Message Boards Message Boards

Mathematica for nuclear medicine?

One of my relatives is a cardiologist, and he uses a ridiculously expensive medical imaging workstation with proprietary software that costs tens of thousands of dollars. This thing is ancient and runs Solaris. It's called a Pegasys ADAC I believe. Here is a sample of what this type of system analyzes:

Image slices in a nuclear medicine study

These are image slices from a camera that orbits around the patient and looks at their heart. It works by detecting radioactive emissions from isotopes injected into the patient. The proprietary software (examples of software) assembles these slices into a volumetric visualization of blood flow through the patient's heart, and also computes things like blood flow and other relevant measurements. The resolution has not seemed to improve in 20 years despite advances in camera technology and other areas, so he is stuck using this old technology.

I was intrigued by this page on the Wolfram site regarding volumetric visualization and manipulation of 3D visuals based on image slices.

It appears Mathematica can do exactly the same thing that the medical imaging system does, for about 1/30th the price, and produce better results, and also not lock someone into a proprietary file format. Also, Mathematica would permit many additional features such as statistical analyses and various other features across large volumes of collected patient data. Currently, that is a manual process involving data entry into Excel spreadsheets, if it is even done at all.

Has anyone used Mathematica for nuclear medicine?

POSTED BY: Andrew Watters
2 Replies

Has anyone used Mathematica for nuclear medicine?

Well, it depends highly on what you mean by the term "use"! One can do nice things with those data and Mathematica - no question! But when it comes to primary diagnosis one just has to use tools certified for exactly this purpose. It simply is a legal matter - and exactly this makes medical software so ridiculously expensive, true!

One more remark: Nuclear medical images are typically "unsharp" for some physical reasons. In this sense they cannot be compared with e.g. CT images, because they do not show morphology but a marker distribution.

POSTED BY: Henrik Schachner
Reply to this discussion
Community posts can be styled and formatted using the Markdown syntax.
Reply Preview
Attachments
Remove
or Discard

Group Abstract Group Abstract