What is the warning means?

zhangjb39 · August 29, 2022, 5:07pm

When I import a Nicolet data to BS, I get a warning of "using first instance ac for all segments, see the doc for info". So which document should I refer?

Raymundo.Cassani · August 29, 2022, 6:20pm

Hi @zhangjb39,

This warning:

Warning: Multiple TSinfo packets detected; using first instance ac for all segments. See documentation for info.

comes from the EEG Nicolet reader script, which was not written by members of the Brainstorm team.

This makes reference to the WARNING comment at the beginning of Nicolet reader script:

github.com

brainstorm-tools/brainstorm3/blob/master/external/other/NicoletFile.m

classdef NicoletFile < handle
  % NICOLETFILE  Reading Nicolet .e files.
  %
  %   GETTING DATA
  %   You can load data from the .e file using the GETDATA method. The
  %   inputs to the method are the object, the segment of the data file
  %   that you want to load data from, the min, and max index you want to
  %   retrieve, and a vector of channels that you want to retrieve.
  %
  %   Example:
  %     OUT = GETDATA(OBJ, 1, [1 1000], 1:10) will return the first 1000
  %     values on the first 10 channels of the first segment of the file.
  %
  %   GETTING Nr OF SAMPLES
  %     Use the GETNRSAMPLES method to find the number of samples per
  %     channel in each data segment.
  %
  %   WARNING! 
  %   The .e format allows for changes in the TimeSeries map during the
  %   recording. This results in multiple TSINFO structures. Depending on

This file has been truncated. show original

Best,
Raymundo

@Francois, it seems that there is a newer version of the Nicolet reader that supports multiple and differing TsInfos (Adds support for multiple and differing TsInfos. · ieeg-portal/Nicolet-Reader@2874014 · GitHub). Have you check it out?

zhangjb39 · August 30, 2022, 12:37am

Thank you

Francois · August 31, 2022, 6:58am

No, I haven't looked at this, but I'm not sure where this update I pushed in October 2018 came from, while this multiple TSInfos fix was posted in April 2018. I'll investigate.

@zhangjb39 Can you please share an example file (preferably short, for data transfer considerations)?

Zip the .e file and upload it somewhere of your choice (that can be accessed easily outside of China)
Post the download link here
If this is clinical data, send it to me by private message instead of public post.

Francois · September 5, 2022, 11:32am

Thank you for the example file.

I updated the NicoletFile reader with the latest version supporting multiple TsInfos:
brainstorm3/NicoletFile.m at master · brainstorm-tools/brainstorm3 · GitHub

However, this doesn't fix the problem with the file that you shared.
At line 454: brainstorm3/NicoletFile.m at master · brainstorm-tools/brainstorm3 · GitHub
nrIdx(2) = 1.1744e+09
Which means that there would be billions of channels... the for loop at line 455 never returns because of this.

I can't say whether this is a problem with your file or with the reader.
Do you have issues with all your .e files in Brainstorm, or is it only this one?
Can you open the file with the Nicolet viewer?

One solution would be to use the Nicolet software to convert the .e file to another file format, maybe EDF+, and process the .edf file in Brainstorm instead.

zhangjb39 · September 5, 2022, 12:29pm

Not just one. Some of them are in such condition. I found there are some pause among these data. I can open it with Nicolet viewer (smoothly). The problem is that I have no access to the Nicolet software which able to convert the .e file to another file format.....

Francois · September 5, 2022, 1:19pm

I'm afraid I have no other solution to offer.

The function NicoletFile does not seem to be actively maintained. A question I asked in 2020 has never been addressed:
https://github.com/ieeg-portal/Nicolet-Reader/issues

zhangjb39 · September 5, 2022, 2:15pm

Thank you. No matter what. This is an ending for this question in some degree.