Pith. sign in

REVIEW 4 major objections 6 minor 26 references

Towards Affordable Tumor Segmentation and Visualization for 3D Breast MRI Using SAM2

T0 review · 4 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

Pith's one-line read A zero-shot video-segmentation model can turn a single annotated slice of a breast MRI into a full 3D tumor segmentation.

desk verdict Useful empirical comparison of SAM2 propagation strategies for breast MRI, but the mask-based tracking result is likely inflated by including the ground-truth prompt slice in the Dice computation. read the letter →

arxiv 2507.23272 v1 pith:JC4VQAOO submitted 2025-07-31 cs.CV cs.AI

classification cs.CVcs.AI
keywords SAM2breastMRItumorsegmentationzero-shotvideoobjecttrackingfoundationmodelsvolumetricDiceequitableAI
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

This paper asks whether a general-purpose image and video segmentation model, SAM2, can segment breast tumors across a full 3D MRI volume when given only one annotated slice. The authors report that it can: a single bounding box propagated slice-by-slice yields a mean volumetric Dice of 0.57, and using a segmentation mask as the prompt raises the mean to 0.71, with no training or fine-tuning. They compare three propagation directions and find that starting at the central tumor slice and moving outward outperforms bottom-to-top and top-to-bottom propagation. The practical stake is that an open-source, zero-shot model could lower the cost barrier for 3D breast MRI interpretation in settings where commercial medical AI is unaffordable. The paper also identifies fragmented multi-lesion tumors as the main failure mode.

What carries the argument

The central mechanism is SAM2's tracking functionality, which takes the predicted mask from the previous frame as contextual guidance for the next frame. Applied slice-by-slice to axial MRI slices, this turns a single bounding-box or mask prompt into a chain of segmentations spanning the tumor volume. The paper's novel procedural choice is the center-outward traversal: initializing at the slice where the tumor is largest and clearest, then propagating both upward and downward, reduces long-range tracking errors compared with starting at an end slice.

What would settle it

Re-run the full pipeline on a subset of MAMA-MIA cases after independently verifying or re-registering the expert masks to the DCE-MRI slices; if mean volumetric Dice changes materially or the center-outward advantage over directional propagation disappears, the reported performance rests on an alignment artifact rather than on SAM2's tracking ability.

Watch

Extended reading notes

Core claim

The central discovery is that SAM2, a vision foundation model trained for image and video segmentation rather than for volumetric medical data, can be reused as a 3D tumor segmenter by treating a breast MRI as a stack of 2D frames and propagating a single-slice prompt across adjacent slices. On the MAMA-MIA expert-segmented subset of the Duke Breast Cancer Dataset, the center-outward propagation strategy achieves mean volumetric Dice of 0.57 with bounding-box prompts and 0.71 with mask prompts, and it wins for the majority of the 279 patients against the two directional strategies. The authors additionally find that tumor size, volume, and slice count show no meaningful correlation with Dice score, while multiple small scattered lesions produce poor segmentations.

Load-bearing premise

The evaluation assumes that the MAMA-MIA expert voxel segmentations are correct and are aligned with the pre-contrast DCE-MRI slices extracted using the procedure of refs. [5,12,14,15], and the paper does not independently validate this alignment or provide the extraction pipeline.

Editorial extensions

If this is right

  • If confirmed, a zero-shot SAM2 workflow can produce clinically plausible 3D tumor volumes from a single annotation, removing the need for large annotated training sets or task-specific fine-tuning.
  • Center-outward propagation should be the default traversal order for SAM2-based slice-wise medical tracking, since it maximizes per-patient Dice and consistency.
  • Using a segmentation mask rather than a bounding box as the initial prompt improves mean volumetric Dice from 0.57 to 0.71, so interactive annotators should prefer mask prompts whenever available.
  • The absence of correlation between Dice and tumor volume or slice count implies that failure is not simply a matter of tumor size; the identified multi-lesion failure mode suggests an explicit pre-selection or multi-prompt handling is needed for fragmented tumors.
  • Deployment cost in low-resource settings is lowered because the pipeline needs only one human annotation and open-source components.

