Customize TotalSegmentator Outputs to Fit Your Needs

Did you know you don’t have to run the full TotalSegmentator model every time?

By default, the “total” task runs five different models, each responsible for segmenting a group of anatomical structures — these are called tasks (like vessels, bones, organs, etc.). The combined result gives you the full segmentation.

But sometimes, you don’t need all of that.

Run Only the Models You Need

Let’s say you only want the scapula, femur, and humerus, there’s no need to run the full model. Instead, you can choose to run only the task that includes those structures.

  • This saves time
  • Reduces unnecessary output
  • And gives you exactly what you need

Know What Each Task Segments

Not sure which model contains the structure you’re looking for? Check the full list of available classes segmented by task here:
TotalSegmentator model classes

Already Integrated in MedRouter

At MedRouter, we’ve already built this feature into the app. You can:

  • Choose only the tasks you need
  • Avoid running unnecessary models
  • Get results faster

In fact, the output in the example shown in our latest test comes from Task 294: a fully custom task that requires manual setup in TotalSegmentator. But once it’s configured, you get highly specific results like these:

🔗 Example output

💭 We’re planning to share more tips like this to help you get the most out of TotalSegmentator. Stay tuned!

Latest Updates

Ready to get started?

Unlock AI-powered imaging, enhance your workflow today!

This platform is not FDA-approved and does not provide diagnostic services. It is intended for informational purposes only and should not be used as a substitute for professional medical advice, diagnosis, or treatment