High values in Kurtosis measures

[JRNeurons]

Hello,
I am using Pydesigner to process dMRI, and I am getting some wildly high or low values in mean kurtosis, radial kurtosis, WMTI_eas_RD, and WMTI_eas_tort. (normal values for mk and rk range from 0.5 to 2, and abnormal values can be of an order between +/- ~1e4 and +/-1e9. I have included a screenshot of the nifti file of the RK. You can see that there are a few very bright spots and that the scale is extremely high.

I only see this in a few of our scans. Has anyone else had any experience troubleshooting these issues? We use the same parameters to process all of our scans.
Thanks,
JR