SynPo: Boosting Training-Free Few-Shot Medical Segmentation via High-Quality Negative Prompts

Haoke Xiao; Jiaxing Chai; Rong Wang; Yongcun Zhang; Yufei Liu; Zhiming Luo; Zijie Meng

REVIEW 5 major objections 5 minor 19 references

SynPo: Boosting Training-Free Few-Shot Medical Segmentation via High-Quality Negative Prompts

T0 review · 5 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Negative prompts placed inside the organ, just below target confidence, let training-free segmentation match trained methods.

desk verdict A clever negative-prompt heuristic with real gains, but a load-bearing equation is underspecified and the CHAOS numbers do not add up. read the letter →

arxiv 2506.15153 v2 pith:HNMH6VXH submitted 2025-06-18 cs.CV

Yufei Liu , Haoke Xiao , Jiaxing Chai , Yongcun Zhang , Rong Wang , Zijie Meng , Zhiming Luo This is my paper · ORCID

classification cs.CV

keywords few-shotmedicalimagesegmentationtraining-freeprompt-basednegativepromptselectionconfidencemapsynergySegmentAnythingModelDINOv2featuresGaussianbandabdominalCTandMRI

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

SynPo's central claim is that training-free few-shot medical segmentation fails not because the promptable segmenter is weak but because negative prompts are chosen badly. The paper proposes placing negative points inside the anatomical region, at confidence values a small distance below the target, and fusing DINOv2's semantic confidence map with SAM-ViT's spatial confidence map to make that placement reliable. On Synapse-CT and CHAOS-MRI, this prompt-only method reaches mean Dice of 79.91 and 81.15, surpassing the training-free baselines ProtoSAM and PerSAM and matching the training-based state of the art such as GMRD. If the claim holds, prompt engineering alone can substitute for training in low-annotation medical settings.

What carries the argument

The load-bearing object is the Synergy confidence map, defined as $\mathrm{SynMap} = \delta_{S\text{-}D}(S_{\mathrm{SAM}}\odot S_{\mathrm{DINO}}) + \delta_S S_{\mathrm{SAM}} + \delta_D S_{\mathrm{DINO}}$, which multiplies the two confidence maps so that only pixels where both encoders agree retain high confidence while a weighted sum preserves each encoder's strengths. The negative-confidence distribution is fit to the fused map for background features, and the point selection restricts negative candidates to the interval $[\mu - \alpha\sigma, \mu - \beta\sigma]$. K-means then condenses the selected coordinates into the final positive and negative prompts. This machinery is what lets the paper turn fusion and point placement into performance.

What would settle it

Run SynPo on a new organ or modality with a control that draws negative points from the background or uniformly; if the inside-the-organ band $[\mu - \alpha\sigma, \mu - \beta\sigma]$ does not beat the control in mean Dice, the central claim is falsified. A direct minimal test is to replace Eq. (9)'s band with $[\mu - 3\sigma, \mu - 2\sigma]$ on the same datasets, since the margin narrowing or reversing would show the chosen band itself is load-bearing.

Watch

Extended reading notes

Core claim

The discovery is a placement rule for negative prompts in point-promptable segmentation. Instead of choosing pixels least similar to the target, which tend to fall in background that SAM already ignores, SynPo models the distribution of negative confidence values with a Gaussian and draws negative points from the band $[\mu - \alpha\sigma, \mu - \beta\sigma]$, i.e. pixels that are less similar than the target but still within or near the anatomical region. These negative points, together with top-ranked positive points condensed by K-means, are fed to SAM. On the four-organ abdominal benchmarks the rule yields 79.91 mean Dice on Synapse-CT and 81.15 on CHAOS-MRI, outperforming training-free baselines (ProtoSAM 73.45 and PerSAM 62.34 on Synapse) and reaching the level of trained methods such as GMRD (78.52 on Synapse).

Load-bearing premise

The load-bearing premise is that SAM is best guided by negative points placed inside the anatomical region at confidence values just below the target, rather than by background points; if that behavior does not hold for other organs or modalities, the gains collapse.

