Dear Sharon,
for fitting the DCE MRI data, a Levenberg-Marquardt fitting routine is used. The starting parameters are the initial parameters for this optimizer. So if you set i.e. Ktrans to 15, the search for the most likely Ktrans in the parameter space will start at this value. Setting such initial parameters can be useful to prevent optimization resulting in local minima. This does not mean, however, that the resulting Ktrans is constrained to values larger than 15. Setting constraints for the resulting parameters is also possible in the box "Constraints". Here, the result of the optimizer is not allowed above/below the stated upper/lower values. Remark on the units: In MITK, Ktrans is calculated in the unit of ml/min/100ml. You might have to do a conversion of the values, if you are using i.e. instead ml/min/ml. The formula of the relative signal enhancement is given by: c(t) = k * (S(t)-S_0)/S_0, where c(t) is the contrast agent concentration, k the conversion factor, S(t) the dynamic DCE MRI signal, and S_0 the baseline signal. I hope this answered your questions regarding start parameter and relative signal enhancement? Please just let me know if something is not clear. We will look into the DICOM reading error and get back to you on that. Best wishes, Ina Kompan
_______________________________________________ mitk-users mailing list mitk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mitk-users
