Dear Julia,
On 18 Nov 2016, at 13:58, Binder, Julia wrote:
> Dear Prof. Gaser,
>
> thank you very much for this helpful answer.
>
> Do I understand you correctly that I could use the newest version to
> start the data preprocessing from scratch with the new data set and
> also use the newer version for the older data set (that had been
> preprocessed with the older version) where I only perform the
> statistics now. The two data sets do not have to the comparable as
> they contain very different samples. Hence, the preprocessing for the
> older data set would have been done with the older version, the stats
> with the newer (to take advantage of the new features)?
If you don’t want to analyze the old and new data together in one
analysis it doesn’t matter which preprocessing you have used. It is
only important to not mix data with different preprocessings (where the
versions differ too much) because then you might have a nuisance effect
simply due to different preprocessing versions. The analysis in CAT12
has not changed, but some new features were included. Thus, you can use
the new features with the newer CAT12 version for data which were
preprocessed with older or newer versions with the prerequisite that the
analysis is done separately.
Best,
Christian
>
> Cheers,
> Julia
> ________________________________________
> Von: Christian Gaser [[log in to unmask]]
> Gesendet: Freitag, 18. November 2016 11:54
> An: [log in to unmask]; Binder, Julia
> Cc: Christian Gaser
> Betreff: Re: cat12 versions
>
> Dear Julia,
>
> there are some substantial changes between these versions. Thus, I
> would strongly recommend to not mix the different versions if version
> numbers are so different.
>
> Anyway, there is nothing wrong with the old version. We have changed
> many functions in the mean time regarding skull-stripping and
> intensity scaling (LAS) which caused problems with some data. If you
> are happy with the current results you can keep that older version.
> However, there are also many new functions in the newer versions and
> it's maybe worth to try. As alternative you can also keep the older
> version for preprocessing to be compatible, but use the newest version
> for analyzing the data.
>
> Best,
>
> Christian
>
> On Thu, 17 Nov 2016 12:51:04 +0000, Binder, Julia
> <[log in to unmask]> wrote:
>
>> Dear cat12 experts,
>>
>> I have been analyzing a data set with the cat12 version r973
>> (2016-07-25). I will now start to analyze a new data set and I wonder
>> whether I should update the cat12 version. Would this interfere with
>> the analyzes of the first data set which I will still continue? What
>> do you recommend?
>>
>> I would be very happy if somebody could comment on this.
>>
>> Cheers,
>> Julia
>>
>>
>>
>
|