MedSAM2 · azureml v3

<!-- DO NOT CHANGE MARKDOWN HEADERS. IF CHANGED, MODEL CARD MAY BE REJECTED BY A REVIEWER --> <!-- `description.md` is required. --> MedSAM2 is a breakthrough medical image segmentation foundation mo…

azureFoundry/azureml/MedSAM2@3

MedSAM2 is a breakthrough medical image segmentation foundation model that addresses the critical need for efficient and accurate segmentation in precision medicine. Built upon the Segment Anything Model (SAM) 2.1 architecture, it uniquely bridges the gap between 2D and 3D medical image segmentation, offering a unified solution for both volumetric scans and video data. The model stands out for its ability to reduce manual annotation costs by over 85% while maintaining high accuracy across diverse medical imaging applications.

The model has been extensively trained on over 455,000 3D image-mask pairs and 76,000 annotated video frames, spanning multiple organs, pathologies, and imaging protocols. MedSAM2's versatility has been validated through comprehensive user studies involving 5,000 CT lesions, 3,984 liver MRI lesions, and 251,550 echocardiogram vide

Specifications & capabilities

Provider specifications & access

1–1 / 1

MedSAM2 · azureml v3 ↗

source model id
azureml/MedSAM2@3
hosting
foundry-managed-compute
platforms
  • Microsoft Foundry
license
apache-2.0
architecture
input_modalities
    output_modalities
      capabilities

        Original sources

        1–3 / 3

        Microsoft Foundry · Model page ↗

        Microsoft Foundry · Model documentation ↗

        Microsoft Foundry · Catalog ↗

        Source links

        1–1 / 1

        Explore more models

        Models

        Loading
        EVIDENCE / COMPARISON

        Benchmarks

        Explore every benchmark

        Loading

        These charts contain model-linked results. The charts above also include unlinked source profiles. Compare scores within each evaluation. Model quality, usage, transparency and system performance measure different things.

        Install TensorTrace