Editorial extensions

If this is right

Training-free few-shot segmentation can reach trained-method accuracy on abdominal CT and MRI when negative prompts are selected from inside the anatomical region, suggesting that annotation cost can be cut without retraining.
The confidence-map fusion rule is transferable: the same fixed weights ($\delta_{S\text{-}D}=0.8$, $\delta_S=\delta_D=0.1$) work on both Synapse-CT and CHAOS-MRI, so the method does not need per-dataset retuning.
Replacing PerSAM's negative selection with SynPo's band rule adds 13.18 Dice on CHAOS-MRI, implying that negative prompt placement, not the positive points alone, is the dominant source of gain in this paradigm.
The Noise-aware Refine module adds only a smaller gain (+1.19 Dice) after the point-selection gains, so future work can focus on prompt placement rather than refinement.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

If the negative-band principle is model-agnostic, the same Gaussian-band selection should transfer to other promptable segmentation models; a cross-backbone test would separate the principle from SAM-specific behavior.
The paper fixes $\alpha$ and $\beta$ globally; per-organ tuning might raise scores further, or reveal that the band is really selecting boundary-adjacent pixels rather than simply 'less similar' pixels.
Because the synergy weight on the product term is 0.8, an ablation that varies $\delta_{S\text{-}D}$ would clarify how much of the gain comes from the fusion as opposed to the negative-prompt rule.
The pilot evidence for the core premise is qualitative; a systematic quantitative study across organs and modalities would test whether inside-the-organ negative points are always better than background negatives.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Short version: the core idea—pick negative prompts inside the organ, just below target confidence—is genuinely interesting, and the reported gains over PerSAM/ProtoSAM look large. But the paper has a central ambiguity in Eq. (9) plus an arithmetic slip, and until those are resolved I would not trust the headline numbers.

What is new: fusing DINOv2 and SAM confidence maps is sensible; using a Gaussian band on that map to choose negative points is a concrete, training-free heuristic. The ablation replacing PerSAM's negatives with this strategy (+13.18 Dice on CHAOS) is the most convincing evidence. The method is training-free and competitive with training-based approaches, which matters for medical imaging.

Soft spots, in order of severity. First, the CHAOS-MRI mean is wrong: 80.30, 77.32, 77.32, 83.04 average to 79.5, not 81.15. That needs a correction. Second, and more serious, Eq. (9) defines p_i as a pixel on the synergy map but thresholds its confidence using mu and sigma from SynMap_neg, a different distribution on a different scale. The final alpha and beta are not disclosed, and Fig. 4 suggests the best CHAOS setting is [mu, mu+1.5sigma], which on the background-similarity map would select more background-like pixels—contradicting the paper's "inside the anatomical region" rationale. Without code or disclosed hyperparameters, the reader cannot tell which variant actually produced the gains. This is not a nitpick; the negative-prompt selection is the paper's reason to exist. Third, hyperparameters appear to be tuned on evaluation data, and most baselines are quoted from other papers without matched folds or error bars. Those are common, perhaps acceptable, but should be acknowledged.

Credit where due: the pilot experiment (Fig. 1(3)) is a nice sanity check, the confidence-map fusion is well motivated, and the reported results, if reproducible, would be a useful data point for training-free medical segmentation.

Bottom line: worth a serious referee, but not in current form. I would ask for code, a corrected Eq. (9) with explicit use of the right map, final hyperparameter values, and a rerun with validation-based selection. Then the claim "comparable to training-based SOTA" might hold. As it stands, I would be uncomfortable citing it.

Referee Report

5 major / 5 minor

Summary. This paper proposes SynPo, a training-free few-shot medical image segmentation method built on SAM and DINOv2. It combines DINOv2 semantic features and SAM spatial cues into a confidence map, selects positive point prompts from high-confidence pixels, and selects negative point prompts from a Gaussian-defined confidence band via K-means clustering, then refines the coarse mask. Experiments on Synapse-CT and CHAOS-MRI report mean Dice values of 79.91 and 81.15, claiming parity with training-based few-shot methods and large gains over PerSAM and ProtoSAM.

Significance. If the results hold, the paper would be significant: it suggests that prompt engineering alone, using frozen LVMs, can match training-based few-shot segmentation on abdominal CT and MRI. The strengths are the simple training-free framework, the concrete hypothesis that informative negative prompts lie inside the anatomical region below the target confidence, and an ablation isolating the negative-point strategy. However, as written, the empirical case is not yet persuasive: the headline CHAOS-MRI mean is arithmetically inconsistent, the negative-selection equation mixes confidence maps with incompatible statistics, and key hyperparameters and comparison protocols are undisclosed.

major comments (5)

