sensors, ask an experienced MEG user to help you! Otherwise, tune channel (Squiddler -> Mode -> Tune) so that the characteristic curve in xy display (raw data display -> xy) is pure sinusoidal curve (no flat parts) and 1/3 of the curve is below the zero. Use bias and offset for that. Go back to the measurement mode (Squiddler -> Mode -> Meas)
* Especially take care of the magnetometers used as zero detectors in MaxShield (raw data display -> Selection -> MaxAll), their quality affects the other channels as well through feedback coils.
* If these operations do not enhance the quality of a signal, mark that channel as bad in the beginning of each measurement session (e.g. block) form On-line -> mark bad channels.
9. If data seems to be “unstable”, check that the gantry position is set correctly (Settings -> Gantry position in the megacq screen). If it is, try to turn MaxShield off for a while, check if that helps, and turn it on again.
10. If the problem looks more serious (spread of noise to other channels, unstability, etc) or persists, please reach an experienced MEG user (Etienne or Virginie)
During the measurement 1. Inspect the quality of the data!!
2. If individual bad channel appears during the recording, mark it as bad channel (On-line -> mark bad channels).
3. If you suddenly see many bad channels (strangely behaving), turn SSP vectors off to see if the problem is only in few channels, and mark them as bad channels
4. If the interference seems to be external (e.g. seen in most of the channels), it will be removed in MaxFilter processing -> do not panic if the channels are not saturated.
5. If you feel uncertain about something, ask Etienne or Virginie to come and solve the problem!
Problem with Audiophone If you have a problem to communicate with the subject inside the Magnetic Shielded Room (MSR), push the button X to turn off the device and push on the button "MSR". If the audiophone rings continuously, disconnect it (the plug is on the back) and reconnect it. Problem with the fORP responses handles If you don't have the subject's responses, check on the trigger unit box (on the shelf) on top of the desktop if the green LEDs (close to cables) are well light on. If not, check the power supply connection and the optical fiber connection If yes, move a little the optical fiber connector in the plug or try to disconnect it and reconnect it. If you have two TTL signals when the subject gives his response, move a little the optical fiber connector in the plug and test again. |