CoChat

The DOI matches its Crossref record; 51 of the 51 references it cites were found in a registry or an index; 13 of the 26 sources its claims cite were checked for support.

Segment anything in medical images

Open access
Ma, Jun, He, Yuting, Li, Feifei, Han, Lin, You, Chenyu, Wang, Bo · 2024 · Nature Communications

Abstract

Medical image segmentation is a critical component in clinical practice, facilitating accurate diagnosis, treatment planning, and disease monitoring. However, existing methods, often tailored to specific modalities or disease types, lack generalizability across the diverse spectrum of medical image segmentation tasks. Here we present MedSAM, a foundation model designed for bridging this gap by enabling universal medical image segmentation. The model is developed on a large-scale medical image dataset with 1,570,263 image-mask pairs, covering 10 imaging modalities and over 30 cancer types. We conduct a comprehensive evaluation on 86 internal validation tasks and 60 external validation tasks, demonstrating better accuracy and robustness than modality-wise specialist models. By delivering accurate and efficient segmentation across a wide spectrum of tasks, MedSAM holds significant potential to expedite the evolution of diagnostic tools and the personalization of treatment plans.

Overview

CoChat Summary

TL;DR

MedSAM is a promptable foundation model for universal medical image segmentation, developed by fine-tuning SAM on 1,570,263 medical image-mask pairs across 10 imaging modalities and over 30 cancer types. It outperformed SAM and often matched or surpassed modality-wise specialist models across 86 internal and 60 external validation tasks.

Key findings

  • MedSAM was trained on 1,570,263 image-mask pairs covering 10 imaging modalities and over 30 cancer types, and evaluated on 86 internal and 60 external segmentation tasks.
  • On external validation, MedSAM achieved a median DSC of 87.8% (IQR 85.0–91.4%) for nasopharynx cancer segmentation, improving over SAM, specialist U-Net, and DeepLabV3+ by 52.3%, 15.5%, and 22.7%, respectively.
  • MedSAM ranked first on most internal and external tasks, while SAM ranked last on almost all tasks; specialist models showed limited generalization on unseen targets.
  • In a human annotation study, MedSAM-assisted annotation reduced time by 82.37% and 82.95% for two experts, and scaling training data from 10K to 100K to the full dataset improved performance.

Methods

  • Curated public medical image segmentation datasets into 1,570,263 image-mask pairs across 10 modalities; images were normalized and resized to 1024 × 1024 × 3, with 3D CT/MR handled as 2D slices.
  • MedSAM follows the SAM architecture with a ViT-Base image encoder, prompt encoder for bounding boxes, and mask decoder; it was initialized from SAM, fixed the prompt encoder, and trained the image encoder and mask decoder with combined dice and cross-entropy loss using AdamW, batch size 160, and 150 epochs on 20 A100 GPUs.
  • Internal validation used an 80%/10%/10% training/tuning/validation split, while external validation datasets were held out; comparisons included SAM, modality-wise U-Net (nnU-Net), and DeepLabV3+ specialists, with DSC and normalized surface distance metrics and Wilcoxon signed-rank tests.

Limitations

  • The paper states modality imbalance in the training set, with CT, MRI, and endoscopy dominating, potentially affecting less-represented modalities such as mammography.
  • The paper states difficulty segmenting vessel-like branching structures because bounding box prompts can be ambiguous, e.g., arteries and veins sharing the same box in eye fundus images.
  • Evident from the text: the human annotation study used only 10 adrenal tumor cases (733 slices) and two radiologists, so the reported time savings may not generalize broadly.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1038/s41467-024-44824-z

