Questions to consider before buying a 3D printer

[curious aardvark]

couple of things:
1) you absolutely need to know what your budget is. Metal printers start at around the $50,000 mark.

2) sla and fdm are probably not going to be much use to you, other than for 'rough' prototyping. For final use you'll most likely need a nylon powder based sls system - starting at around the $10,000 mark.
A polyjet system might also be suitable as they tend to have more available materials.

3) for small parts like the nozzle thing you posted the picture of - scanning is not the way to go. Simply design from scratch. Looking at that part, it would take about an hour for me to replicate with a set of digital calipers and my cad software. It would take much much longer to try and get an accurate scan and convert it into a model. And even then it would be nowhere near as accurate as the designed from scratch model.

4) there is software that can convert mri/cat scans into printable 3d models. For things like eye sockets - that's currently your best option. Or as awerby suggests possibly use a soft silicon to make a mould.

Industrial quality scanners start at around the $20,000 mark.
For most 'things' it is quicker and more accurate to design from scratch.

But you need to get a definite top end of your budget before you can even begin to start considering equipment.

[mike@aravind.org]

.......4) there is software that can convert mri/cat scans into printable 3d models. For things like eye sockets - that's currently your best option. .......

Again, a lot of useful advice. I really need the people back at the hospital in India to consider what you just written.

Regarding the one sentence I quoted above, after doing a search, I found this:

http://www.instructables.com/id/How-...-a-3D-printab/

This may or may not solve my initial thought, 3D printing a prosthetic eye to fit a patient's eye socket, but even if we continued with a manual process as we do now, this would give us a realistic scale replica of the patient's eye socket, so the prosthetic eye can be worked on with less inconvenience to the patient. ....but it seems to me that once this scan was converted to data, we could potentially print an eye that would fit correctly into this socket.

(This has nothing to do with the questions I'm involved in regarding scanning and printing, but it does show what is possible:
https://www.fastcompany.com/3023276/...-a-3-d-printer
.....which would make me think something like the Stratasys Mojo, or something similar, might be capable of this kind of result.)

[mike@aravind.org]

(To visualize what I'm working with, this YouTube video shows the structure into which the prosthetic eye will need to fit:
https://www.youtube.com/watch?v=HKEA4p5k66U
I am not sure how much of this structure needs to be used to properly support a prosthetic eye. I will ask.)