[Section 3.2, Table 1] The reported mean Dice for CHAOS-MRI is internally inconsistent. The four organ scores in the SynPo row, 80.30 (spleen), 77.32 (liver), 77.32 (left kidney), and 83.04 (right kidney), average to 79.50, not the reported 81.15. The Synapse-CT row averages to 79.93, matching the reported 79.91, so the inconsistency is specific to CHAOS-MRI. This affects the abstract, Section 3.2, and the central claim of parity with GMRD (82.90). Please correct the arithmetic and re-check all summary statistics.
[Section 2.3, Eq. (9)] Eq. (9) selects negative pixels using c_i, defined as the confidence of a pixel p_i in SynMap, but thresholds c_i against μ and σ estimated from SynMap_neg in Eq. (6). SynMap and SynMap_neg are different maps with different value distributions; applying background-similarity statistics to target-similarity values has no stated justification. If the selection is intended on SynMap_neg, the notation p_i=(x_i,y_i,c_i) should be changed; if it is intended on SynMap, the thresholding rule needs a distributional argument or empirical calibration. This ambiguity is central because Eq. (9) is the core negative-prompt selection mechanism.
[Section 2.3 and Section 3.3] The final configuration of the Point Selection Module is underspecified. The paper does not report the values of α, β, γ_1, γ_2, K_p, or K_n used for the results in Table 1 and the ablations in Tables 2 and 3. Fig. 4 sweeps α with β = α - 1.5 but does not state which point of the sweep was selected or whether the same values were used on Synapse-CT and CHAOS-MRI. Without these parameters, Eqs. (9)-(11) cannot be evaluated or reproduced, and the reported +12.04 and +13.18 gains cannot be traced to a specific selection rule.
[Section 3.2, Table 1] The comparison protocol does not support the strength of the claims. Except for PerSAM and SynPo, all baseline numbers are sourced from their original papers, so the comparison is not matched in folds, image preprocessing, or inference settings. No error bars are given for the training-based methods, and no significance test is reported. The statement that SynPo 'outperforms' training-based methods in spleen and right kidney on Synapse-CT should be softened or backed by a matched evaluation.
[Section 3.3, Table 2] The ablation 'SAM + DINO + PSM' improves from 66.70 to 78.74 Dice, but this row combines the confidence-map fusion, positive-point selection, and negative-point selection. Since the central claim is about negative-prompt quality, the ablation should separate the negative-point selection from the positive-point and clustering components. Table 3 does this on CHAOS-MRI by replacing only PerSAM's negative selection, but the Synapse-CT ablation does not isolate the contribution of Eq. (9).

minor comments (5)

[Section 2.3, Eq. (10)] In Eq. (10), the condition 'if |P_neg| ≤ γ_2·K_n' should refer to |P'_neg|, since P_neg is the final clustered set defined later in Eq. (11).
[Section 2.2, Eq. (3)] In Eq. (3), the notation 'Ms' appears to denote the inverse of the support mask; using an explicit overbar, e.g., \bar{M}_S, would avoid confusion with the original support mask M_S.
[Section 2.2 and Section 2.3] The spatial dimensions h, w, H, and W are not consistently fixed; please clarify whether the confidence maps are 64×64 and how the point coordinates are rescaled when they are fed into SAM.
[Fig. 4] The label 'RawData' in Fig. 4 is not defined in the text; please specify which dataset and preprocessing it refers to, and explain why it is plotted alongside CHAOS.
[Throughout] There are several typographical issues, including 'yeiled' in Section 2.4 and a stray '3' in the ProtoSAM standard-deviation row of Table 1; a careful proofread is needed.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: SynPo's reported Dice gains are measured on external benchmarks, not derived from assumptions that contain the outcome.

full rationale

I walked the claimed derivation chain. SynPo is an empirical prompt-selection pipeline: Eq. (4)-(5) fuse two fixed feature-similarity maps, Eq. (6) fits a Gaussian to the background-similarity map, and Eqs. (7)-(12) select and cluster point prompts before feeding them to SAM. None of these steps derive a segmentation result from an assumption that already contains the reported Dice scores. The headline numbers (79.91 Synapse, 81.15 CHAOS) are measurements on external benchmarks against external baselines, not predictions recovered from fitted parameters. The hyperparameters delta_S-D, delta_S, delta_D and the alpha/beta band are tuned, but tuning hyperparameters on evaluation data is a fitting practice, not a circular derivation unless the 'prediction' is the fitted value itself, which is not the case here. The self-citations ([12], [13], [16]) are non-load-bearing: they support a general process description and a peripheral remark about over-segmentation, and none is used as a uniqueness theorem or as a substitute for the paper's measurements. The notation mismatch in Eq. (9) (c_i defined on the synergy map while mu, sigma come from SynMap_neg) is a real reproducibility/correctness concern, but it is not circularity: it does not exhibit an output that equals an input by construction. No circular step meets the required evidentiary standard, so the appropriate finding is no significant circularity.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The method introduces no new postulates beyond the heuristic distributional and complementarity assumptions above. Its contribution is in the prompt-selection engineering, which relies on a modest set of manually tuned hyperparameters that are not fully specified.

