Thanks again everyone, this has been very useful. I will start testing/comparing methods next week, and will likely get back to some of you if I run into questions. Best! Christy
On Friday, February 24, 2023 at 3:46:42 AM UTC+1 Vera Weisbecker wrote: > Hi Christy, > > > > I have used endomaker a bit for birds and it has been a revelation, it > takes seconds to create pretty good endocasts that need a little bit of > fixing in something like geomagic. If I had to do mammals now on the normal > endomaker package, I would work out the fastest way of cropping off all > the bone that isn’t related to surrounding the brain, then wrap the outside > skull/close off the foramen magnum, and run endomaker. It would still be > way faster than manual segmentation. > > > > Cheers, > > > > Vera > > > > *From:* [email protected] <[email protected]> *On Behalf Of > *Paul Penkalski > *Sent:* Friday, 24 February 2023 7:09 AM > *To:* Christy Hipsley <[email protected]> > *Cc:* Morphmet <[email protected]> > *Subject:* Re: [MORPHMET2] programs for generating cranial endocasts > > > > Hi Christy, > Another good option is ITK-SNAP: http://www.itksnap.org/ > > Although it isn't designed specifically to segment endocraniums, it > has several automatic algorithms that I've used to segment fossil > braincases automatically and it worked really well. You may have to > clean up the segmentation a little at the end, but that's true for any > program. Then you can save the segmentation as a 3D model or mesh. > > You might download it and do a quick test segmentation, and you'll see > how straighforward it is. I had tried 3D Slicer prior to that and > couldn't get it to do anything useful. Then someone recommended > ITK-SNAP. Of course there are a few commercial apps too, but they > aren't any easier to use, and not really any more powerful than ITK, > and obviously more expensive. > > Downsampling the scan is a good idea. I often do that. You can also > crop it, e.g. if it's a scan of the whole skull and you only need to > segment the endocranium. I like IrfanView which can do batch crop and > resize and has a DICOM plugin (and it's free, at least the older > versions). 0.1 mm is pretty high resolution unless you're scanning > mice! > > Cheers, > Paul P. > > > On 2/23/23, Murat Maga <[email protected]> wrote: > > Without the data it is hard to tell, but often those crashes are memory > > related. Memory usage during segmentation increases 8-10 folds based on > the > > segmentation effect being used. So if your dataset is 2GB, you need > minimum > > of 16-20GB of "available" RAM for Slicer to work properly. If you can > share > > your dataset, I can take a look. > > > > Also for endocast segmentation there is no need to work with full > > resolution images. You can downsample your image (via the CropVolume) by > 2, > > and then try. That will reduce your data volume by 8 folds. Your > > experiments will go faster too. > > > > The other important parameters is the size of the largest hole to plug, > and > > the smoothing. Default parameters we included worked fine for large apes, > > but I am not sure for seal skulls. > > > > > > > > On Thursday, February 23, 2023 at 7:24:30 AM UTC-8 Christy Hipsley wrote: > > -- > You received this message because you are subscribed to the Google Groups > "Morphmet" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > > To view this discussion on the web visit > https://groups.google.com/d/msgid/morphmet2/[email protected] > > <https://groups.google.com/d/msgid/morphmet2/CAJveYNT5LrdYjdMLhpoc_zvgj+CQw=m4kzcbgsr3ban6hb0...@mail.gmail.com> > . > -- You received this message because you are subscribed to the Google Groups "Morphmet" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/morphmet2/1c0de0e2-6fbf-4347-bfb5-1916564f73c0n%40googlegroups.com.
