MRI / CTSegmentation

MRSegmentator

MRSegmentator does automatic segmentation on MRI and CT: 40 organs, large vessels, bones and muscles from the chest to the hips. Its body composition task outlines subcutaneous, visceral and gluteofemoral fat and the abdominal muscles on MRI.

Run this model →

Use MRSegmentator in your browser: it runs in the cloud. No coding, no installation.

You upload
MRI / CT scan
You get
Outlined structures
Tasks
3
Cost per million voxels
€0.006

Is it right for your scan?

Intended use

Segmenting and measuring organs and muscles across groups of patients, body composition studies, or a first draft to correct by hand where needed.

Known limitations

Does not outline the head, neck or lower legs. Least reliable on small structures such as the adrenal glands and the portal and splenic veins. Body composition was checked against human outlines on Dixon and T2-HASTE scans only. It underestimates subcutaneous fat when the body is wider than the field of view, can misjudge gluteofemoral fat in very lean people, and was not tested on severe muscle wasting.

Available tasks (3)

Anatomy (MRI)

Default

40 structures on T1-weighted, T2-weighted and Dixon scans.

Input
MRI scan.
Output
Segmentation label map.

Anatomy (CT)

The same 40 structures.

Input
CT scan.
Output
Segmentation label map.

Body composition

Abdominal subcutaneous fat runs from the ninth thoracic vertebra to the top of the gluteus maximus at the back and the pubic symphysis at the front; gluteofemoral fat from there to the knee joint. For total subcutaneous fat, add classes 1, 9 and 10.

Input
MRI scan.
Output
Segmentation label map.

After the AI task

Correct the result in your browser, then measure each structure and follow it across your study. See what is measured

Research details

Built by
Hartmut Häntze, Keno Bressem, Lisa Adams and colleagues, Charité Berlin and Technical University of Munich, Germany. Nalvera.AI runs their published model and is independent of the authors.
Version
2.0.0 · Apache-2.0 · released 6 Jul 2026
Training data
Anatomy learned from 3,749 MRI and CT scans, body composition from Dixon and T2-HASTE MRI. The authors trained five networks on different parts of these scans. Body composition runs all five. Anatomy runs two (folds 2 and 3), which is faster. On 120 public AMOS scans, two and five networks both matched human outlines of 13 structures at a Dice overlap of 0.785 on MRI and 0.813 on CT (1 is a perfect match), and no scan lost more than 0.013.

Citations & references

Questions about MRSegmentator

How do I get fat and muscle volumes from an MRI?

Run the Body composition task from a study. Nalvera.AI then lists the volume and mean signal of each class, and a chart shows how they change over time.

Other models

Nalvera iSegCT / MRI · Interactive segmentationNalvera’s in-house model for interactive CT and MRI segmentation. Define the structures you need and refine their outlines directly in your browser.MAISI-v2CT / MRI · GenerationGenerate synthetic CT or brain MRI images without an input scan. Choose the image size and anatomy for research dataset creation.MerlinCT · Report generation · AnalysisExplore abdominal CT scans through generated research reports and predicted findings. Includes tasks for estimating five-year disease risk.MOOSE CTCT · SegmentationFocused CT segmentation for heart chambers, lung lobes, muscles, and other anatomical regions.MuscleMap MRIMRI · SegmentationSegment muscles and bones on whole-body MRI scans across different MRI contrasts, with coverage from the neck through the trunk to the legs.NV-Reason-CTCT · Report generationWrite structured chest or abdominal CT reports, or a step-by-step reasoning walkthrough that states its observations and uncertainty.NV-Segment-CTCT · (Interactive) SegmentationNVIDIA’s CT segmentation model: 117 organs, bones, vessels, and muscles in one pass, plus tasks for the airway and the liver vessels.TotalSegmentator CTCT · SegmentationWhole-body and regional CT segmentation, with additional tasks for individual organs, bones, muscles, and vessels.TotalSegmentator MRIMRI · SegmentationSegment organs, bones, vessels, and muscles on MRI scans. Choose whole-body segmentation or dedicated tasks for the body, spine, or liver.

Citing Nalvera.AI

Please cite the platform alongside the model's own paper.

@misc{nalvera,
  title        = {Nalvera.AI: a cloud platform for medical-imaging AI},
  author       = {{van der Heyden, Brent}},
  year         = {2026},
  howpublished = {\url{https://nalvera.ai}},
  note         = {Model: MRSegmentator v2.0.0. Accessed YYYY-MM-DD}
}
Research use only — not a medical device. Outputs are not validated for clinical diagnosis or treatment.
Run this model → ← Back to models