free parameters (4)

delta_S-D, delta_S, delta_D = 0.8, 0.1, 0.1
Fusion weights in Eq. (4) are fixed across datasets without a tuning procedure; they control the synergy map and are chosen manually.
alpha, beta = Unspecified in final method; explored in Fig. 4
Confidence interval bounds in Eq. (9). The parameter experiment tunes them on CHAOS and RawData, but the final values used for Tables 1-3 are never stated.
gamma_1, gamma_2 = Not reported
Clustering scaling factors used to oversample before K-means (Eq. 7, Eq. 10); no values or selection procedure are described.
K_p, K_n = Not reported
Desired numbers of positive and negative point prompts; never specified in the paper.

assumptions (4)

domain assumption Cosine similarity between support and query features measures semantic correspondence.
Used throughout to build confidence maps (Eq. 2); inherited from ProtoSAM and PerSAM without revalidation on medical data.
domain assumption DINOv2 features provide semantic but location-blurred cues, while SAM-ViT features provide spatial precision.
The motivating claim of the Confidence Map Synergy Module (Section 2.2, Fig. 1(2)).
domain assumption Negative confidence map values follow a Gaussian distribution.
Eq. (6) models pixel values with a Gaussian; no normality check is reported, and the subsequent interval selection relies on this distribution.
standard math K-means centroids are representative prompt locations.
Used to downsample candidates to Kp/Kn prompts (Eq. 8, Eq. 11); standard clustering assumption.

how reviews work

0 comments

Cite this review

Pith. "Pith review of SynPo: Boosting Training-Free Few-Shot Medical Segmentation via High-Quality Negative Prompts." pith.science (2026). https://pith.science/paper/HNMH6VXH

@misc{pith2026250615153,
  author       = {Pith},
  title        = {Pith review of: SynPo: Boosting Training-Free Few-Shot Medical Segmentation via High-Quality Negative Prompts},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HNMH6VXH}},
  note         = {Machine review of arXiv:2506.15153}
}

read the original abstract

The advent of Large Vision Models (LVMs) offers new opportunities for few-shot medical image segmentation. However, existing training-free methods based on LVMs fail to effectively utilize negative prompts, leading to poor performance on low-contrast medical images. To address this issue, we propose SynPo, a training-free few-shot method based on LVMs (e.g., SAM), with the core insight: improving the quality of negative prompts. To select point prompts in a more reliable confidence map, we design a novel Confidence Map Synergy Module by combining the strengths of DINOv2 and SAM. Based on the confidence map, we select the top-k pixels as the positive points set and choose the negative points set using a Gaussian distribution, followed by independent K-means clustering for both sets. Then, these selected points are leveraged as high-quality prompts for SAM to get the segmentation results. Extensive experiments demonstrate that SynPo achieves performance comparable to state-of-the-art training-based few-shot methods.

Figures

Figures reproduced from arXiv: 2506.15153 by the authors.

**Figure 1.** Challenges in Training-Free Few-Shot Segmentation. (1) The universal process of point promptable segmentation model for training-free few-shot. (2) Different Confidence Maps (C. Map) vs. Ground Truth. In the confidence map generated by DINOv2 features, irrelevant regions on the right are mistakenly identified as "similar". The confidence map from SAM-ViT features exhibits less clear differentiation in values. Our … view at source ↗

**Figure 2.** (1)Overview of SynPo Architecture. (2) Illustration of Confidence Map Synergy. (3) Point Selection Module Diagram. To address these limitations, we propose SynPo, a novel training-free method with Confidence Map Synergy Module and Point Selection Module. First, drawing from SAM’s ability to capture precise low-level spatial information [6,10], which can be a supplement of features from DINOv2, we introduce the Conf… view at source ↗

