The DOI matches its Crossref record; 43 of the 43 references it cites were found in a registry or an index; none of the 1 sources its claims cite is checked for support yet.
Segment anything model for medical image analysis: An experimental study
Open accessAbstract
Training segmentation models for medical images continues to be challenging due to the limited availability of data annotations. Segment Anything Model (SAM) is a foundation model trained on over 1 billion annotations, predominantly for natural images, that is intended to segment user-defined objects of interest in an interactive manner. While the model performance on natural images is impressive, medical image domains pose their own set of challenges. Here, we perform an extensive evaluation of SAM’s ability to segment medical images on a collection of 19 medical imaging datasets from various modalities and anatomies. In our experiments, we generated point and box prompts for SAM using a standard method that simulates interactive segmentation. We report the following findings: (1) SAM’s performance based on single prompts highly varies depending on the dataset and the task, from IoU=0.1135 for spine MRI to IoU=0.8650 for hip X-ray. (2) Segmentation performance appears to be better for well-circumscribed objects with prompts with less ambiguity such as the segmentation of organs in computed tomography and poorer in various other scenarios such as the segmentation of brain tumors. (3) SAM performs notably better with box prompts than with point prompts. (4) SAM outperforms similar methods RITM, SimpleClick, and FocalClick in almost all single-point prompt settings. (5) When multiple-point prompts are provided iteratively, SAM’s performance generally improves only slightly while other methods’ performance improves to the level that surpasses SAM’s point-based performance. We also provide several illustrations for SAM’s performance on all tested datasets, iterative segmentation, and SAM’s behavior given prompt ambiguity. We conclude that SAM shows impressive zero-shot segmentation performance for certain medical imaging datasets, but moderate to poor performance for others. SAM has the potential to make a significant impact in automated medical image segmentation in medical imaging, but appropriate care needs to be applied when using it. Code for evaluation SAM is made publicly available at https://github.com/mazurowski-lab/segment-anything-medical-evaluation.
Overview
AI snapshot
Written from the paper’s full text.
TL;DR
This experimental study evaluates the Segment Anything Model (SAM) for zero-shot medical image segmentation across 19 public datasets and 28 tasks. SAM shows highly variable performance, performs best with box prompts, and beats RITM/SimpleClick/FocalClick in most single-point settings, but gains little from iterative point prompts and remains moderate-to-poor on many medical tasks.
Key findings
- SAM’s single-prompt performance varied widely by dataset/task, from IoU=0.1135 for spine MRI to IoU=0.8650 for hip X-ray in the abstract; the Results section reports a range from IoU=0.1136 to IoU=0.9118.
- Box prompts were clearly superior to point prompts. Mode 4 (one box per object part) had the best average performance at IoU=0.6542; Mode 3 (single box around the entire object) averaged IoU=0.5891.
- SAM outperformed RITM, SimpleClick, and FocalClick on 24 of 28 tasks with a single non-iterative point prompt, and on 26 of 28 in oracle mode. Average IoU was 0.4595 for SAM, 0.5137 for SAM oracle, 0.2240 for FocalClick, 0.1910 for SimpleClick, and 0.1322 for RITM.
- With iterative point prompts, SAM improved only slightly and was surpassed by SimpleClick and RITM when five or more points were provided. The correlation between average object size and SAM performance was small and non-statistically significant.
Methods
- The authors curated 19 publicly available radiological medical imaging datasets covering X-ray, MRI, CT, ultrasound, and PET, yielding 28 segmentation tasks; for 3D modalities, 2D slices were used as inputs.
- They evaluated five non-iterative prompting modes using points or boxes, iterative point prompting guided by an error map, “segment everything” mode, and prompt ambiguity/oracle outputs.
- Performance was measured with IoU against ground-truth masks, using SAM’s highest-confidence mask; SAM was compared with RITM, SimpleClick, and FocalClick.
Limitations
- The paper states that text-based prompting is not commented on because it remains proof-of-concept, and that “segment everything” mode is not elaborated on. It also states future work should include prompt engineering and extension to 3D segmentation.
- The text makes evident that the study focuses on radiological images and uses 2D slices for 3D modalities, which may not reflect full 3D medical segmentation performance.
