CT / MRIGeneration

MAISI-v2

MAISI-v2, NVIDIA's model for synthetic CT and brain MRI (also published as NV-Generate-CTMR), generates images without an input scan. Create CT images with optional anatomy masks, or brain MRI images in one or several contrasts.

Run this model →

Use MAISI-v2 in your browser: it runs in the cloud. No coding, no installation.

You upload
No scan needed
You get
Synthetic image
Tasks
2
Cost per million voxels
€0.008

Is it right for your research?

Intended use

Creating synthetic (artificial) 3D CT or brain MRI images for research, for example to build or enlarge training datasets, optionally with matching CT organ masks. Synthetic images must never be used as real patient data.

Known limitations

Image quality depends strongly on the image size and spacing you choose: settings unlike its training data can give unrealistic images, even when they are accepted. The CT mask comes from a real anatomy template for the body region you choose; organ sizes cannot be set. Brain MRI in several contrasts creates each contrast separately from the same starting point; it does not convert a real scan.

Available tasks (2)

CT generation — image / image + mask

Default

Synthetic CT generation (rflow-ct). Image-only, or image + an anatomy mask taken from a real scan of the chosen body region, optionally with a tumour type.

Input
Choose the generation settings.
Output
Synthetic CT image, optionally with a paired anatomy mask.
View 124 output classes
1 liver3 spleen4 pancreas5 right kidney6 aorta7 inferior vena cava8 right adrenal gland9 left adrenal gland10 gallbladder11 esophagus12 stomach13 duodenum14 left kidney15 bladder17 portal vein and splenic vein19 small bowel22 brain23 lung tumor24 pancreatic tumor25 hepatic vessel26 hepatic tumor27 colon cancer primaries28 left lung upper lobe29 left lung lower lobe30 right lung upper lobe31 right lung middle lobe32 right lung lower lobe33 vertebrae L534 vertebrae L435 vertebrae L336 vertebrae L237 vertebrae L138 vertebrae T1239 vertebrae T1140 vertebrae T1041 vertebrae T942 vertebrae T843 vertebrae T744 vertebrae T645 vertebrae T546 vertebrae T447 vertebrae T348 vertebrae T249 vertebrae T150 vertebrae C751 vertebrae C652 vertebrae C553 vertebrae C454 vertebrae C355 vertebrae C256 vertebrae C157 trachea58 left iliac artery59 right iliac artery60 left iliac vena61 right iliac vena62 colon63 left rib 164 left rib 265 left rib 366 left rib 467 left rib 568 left rib 669 left rib 770 left rib 871 left rib 972 left rib 1073 left rib 1174 left rib 1275 right rib 176 right rib 277 right rib 378 right rib 479 right rib 580 right rib 681 right rib 782 right rib 883 right rib 984 right rib 1085 right rib 1186 right rib 1287 left humerus88 right humerus89 left scapula90 right scapula91 left clavicula92 right clavicula93 left femur94 right femur95 left hip96 right hip97 sacrum98 left gluteus maximus99 right gluteus maximus100 left gluteus medius101 right gluteus medius102 left gluteus minimus103 right gluteus minimus104 left autochthon105 right autochthon106 left iliopsoas107 right iliopsoas108 left atrial appendage109 brachiocephalic trunk110 left brachiocephalic vein111 right brachiocephalic vein112 left common carotid artery113 right common carotid artery114 costal cartilages115 heart116 left kidney cyst117 right kidney cyst118 prostate119 pulmonary vein120 skull121 spinal cord122 sternum123 left subclavian artery124 right subclavian artery125 superior vena cava126 thyroid gland127 vertebrae S1128 bone lesion132 airway

MRI brain generation — image / cross-contrast

Synthetic brain MRI generation (rflow-mr-brain). Image-only in a chosen sequence (T1/T2/FLAIR/SWI, whole-brain or skull-stripped), or cross-contrast: several sequences in one task, each a separate synthetic brain, so the images do not line up with each other.

Input
Choose the generation settings.
Output
One synthetic brain MRI image per selected sequence.

Step-by-step guide: how to generate synthetic CT and brain MRI images

Research details

Built by
NVIDIA (Can Zhao and colleagues). Nalvera.AI runs their published model and is independent of the authors.
Version
1.1 · Open Model License · released 23 Aug 2026
Training data
MAISI-v2. CT weights trained on large multi-institutional CT cohorts. Brain MRI weights are NVIDIA's NV-Generate-MR-Brain v1 (August 2026), trained on MR-RATE batches 0–27: 73,516 unique subjects / 665,371 scans across T1w, T2w, FLAIR, SWI and MRA, whole-brain and skull-stripped. Adult human anatomy.

Citations & references

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.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.MRSegmentatorMRI / CT · SegmentationSegment 40 organs, vessels, bones, and muscles from the chest to the hips on MRI or CT, with a body composition task for fat and abdominal muscles on MRI.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: MAISI-v2 v1.1. 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