Line width soft constraints in AMARES

[Ronald Ouwerkerk]

I don’t understand how soft constraints work in AMARES within jMRUI 5.2.
I did a test on a tiny choline peak in a liver spectrum with massive fat signals. I have a seventeen peak model that usually works, but I kept getting very broad peaks fitted to the almost non-existent choline peak. The soft constraints on line-width did not seem to help. I then varied the soft constraint upper boundary for this peak (chem shift is limited 3.24ppm +/- 0.2 ppm) and found weird results. At low upper limits the fitted line width exceeds the boundary but it is limited and increases with increasing the upper limit (so does the amplitude) then a plateau is reached. When the soft constraint upper limit is increased further I get a different smaller result and this is equal to the result without constraint on the line width. See results table below.

Can anyone comment on the results below? What is the reference for the paper where the soft constraint feature was published and explained?

table: soft constraint effect LW in AMARES
soft max lw fitted lw fitted amp
10 18.9 15.97
15 28.3 58.43
20 37.7 64.62
25 47.2 85.54
30 49.4 110.96
35 54.2 106.44
36 54.2 106.22
40 54.2 106.23
45 54.2 106.23
50 54.2 106.26
51 54.2 106.24
60 48.8 89.1
70 48.8 89.1
100 48.8 89.1
estimated 48.8 89.1

Ronald Ouwerkerk
National Institutes of Health (NIH)

[Michal Jablonski]

Dear Ronald Ouwerkerk,

Could you send me the your data. I will try to find out if there is any problem with AMARES.

Thank you

MJ

• This reply was modified 8 years ago by Michal Jablonski.
• This reply was modified 8 years ago by Michal Jablonski.

[Author]

Posts