Questions for diffusion-related bi-tensor modelling using QIT

[cyyyuyan]

Dear Dr. Ryan Cabeen,

I am a beginner of Quantitative Imaging Toolkit (QIT) and diffusion data-related analysis. Recently, while conducting diffusion data analysis, I learned from literature that the analysis related to diffusion should take into account the influence of perivascular spaces (PVS). I am interested in analyzing the impact of PVS on tissue signals in our own collected multi-shell diffusion data.

I have learned a paper where the authors utilized the bi-tensor model in the QIT to differentiate PVS signals from tissue signals. That is, they established two tensors for each voxel, one corresponding to tissue and the other to PVS compartment. I am wondering if the "VolumeBitensorFit" model in QIT is the one used for this type of analysis.

I have some difficulties understanding the output results from this model. For instance, I am not clear on the difference between the outputs "bitensor_fFA.nii.gz" and "bitensor_tFA.nii.gz". I am uncertain how these outputs correspond to PVS and tissue. However, I could not find an explanation for the output results by using the command "qit VolumeBitensorFit --help". Could you please tell me what these outputs represents? The example results are as follows (the method I choose is Anisotropic):

Additionally, I want to know if I need to further analyze the results obtained from "VolumeBitensorFit" using the script which named "pvs-qitdiff", in order to differentiate the signals of the PVS and tissue? I am looking forward to your guidance and appreciate your help in advance.

Best regards

[cyyyuyan]

Hi QIT Experts,
I just wanted to follow up with my questions below to see if anyone know the answers to any of them. Thanks in advance for the help.

Best
yuyan

Dear Dr. Ryan Cabeen,

I am a beginner of Quantitative Imaging Toolkit (QIT) and diffusion data-related analysis. Recently, while conducting diffusion data analysis, I learned from literature that the analysis related to diffusion should take into account the influence of perivascular spaces (PVS). I am interested in analyzing the impact of PVS on tissue signals in our own collected multi-shell diffusion data.

I have learned a paper where the authors utilized the bi-tensor model in the QIT to differentiate PVS signals from tissue signals. That is, they established two tensors for each voxel, one corresponding to tissue and the other to PVS compartment. I am wondering if the "VolumeBitensorFit" model in QIT is the one used for this type of analysis.

I have some difficulties understanding the output results from this model. For instance, I am not clear on the difference between the outputs "bitensor_fFA.nii.gz" and "bitensor_tFA.nii.gz". I am uncertain how these outputs correspond to PVS and tissue. However, I could not find an explanation for the output results by using the command "qit VolumeBitensorFit --help". Could you please tell me what these outputs represents? The example results are as follows (the method I choose is Anisotropic):

Additionally, I want to know if I need to further analyze the results obtained from "VolumeBitensorFit" using the script which named "pvs-qitdiff", in order to differentiate the signals of the PVS and tissue? I am looking forward to your guidance and appreciate your help in advance.

Best regards