- The paper states that oracle performance assumes knowledge of the true mask and is biased when there is no additional user interaction after initial prompts; iterative prompting also uses the oracle prediction to guide the next prompt.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
12 Own result · 5 No basis given · 2 Own interpretation · 1 Not checked yet · 1 Own method
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“(1) SAM's performance based on single prompts highly varies depending on the dataset and the task, from IoU=0.1135 for spine MRI to IoU=0.8650 for hip X-ray.”
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“(3) SAM performs notably better with box prompts than with point prompts.”
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“When used in oracle mode, SAM was better than all other methods for 26 out of 28 tasks.”
§ 3. Results > 3.2. Comparing SAM to other interactive segmen-
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Moreover, as expected, prompts where each sep- arate part of the object is indicated separately are generally superior to those where only one part is indicated, or all parts are outlined in one box.”
§ 3. Results > 3.1. Performance of SAM for different modes of use
hedgedcorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This figure illustrates a high variability of SAM's performance ranging from near- perfect for well-circumscribed objects with unambiguous prompts to very poor, particularly for objects with ambigu- ous prompts.”
§ 3. Results > 3.1. Performance of SAM for different modes of use
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“If SAM used Mode 3 (single box), it outperformed all other methods in all but one task.”
§ 3. Results > 3.2. Comparing SAM to other interactive segmen-
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“These modes also showed different performances in such scenarios.”
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Following these two trends, Mode 4, where each part of an object is indicated by a separate box, showed the best performance with an average IoU of 0.6542.”
§ 3. Results > 3.1. Performance of SAM for different modes of use
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“The magnitude of this improvement was highly dependent on the dataset.”
§ 3. Results > 3.1. Performance of SAM for different modes of use
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Additionally, the oracle mode showed a moderate im- provement over the default mode.”
§ 3. Results > 3.1. Performance of SAM for different modes of use
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“It ranges from an impres- sive IoU of 0.9118 to a very poor IoU of 0.1136.”
§ 3. Results > 3.1. Performance of SAM for different modes of use
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This was particularly pro- nounced for datasets where objects typically consist of more than one part.”
§ 3. Results > 3.1. Performance of SAM for different modes of use
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“First, SAMs performance varies widely across different datasets.”
§ 3. Results > 3.1. Performance of SAM for different modes of use
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“When applying SAM on medical images, we found a consistent tendency of SAM in interpreting the point prompts.”
§ 3. Results > 3.4. Performance of SAM in the presence of ambi-
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Given this prompt-based nature, in its basic form, SAM cannot be used the same way as most segmentation models in medical imaging where the input is simply an image and the output is a segmentation mask or multiple masks for the desired object or objects.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- It may state two separate claims as one. 5
- The quote only partly states this claim. 1
- Probably not a substantial claim. 2
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“We find that this mode provides mixed usefulness and that the results are somewhat dependent on the number of prompts.”
§ 3. Results > 3.6. Segment-everything mode for medical images
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This is due to the fact that SAM does not appear to draw hardly any benefits from additional information provided through the interactive points after two or three points have been provided.”
§ 3. Results > 3.3. Performance of SAM and other methods for
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one. The quote only partly states this claim.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Some datasets also showed high variability of object size within the dataset itself (e.g., kidneys in ultrasound images) while others showed relative consistency of object size (such as bones in hip X-ray images).”
§ 3. Results > 3.5. Performance of SAM for objects of different
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Overall, SAM shows promise for use in medical images, as long as suitable prompting strategies are used for the dataset and task of choice.”
plainconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“Note that the correlation analysis looks at average object sizes in the image and does not consider the variation of object sizes in the individual datasets.”
§ 3. Results > 3.5. Performance of SAM for objects of different
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
In the paper
“The recently developed Segment Anything Model is a foundation model that has achieved promising zero-shot segmentation performance on a variety of natural image datasets [18].”
What the sources say
- Not checked yetcites [18] Jang, J., Hwang, D., 2022. M3T: Three-dimensional Medical image classifier using Multi-plane and Multi-slice Transformer. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 20718–20729. source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim.
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?
The cited source doesn’t support the claim as written.
In the citing paper
“Most previous studies (Ji et al., 2023a; Mazurowski et al., 2023; Zhou et al., 2023) evaluated SAM with limited or even only one type of testing mode / strategy.”
From Segment anything model for medical images? (2024) · cites this paper as Mazurowski et al., 2023
Checked by a reasoning model — may be wrong.