Cites

51 of 51 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Fig. 1 MedSAM is trained on a large-scale dataset that can handle diverse segmentation tasks. The dataset covers a variety of anatomical structures, pathological conditions, and medical imaging modalities. The magenta contours and mask overlays denote the expert annotations and MedSAM segmentation results, respectively.
Fig. 1 MedSAM is trained on a large-scale dataset that can handle diverse segmentation tasks. The dataset covers a variety of anatomical structures, pathological conditions, and medical imaging modalities. The magenta contours and mask overlays denote the expert annotations and MedSAM segmentation results, respectively.CC BY · PubMed Central
Fig. 2 Overview of the modality distribution in the dataset and the network architecture. a The number of medical image-mask pairs in each modality. b MedSAM is a promptable segmentation method where users can use bounding boxes to specify the segmentation targets. Source data are provided as a Source Data file.
Fig. 2 Overview of the modality distribution in the dataset and the network architecture. a The number of medical image-mask pairs in each modality. b MedSAM is a promptable segmentation method where users can use bounding boxes to specify the segmentation targets. Source data are provided as a Source Data file.CC BY · PubMed Central
Fig. 3 Quantitative and qualitative evaluation results on the internal validation set. a Performance distribution of 86 internal validation tasks in terms of median dice similarity coefficient (DSC) score. The center line within the box represents the median value, with the bottom and top bounds of the box delineating the 25th and 75th percentiles, respectively. Whiskers are chosen to show the 1.5 of the interquartile range. Up-triangles denote the minima and down-triangles denote the maxima. b Podium plots for visualizing the performance correspondence of 86 internal validation tasks. Upper part: each colored dot denotes the median DSC achieved with the respective method on one task. Dots corresponding to identical tasks are connected by a line. Lower part: bar charts represent the frequency of achieved ranks for each method. MedSAM ranks in the first place on most tasks. c Visualized segmentation examples on the internal validation set. The four examples are liver cancer, brain cancer, breast cancer, and polyp in computed tomography (CT), (Magnetic Resonance Imaging) MRI, ultrasound, and endoscopy images, respectively. Blue: bounding box prompts; Yellow: segmentation results. Magenta: expert annotations. Source data are provided as a Source Data file.
Fig. 3 Quantitative and qualitative evaluation results on the internal validation set. a Performance distribution of 86 internal validation tasks in terms of median dice similarity coefficient (DSC) score. The center line within the box represents the median value, with the bottom and top bounds of the box delineating the 25th and 75th percentiles, respectively. Whiskers are chosen to show the 1.5 of the interquartile range. Up-triangles denote the minima and down-triangles denote the maxima. b Podium plots for visualizing the performance correspondence of 86 internal validation tasks. Upper part: each colored dot denotes the median DSC achieved with the respective method on one task. Dots corresponding to identical tasks are connected by a line. Lower part: bar charts represent the frequency of achieved ranks for each method. MedSAM ranks in the first place on most tasks. c Visualized segmentation examples on the internal validation set. The four examples are liver cancer, brain cancer, breast cancer, and polyp in computed tomography (CT), (Magnetic Resonance Imaging) MRI, ultrasound, and endoscopy images, respectively. Blue: bounding box prompts; Yellow: segmentation results. Magenta: expert annotations. Source data are provided as a Source Data file.CC BY · PubMed Central
Fig. 4 Quantitative and qualitative evaluation results on the external validation set. a Performance distribution of 60 external validation tasks in terms of median dice similarity coefficient (DSC) score. The center line within the box represents the median value, with the bottom and top bounds of the box delineating the 25th and 75th percentiles, respectively. Whiskers are chosen to show the 1.5 of the interquartile range. Up-triangles denote the minima and down-triangles denote the maxima. b Podium plots for visualizing the performance correspondence of 60 external validation tasks. Upper part: each colored dot denotes the median DSC achieved with the respective method on one task. Dots corresponding to identical tasks are connected by a line. Lower part: bar charts represent the frequency of achieved ranks for each method. MedSAM ranks in the first place on most tasks. c Visualized segmentation examples on the external validation set. The four examples are the lymph node, cervical cancer, fetal head, and polyp in CT, MR, ultrasound, and endoscopy images, respectively. Source data are provided as a Source Data file.
Fig. 4 Quantitative and qualitative evaluation results on the external validation set. a Performance distribution of 60 external validation tasks in terms of median dice similarity coefficient (DSC) score. The center line within the box represents the median value, with the bottom and top bounds of the box delineating the 25th and 75th percentiles, respectively. Whiskers are chosen to show the 1.5 of the interquartile range. Up-triangles denote the minima and down-triangles denote the maxima. b Podium plots for visualizing the performance correspondence of 60 external validation tasks. Upper part: each colored dot denotes the median DSC achieved with the respective method on one task. Dots corresponding to identical tasks are connected by a line. Lower part: bar charts represent the frequency of achieved ranks for each method. MedSAM ranks in the first place on most tasks. c Visualized segmentation examples on the external validation set. The four examples are the lymph node, cervical cancer, fetal head, and polyp in CT, MR, ultrasound, and endoscopy images, respectively. Source data are provided as a Source Data file.CC BY · PubMed Central
Fig. 5 The effect of training dataset size and a user study of tumor annotation efficiency. a Scaling up the training image size to one million can significantly improve the model performance on both internal and external validation sets. b MedSAM can be used to substantially reduce the annotation time cost. Source data are provided as a Source Data file.
Fig. 5 The effect of training dataset size and a user study of tumor annotation efficiency. a Scaling up the training image size to one million can significantly improve the model performance on both internal and external validation sets. b MedSAM can be used to substantially reduce the annotation time cost. Source data are provided as a Source Data file.CC BY · PubMed Central

Verification

Claims the paper makes

7 Valid · 3 Not valid · 5 Not checked · 39 Own results and interpretations

25 more claims

Create a free account to see all 54 claims and to verify the 1 reference still open.

See all claims — free account
  • Probably not a substantial claim. 9
  • It may state two separate claims as one. 18
  • The quote only partly states this claim. 1

Create a free account to see these claims.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Citations

Cites

Show all 44 references

Research works printed but not linked: 1

  1. Vaswani, A. et al. Attention is all you need. In Advances in Neural Information Processing Systems, Vol. 30 (Curran Associates, Inc., 2017).

Cited by

See all 2,930

Metadata

Authors
Ma, J., He, Y., Li, F., Han, L., You, C., & Wang, B.
Affiliations
  • University Health Network (CA)
  • Western University (CA)
  • New York University (US)
  • Yale University (US)
Corresponding author
Wang, Bo
Venue
Nature Communications
Published
2024-01-22
Volume
15
Issue
1
Pages
654-654
ISSN
2041-1723 (ISSN-L)
OpenAlex
W4391109864
Cited by
2,913
References
44
Citations by year
2026: 1023 · 2025: 1233 · 2024: 612
FWCI
622.56
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Communications (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • PubMed Central (submittedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada (RGPIN-2020-06189 and DGECR-2020-00294)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › AI in cancer detection
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Medical Image Segmentation Techniques
  • Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › COVID-19 diagnosis using AI
Keywords
medical image segmentation, MedSAM, foundation models, cross-modal generalization, external validation
MeSH terms
Diagnostic Imaging, Image Processing, Computer-Assisted

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (qwen3.8-omni-flash and gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Segment anything in medical images. https://app.cochat.ai/papers/10.1038/s41467-024-44824-z
The paper
Ma, J., He, Y., Li, F., Han, L., You, C., & Wang, B. (2024). Segment anything in medical images. Nature Communications. https://doi.org/10.1038/s41467-024-44824-z

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.