Hi! I'm working with the QUINT pipeline, and when I run the pipeline, I downsize my images using NUTIL, then do the ABA alignment on the downsized images, and then run the quantification using NUTIL. For this final step, I'd like to use the full-sized mask images rather than the downsized ones for better accuracy, but my json file from the registration was generated using the downsized masks. Will this be an issue in any way? Does anyone have experience doing this?
Thank you!

Hi,

Nice to hear you're using the QUINT workflow. This isn't a problem. It's the resolution of the segmentations that decides the accuracy. This means you can use low resolution images to register the sections to the atlas, and then perform segmentation of the section images at a much higher resolution. Nutil will scale the atlas registration to match the resolution of the segmentations. Here are some tips for selecting the optimal resolution for the segmentation step: https://nutil.readthedocs.io/en/latest/Q...

If you have further questions, feel free to contact us at support@ebrains.eu

Best regards,

Sharon Yates

Originally posted by jcs01:

Hi! I'm working with the QUINT pipeline, and when I run the pipeline, I downsize my images using NUTIL, then do the ABA alignment on the downsized images, and then run the quantification using NUTIL. For this final step, I'd like to use the full-sized mask images rather than the downsized ones for better accuracy, but my json file from the registration was generated using the downsized masks. Will this be an issue in any way? Does anyone have experience doing this?
Thank you!