1 likeThe process of .stl file acquisition is as simple as it gets: in 3D VRT you just rightclick and click Save as .stl file, than you choose where to save it (like USB stick) Img 1 and 2. This model making could be done on CT studies made on Philips CTs, but it is possible to import DICOM data from other CT (Siemens) to Philips IntelliSpace Portal from our PACS. Today I spoke to tech guy in my department and he will try to connect Siemens Syngo with Philips IntelliSpace Portal for direct data transfer. I have to try import MR studies into this portal, but that is something I am not giving too much hope because this is totally different modality. For a try I took a study from a young female injured in a car accident with broken ilium bone. Slices were 1,5 mm thick, CTA of aorta and pelvic arteries were performed. CAD model of bone structures (pelvis and spine) looked really good. Surface was quite smooth and there was no cascade artifacts... nothing that couldn't be fixed with smooth modifier or in meshmixer . Img 3, 4 and 5. Since this was young persons study, I am wondering how it would look with older person study (degenerative changes in bones, osteopenia... probably there will be more artifacts, missing parts of bones). The only problem I noticed was with another CAD model of cranium (orbital floor, nasal septum and conchae, ethmoid cellulas), thin or lower density bones, respectively. As mentioned earlier, these are thinner bones so a lot of artifacts have emerged which could make some difficulties later on. Img 6 and 7. For the urgent and bigger models making I would say this is excellent technique that SAVES A LOT of TIME. If you are looking for more detailed and accurate models without big artifacts I would recommend manual/semimanual segmentation. Secondly, I wanted to make review about vascular CAD model made with this technique. Since vascular models are basically nothing but a contrast-blood mixture the final product greatly depends of contrast concentration inside vessel lumen in time of acquisition. This again greatly depends of heart contractility, blood flow etc. Contrast inhomogenicity inside vessels can produce artifacts (like relative thinning of vessel; "stenosis" would be inappropriate word, and missing parts of vessels) Img 8. If some contrast enters the veins merging of the veins and arterie can occur, Img 9. For roughly accurate 3D models, and for urgent need for models this is a very good model making technique, but if you seek a very precise models, again I would recommend manual segmentation. Soft tissue models is something that I have to try and CTA of the brain.
This leaderboard is set to Los Angeles/GMT-07:00