 |
 |
Hi Fabian, It seems that the question refers to the VBM template and to the group ICA. Do you know what is causing subjects to miss follow-up visits? If subjects don't return due to reasons entirely independent from the study (e.g., weather, traffic, strikes, etc), then you can safely use for the template all three observations of the subjects that have complete data, i.e., those without any missing visit. If subjects miss follow-up visits due to some other cause that may be associated with variables in the model, then the above will produce a biased template. You could consider taking just the first timepoint for all subjects. It will bias towards that timepoint, potentially penalising later timepoints, but shouldn't be too bad if analyses will compare slopes along time among different groups. All the best, Anderson On 28 June 2017 at 06:34, Fabian Wagner <[log in to unmask]> wrote: > Hello! > > I plan on extracting structural networks in a longitudinal study with 3 > timepoints. My problem is that I lose people on every time point and > therefore some have 3 timepoints, while others have 2. (29 of 68 people > have 3 timepoints) > > As the mixed models approach could deal with some "missing" values" > afterwards with statistical analysis(i realize it is more than half of > FU3), I still don't know if either vbm or ICA would be possible with > different group sizes in a repeated measures design. > > Within my baseline computations I used the normal vbm approach from the > user guide and used the resulting 4D data for a single session ICA with > concatenated data. Is something like this possible with a repeated measures > design, where some people offer more (timepoints) to the resulting gray > matter results and networks? > > In my mind, doing vbm and ICA per each timepoint, the resulting data would > be not as comparable as just cutting down the sample to have no missing > timepoints and then putting them together through the calculations. > I hope my chibberish is understandable, I don't know if I really > understand my own problem. > > Greetings and thank you for all the helping you already did (I learned a > lot from jiscmail). > > Fabian >