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Thus, the correlation between SAM's segmentation performance and medical objects' attributes has not been conducted carefully (He et al., 2023; Mazurowski et al., 2023; Ma and Wang, 2023).”
From Segment anything model for medical images? (2024) · cites this paper as Mazurowski et al., 2023
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“For example, the concurrent studies22,23 conducted a comprehensive assessment of SAM across a diverse array of medical images, underscoring that SAM achieved satisfactory segmentation outcomes primarily on targets characterized by distinct boundaries. However, the model exhibited substantial limitations in segmenting typical medical targets with weak boundaries or low contrast.”
From Segment anything in medical images (2024) · cites this paper as [22]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ 3. Results > 3.1. Performance of SAM for different modes of use
“Mode 4, where each part of an object is indicated by a separate box”
§ C. Performance of other competing methods
“M od e 2: 1 b ox a t e ac h re gi on”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ 2. Methodology > 2.1. Datasets
“We compiled and curated a set of 19 publicly available medical imaging datasets for image segmentation.”
§ C. Performance of other competing methods
“18 m ed ic al im ag in g da ta se ts”
Citations
Cites
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., et al. (2020) · arXiv · doi:10.48550/arxiv.2010.11929 · Cited by 47,472 · cited 1×
“Based on the framework proposed by RITM, SimpleClick replaced the HRNet-18 backbone in RITM with a plain-ViT [9] and included a “click embedding” layer symmetric to patch embedding.”
MethodThe multimodal brain tumor image segmentation benchmark (BRATS)
Menze, B. H., Jakab, A., Bauer, S., Kalpathy-Cramer, J., Farahani, K., Kirby, J., Burren, Y., Porz, N., Slotboom, J., Wiest, R., et al. (2015) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2014.2377694 · Cited by 6,781 · cited 1×
“MRI-Brain The Multimodal Brain Tumor Image Segmentation Benchmark (BraTS) [11] MRI 3 Three parts of brain tumor 13,107”
BackgroundSegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers
Xie, E., Wang, W., Yu, Z., Anandkumar, A., Alvarez, J. M., & Luo, P. (2021) · arXiv · doi:10.48550/arxiv.2105.15203 · Cited by 5,090 · cited 2×
“Fo-calClick replaced the HRNet-18 backbone in RITM with a SegFormer [37] and restricted the update responding to new clicks to occur only as local.”
MethodDeep High-Resolution Representation Learning for Visual Recognition
Wang, J., Sun, K., Cheng, T., Jiang, B., Deng, C., Zhao, Y., Liu, D., Mu, Y., Tan, M., Wang, X., et al. (2021) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2020.2983686 · Cited by 4,894 · cited 1×
“RITM adopted HRNet-18 [36] as the backbone segmentation model and an iterative mask correction approach based on the { previous mask, new click } set to achieve outstanding performance on multiple natural imaging datasets.”
MethodSegment anything in medical images
Ma, J., He, Y., Li, F., Han, L., You, C., & Wang, B. (2024) · Nature Communications · doi:10.1038/s41467-024-44824-z · Cited by 2,913
Dataset of breast ultrasound images
Al-Dhabyani, W., Gomaa, M., Khaled, H., & Fahmy, A. (2020) · Data in Brief · doi:10.1016/j.dib.2019.104863 · Cited by 2,405 · cited 1×
“US-Breast Dataset of Breast Ultrasound Images [2] Ultrasound 1 Breast 780”
Current Methods in Medical Image Segmentation
Pham, D. L., Xu, C., & Prince, J. L. (2000) · Annual Review of Biomedical Engineering · doi:10.1146/annurev.bioeng.2.1.315 · Cited by 2,352 · cited 1×
“Image segmentation is a central task in medical image analysis, ranging from the segmentation of organs [17], abnormalities [3], bones [8], and others [13].”
BackgroundDeep Learning Techniques for Medical Image Segmentation: Achievements and Challenges
Hesamian, M. H., Jia, W., He, X., & Kennedy, P. J. (2019) · Journal of Imaging Informatics in Medicine · doi:10.1007/s10278-019-00227-x · Cited by 1,710 · cited 1×
“Deep learning has dramatically altered how task is performed and brought significant performance improvements in many domains [6, 15].”
