I'm Kyle, a fourth year PhD student within the Arkell group at hte Australian National University. My PhD is focused on using LAMA in which I've developed two-way statistical analysis for the pipeline and I'm in the midst of incorporating shape and texture analysis into the pipeline via ml analysis of radiomic features.
It goes without saying that your preprint is really interesting and MEMOs is something the field has needed desperately. However, I can't directly utilise MEMOS as we use uCT scans of E14.5 embryos stained with PTA, not the E15.5 embryos used in TCP pipeline. Could I please have some insights on how difficult it would be to train our model? Computational resources shouldn't be a problem for our lab as we are supported by the National Computational Infrastructure - it would be mainly issues with sample size / quality of registrations for training.
Obviously I would be interested in collaborating. I would love to combine MEMOS' auto-segmentation with LAMA's statistical analysis.
Happy to receive emails (kyle.drover@anu.edu.au) or set up a Zoom meeting to discuss futher.
Hi Sara,
I'm Kyle, a fourth year PhD student within the Arkell group at hte Australian National University. My PhD is focused on using LAMA in which I've developed two-way statistical analysis for the pipeline and I'm in the midst of incorporating shape and texture analysis into the pipeline via ml analysis of radiomic features.
It goes without saying that your preprint is really interesting and MEMOs is something the field has needed desperately. However, I can't directly utilise MEMOS as we use uCT scans of E14.5 embryos stained with PTA, not the E15.5 embryos used in TCP pipeline. Could I please have some insights on how difficult it would be to train our model? Computational resources shouldn't be a problem for our lab as we are supported by the National Computational Infrastructure - it would be mainly issues with sample size / quality of registrations for training.
Obviously I would be interested in collaborating. I would love to combine MEMOS' auto-segmentation with LAMA's statistical analysis.
Happy to receive emails (kyle.drover@anu.edu.au) or set up a Zoom meeting to discuss futher.
Kind Regards, Kyle