**Figure 3.** Quality results of different methods. Performance on CHAOS-MRI. SynPo achieves an average Dice score of 81.15, outperforming training-free methods like ProtoSAM (77.83) and PerSAM (61.12), while performing competitively with the top training-based methods like GMRD (82.90). It also excels in spleen segmentation (80.30) and maintains competitive performance on other organs, demonstrating strong generalization. Qualit… view at source ↗

Figures from the paper (1 more)

**Figure 4.** Figure 4: Parameter Experiment of α & β, β = α − 1.5. Additionally, we replaced the negative point selection strategy in PerSAM to evaluate its impact on CHAOS-MRI ( [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

19 extracted references · 5 canonical work pages

[1]

arXiv preprint arXiv:2407.07042 (2024)

Ayzenberg, L., Giryes, R., Greenspan, H.: Protosam: One-shot medical image seg- mentation with foundational models. arXiv preprint arXiv:2407.07042 (2024)

arXiv 2024
[2]

IEEE Transactions on Medical Imaging43(6), 2202–2214 (2024)

Cheng, Z., Wang, S., Xin, T., Zhou, T., Zhang, H., Shao, L.: Few-shot medi- cal image segmentation via generating multiple representative descriptors. IEEE Transactions on Medical Imaging43(6), 2202–2214 (2024)

2024
[3]

arXiv preprint arXiv:2010.11929 (2020)

Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., et al.: An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929 (2020)

arXiv 2020
[4]

Medical Image Analysis78, 102385 (2022)

Hansen, S., Gautam, S., Jenssen, R., Kampffmeyer, M.: Anomaly detection- inspired few-shot medical image segmentation through self-supervision with su- pervoxels. Medical Image Analysis78, 102385 (2022)

2022
[5]

Medical image analysis69, 101950 (2021)

Kavur, A.E., Gezer, N.S., Barış, M., Aslan, S., Conze, P.H., Groza, V., Pham, D.D., Chatterjee, S., Ernst, P., Özkan, S., et al.: Chaos challenge-combined (ct- mr) healthy abdominal organ segmentation. Medical image analysis69, 101950 (2021)

2021
[6]

In: Proceedings of the IEEE/CVF International Conference on Computer Vision

Kirillov, A., Mintun, E., Ravi, N., Mao, H., Rolland, C., Gustafson, L., Xiao, T., Whitehead, S., Berg, A.C., Lo, W.Y., et al.: Segment anything. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 4015–4026 (2023)

2023
[7]

In: Proc

Landman, B., Xu, Z., Igelsias, J., Styner, M., Langerak, T., Klein, A.: Miccai multi-atlas labeling beyond the cranial vault–workshop and challenge. In: Proc. MICCAI multi-atlas labeling beyond cranial vault—workshop challenge. vol. 5, p. 12. Munich, Germany (2015)

work page 2015
[8]

arXiv preprint arXiv:2304.07193 (2023)

Oquab, M., Darcet, T., Moutakanni, T., Vo, H., Szafraniec, M., Khalidov, V., Fernandez, P., Haziza, D., Massa, F., El-Nouby, A., et al.: Dinov2: Learning robust visual features without supervision. arXiv preprint arXiv:2304.07193 (2023)

arXiv 2023

Show all 19 references

[9]

In: Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XXIX 16

Ouyang, C., Biffi, C., Chen, C., Kart, T., Qiu, H., Rueckert, D.: Self-supervision with superpixels: Training few-shot medical image segmentation without annota- tion. In: Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part X...

2020
[10]

arXiv preprint arXiv:2408.00714 (2024)

Ravi, N., Gabeur, V., Hu, Y.T., Hu, R., Ryali, C., Ma, T., Khedr, H., Rädle, R., Rolland, C., Gustafson, L., et al.: Sam 2: Segment anything in images and videos. arXiv preprint arXiv:2408.00714 (2024)

2024 arXiv
[11]

In: Proceedings of SAI Intelligent Systems Conference

Shen, Q., Li, Y., Jin, J., Liu, B.: Q-net: Query-informed few-shot medical image segmentation. In: Proceedings of SAI Intelligent Systems Conference. pp. 610–628. Springer (2023)

2023
[12]

International Journal of Computer Vision133(1), 1–15 (2025)

Tang, L., Jiang, P.T., Xiao, H., Li, B.: Towards training-free open-world segmen- tation via image prompt foundation models. International Journal of Computer Vision133(1), 1–15 (2025)

2025
[13]

arXiv preprint arXiv:2304.04709 (2023) 10 Y

Tang, L., Xiao, H., Li, B.: Can sam segment anything? when sam meets camou- flaged object detection. arXiv preprint arXiv:2304.04709 (2023) 10 Y. Liu et al

