Dear Sahil,
The error may occur during the wavelet-minimum description length (MDL) process. The wavelet-MDL method uses the explanatory variable to decompose the near-infrared spectroscopy (NIRS) signal into low-frequency drift and hemodynamic signal. Here, the explanatory variable represents the predicted NIRS response induced by stimulus or task. Therefore, to remove the low-frequency drift from your resting-state NIRS data, you may want to use a high-pass filter.
A high-pass filter based on a discrete cosine transform (DCT) is currently implemented in SPM, and is available in the SPM for NIRS toolbox. To resolve the problem, in the step of 'specify 1st level', please select the DCT-based high-pass filtering.
Yours sincerely,
Sungho
On Tue, 11 Mar 2014 13:03:59 -0400, Sahil Bajaj <[log in to unmask]> wrote:
>Hi all,
>
>I have been working on NIRS resting data using NIRS_SPM. After extracting
>optode locations in MNI and after specifying 1st level analysis when I try
>to estimate, I am getting following warnings and then error messages.
>Any one working on NIRS data using NIRS_SPM, could you please help me to
>resolve this issue.
>
>.....
>.....
>.....
>Detrending... ch44
>Detrending... ch45
>Detrending... ch46
>Detrending... ch47
>Detrending... ch48
>Warning: Matrix is singular, close to singular or badly scaled.
> Results may be inaccurate. RCOND = NaN.
>> In detrend_wavelet_MDL at 164
> In spm_filter_HPF_LPF_WMDL at 102
> In precoloring at 42
> In NIRS_Estimation>push_est_Callback at 104
> In gui_mainfcn at 96
> In NIRS_Estimation at 17
>Warning: Matrix is singular, close to singular or badly scaled.
> Results may be inaccurate. RCOND = NaN.
>> In detrend_wavelet_MDL at 215
> In spm_filter_HPF_LPF_WMDL at 102
> In precoloring at 42
> In NIRS_Estimation>push_est_Callback at 104
> In gui_mainfcn at 96
> In NIRS_Estimation at 17
>Warning: Matrix is singular, close to singular or badly scaled.
> Results may be inaccurate. RCOND = NaN.
>> In detrend_wavelet_MDL at 215
> In spm_filter_HPF_LPF_WMDL at 102
> In precoloring at 42
> In NIRS_Estimation>push_est_Callback at 104
> In gui_mainfcn at 96
> In NIRS_Estimation at 17
>Detrending... ch49
>Detrending... ch50
>Detrending... ch51
>Detrending... ch52
>Warning: Matrix is singular, close to singular or badly scaled.
> Results may be inaccurate. RCOND = NaN.
>> In detrend_wavelet_MDL at 276
> In spm_filter_HPF_LPF_WMDL at 102
> In precoloring at 42
> In NIRS_Estimation>push_est_Callback at 104
> In gui_mainfcn at 96
> In NIRS_Estimation at 17
>Calculating the trace(RV) & trace(RVRV)... Please wait.
>
>Any help would be greatly appreciated.
>Thanks.
>S
>
>--
>
>-------------------------------------------------
>
>Sahil Bajaj
>Graduate Student
>Neuroscience Research Group
>Department of Physics & Astronomy
>Byrdine F. Lewis School of Nursing & Health Professions
>Georgia State University
>Atlanta, GA, USA.
>E-mail:[log in to unmask]
>
|