BackgroundMedical image analysis using convolutional neural networks: a review
Anwar, S. M., Majid, M., Qayyum, A., Awais, M., Alnowami, M., & Khan, M. K. (2018) · Journal of Medical Systems · doi:10.1007/s10916-018-1088-1 · Cited by 1,431 · cited 1×
“Image segmentation is a central task in medical image analysis, ranging from the segmentation of organs [17], abnormalities [3], bones [8], and others [13].”
BackgroundThe Liver Tumor Segmentation Benchmark (LiTS)
Bilic, P., Christ, P., Li, H. B., Vorontsov, E., Ben-Cohen, A., Kaissis, G., Szeskin, A., Jacobs, C., Mamani, G. E. H., Chartrand, G., et al. (2023) · Medical Image Analysis · doi:10.1016/j.media.2022.102680 · Cited by 1,412 · cited 1×
“CT-Liver The Liver Tumor Segmentation Benchmark (LiTS) [4] CT 1 Liver 5,501”
Background
Research works printed but not linked: 1
- Bradski, 2000, The OpenCV Library, Dr. Dobb’s J. Softw. Tools
Cited by
Sorted by citations
Segment anything in medical images
Ma, J., He, Y., Li, F., Han, L., You, C., & Wang, B. (2024) · Nature Communications · doi:10.1038/s41467-024-44824-z · Cited by 2,913
“These results highlight the potential of MedSAM as a powerful tool for medical image segmentation.”
Segment anything model for medical images?
Huang, Y., Yang, X., Liu, L., Zhou, H., Chang, A., Zhou, X., Chen, R., Yu, J., Chen, J., Chen, C., et al. (2024) · Medical Image Analysis · doi:10.1016/j.media.2023.103061 · Cited by 536
Transfer learning in environmental remote sensing
Ma, Y., Chen, S., Ermon, S., & Lobell, D. B. (2023) · Remote Sensing of Environment · doi:10.1016/j.rse.2023.113924 · Cited by 486
Segment anything model for medical image segmentation: Current applications and future directions
Zhang, Y., Shen, Z., & Jiao, R. (2024) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2024.108238 · Cited by 364
Self-citationOn the challenges and perspectives of foundation models for medical image analysis
Zhang, S., & Metaxas, D. (2023) · Medical Image Analysis · doi:10.1016/j.media.2023.102996 · Cited by 340
Metadata
- Authors
- Mazurowski, M. A., Dong, H., Gu, H., Yang, J., Konz, N., & Zhang, Y.
- Affiliations
- Duke University (US)
- Corresponding author
- Dong, Haoyu
- Venue
- Medical Image Analysis
- Publisher
- Elsevier BV
- Published
- 2023-10
- Volume
- 89
- Pages
- 102918
- ISSN
- 1361-8415 (ISSN-L)
- PMC
- PMC10528428
- OpenAlex
- W4385481295
- Semantic Scholar
- c6ac64877b2bcac2389be4d91a62dea8c6f73844
- Cited by
- 802
- References
- 40
- Citations by year
- 2026: 194 · 2025: 315 · 2024: 242 · 2023: 51
- FWCI
- 72.49
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Licence
- www.elsevier.com/tdm/userlicense/1.0/ (tdm)
- www.elsevier.com/legal/tdmrep-license (tdm)
- www.elsevier.com/open-access/userlicense/1.0/ (am, 320 d)
- doi.org/10.15223/policy-017 (stm-asf)
- doi.org/10.15223/policy-037 (stm-asf)
- doi.org/10.15223/policy-012 (stm-asf)
- doi.org/10.15223/policy-029 (stm-asf)
- doi.org/10.15223/policy-004 (stm-asf)
- Locations
- Medical Image Analysis (acceptedVersion)
- PubMed (publishedVersion)
- PubMed Central (acceptedVersion)
- arXiv (Cornell University) (submittedVersion, cc-by)
- arXiv (Cornell University) (cc-by)
- Funders
- National Institutes of Health
- National Heart, Lung, and Blood Institute (R44HL152825)
- National Institute of Biomedical Imaging and Bioengineering (R01EB031575)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Medical Image Segmentation Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Keywords
- Segment Anything Model, medical image segmentation, zero-shot segmentation, interactive segmentation, point prompts, iterative segmentation, brain tumor segmentation, organ segmentation, medical imaging modalities, foundation models
- MeSH terms
- Brain Neoplasms, Humans, S-Adenosylmethionine, Tomography, X-Ray Computed
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.
A language model drew the claims from the paper on 2026-10-08, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-08 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
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