2023 arXiv
[14]

Advances in Neural Information Processing Systems (2017)

Vaswani, A.: Attention is all you need. Advances in Neural Information Processing Systems (2017)

2017
[15]

In: European Conference on Computer Vision

Wang, H., Zhang, T., Salzmann, M.: Sinder: Repairing the singular defects of dinov2. In: European Conference on Computer Vision. pp. 20–35. Springer (2024)

2024
[16]

Biomedical Signal Processing and Control100, 107069 (2025)

Wang, R., Yang, Z., Song, Y.: Osam-fundus: A training-free, one-shot segmentation framework for optic disc and cup in fundus images. Biomedical Signal Processing and Control100, 107069 (2025)

2025
[17]

Advances in Neural Information Processing Systems36 (2024)

Zhang, J., Herrmann, C., Hur, J., Polania Cabrera, L., Jampani, V., Sun, D., Yang, M.H.: A tale of two features: Stable diffusion complements dino for zero-shot semantic correspondence. Advances in Neural Information Processing Systems36 (2024)

2024
[18]

arXiv preprint arXiv:2305.03048 (2023)

Zhang, R., Jiang, Z., Guo, Z., Yan, S., Pan, J., Ma, X., Dong, H., Gao, P., Li, H.: Personalize segment anything model with one shot. arXiv preprint arXiv:2305.03048 (2023)

2023 arXiv
[19]

In: International Conference on Med- ical Image Computing and Computer-Assisted Intervention

Zhu, Y., Wang, S., Xin, T., Zhang, H.: Few-shot medical image segmentation via a region-enhanced prototypical transformer. In: International Conference on Med- ical Image Computing and Computer-Assisted Intervention. pp. 271–280. Springer (2023)

2023

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.

[1] [1]

arXiv preprint arXiv:2407.07042 (2024)

Ayzenberg, L., Giryes, R., Greenspan, H.: Protosam: One-shot medical image seg- mentation with foundational models. arXiv preprint arXiv:2407.07042 (2024)

arXiv 2024

[2] [2]

IEEE Transactions on Medical Imaging43(6), 2202–2214 (2024)

Cheng, Z., Wang, S., Xin, T., Zhou, T., Zhang, H., Shao, L.: Few-shot medi- cal image segmentation via generating multiple representative descriptors. IEEE Transactions on Medical Imaging43(6), 2202–2214 (2024)

2024

[3] [3]

arXiv preprint arXiv:2010.11929 (2020)

Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., et al.: An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929 (2020)

arXiv 2020

[4] [4]

Medical Image Analysis78, 102385 (2022)

Hansen, S., Gautam, S., Jenssen, R., Kampffmeyer, M.: Anomaly detection- inspired few-shot medical image segmentation through self-supervision with su- pervoxels. Medical Image Analysis78, 102385 (2022)

2022

[5] [5]

Medical image analysis69, 101950 (2021)

Kavur, A.E., Gezer, N.S., Barış, M., Aslan, S., Conze, P.H., Groza, V., Pham, D.D., Chatterjee, S., Ernst, P., Özkan, S., et al.: Chaos challenge-combined (ct- mr) healthy abdominal organ segmentation. Medical image analysis69, 101950 (2021)

2021

[6] [6]

In: Proceedings of the IEEE/CVF International Conference on Computer Vision

Kirillov, A., Mintun, E., Ravi, N., Mao, H., Rolland, C., Gustafson, L., Xiao, T., Whitehead, S., Berg, A.C., Lo, W.Y., et al.: Segment anything. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 4015–4026 (2023)

2023

[7] [7]

In: Proc

Landman, B., Xu, Z., Igelsias, J., Styner, M., Langerak, T., Klein, A.: Miccai multi-atlas labeling beyond the cranial vault–workshop and challenge. In: Proc. MICCAI multi-atlas labeling beyond cranial vault—workshop challenge. vol. 5, p. 12. Munich, Germany (2015)

work page 2015

[8] [8]

arXiv preprint arXiv:2304.07193 (2023)

Oquab, M., Darcet, T., Moutakanni, T., Vo, H., Szafraniec, M., Khalidov, V., Fernandez, P., Haziza, D., Massa, F., El-Nouby, A., et al.: Dinov2: Learning robust visual features without supervision. arXiv preprint arXiv:2304.07193 (2023)

arXiv 2023