Reading between the lines

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

  • Because the method is domain-agnostic in principle, the same single-box tracking recipe could be tested on other volumetric modalities (CT, ultrasound sweeps) and on longitudinal MRI where the 'frames' are time points rather than slices.
  • A natural next step is to replace the human-supplied box with an automatic detector; if the detector finds the tumor-bearing slice and a box, the whole loop becomes annotation-free, though the paper does not evaluate this.
  • The center-outward advantage may reflect a more general principle: seeding propagation at the most informative frame improves error containment in any video-object-tracking setting, which is testable outside medicine.
  • The reported Dice values give a lower-bound style estimate for what a zero-shot tracker can do on unseen volumetric data; a supervised 3D U-Net baseline on the same 279 patients would make the cost-performance trade-off concrete.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper proposes using SAM2, a zero-shot vision foundation model for image and video segmentation, to segment breast tumors in 3D MRI by treating the volume as a sequence of 2D slices. Starting from a single bounding-box prompt on one slice, the authors compare three slice-wise propagation orders (bottom-to-top, top-to-bottom, center-outward) on 279 patients from the MAMA-MIA/Duke breast MRI dataset and report mean volumetric Dice scores of 0.57 for the best bounding-box strategy and 0.71 when an initial segmentation mask is provided on the central slice. They also analyze correlations between Dice and tumor size, volume, and slice count, and identify multi-focal small lesions as a common failure mode.

Significance. The study addresses an accessibility-oriented question with practical relevance: whether an open-source, training-free video segmentation model can provide useful interactive 3D tumor segmentation from minimal annotations. Its strengths include the use of a public benchmark with expert voxel-level masks, the relatively large 279-patient evaluation cohort, and the zero-shot setting with no training or fine-tuning, which makes the pipeline easy to reproduce conceptually. The failure-mode analysis (scattered small lesions) is clinically plausible and actionable. However, the quantitative conclusions are currently conditional on unresolved protocol details, most importantly the scoring of the mask-prompt experiment, and the lack of baselines and statistical tests makes the performance claims difficult to interpret. If the authors address these points, the result would be a solid empirical contribution to the workshop literature on foundational models in medical imaging.

major comments (4)
  1. [Section 3, 'Comparison to Mask-based Tracking'; Eq. (1)] The manuscript does not state the source of the segmentation mask used as the SAM2 prompt in the mask-based tracking experiment. If this mask is the MAMA-MIA ground-truth mask from the central slice, as the described pipeline strongly suggests, then Eq. (1) scores that prompted slice as part of the volume; because the prompt exactly equals ground truth, that slice contributes a per-slice Dice of 1.0 by construction. Since the central slice is selected as the slice with the largest and clearest tumor, this single slice can inflate the volumetric Dice substantially. The reported improvement from 0.57 to 0.71 is therefore not a clean measure of propagation quality. The authors must disclose the mask source and either exclude the prompted slice from both P and G in Eq. (1) or report Dice separately for propagated slices only; the abstract and conclusion should then be updated if the corrected number changes.
  2. [Section 3, 'Quantitative Results' and Fig. 3] The claim that center-outward propagation outperforms the other two strategies is based on histograms, boxplots, and per-patient winner counts, but no statistical test or confidence interval is provided. Given 279 paired observations, a paired test (e.g., Wilcoxon signed-rank) and an effect size should be reported. Moreover, the center slice is chosen as the slice with the largest and clearest tumor, so the comparison confounds propagation direction with prompt difficulty: initializing from an easier slice may explain the advantage even if propagation itself is not better. Please quantify the difference and discuss the selection protocol.
  3. [Section 2.1] All Dice scores depend on the exact alignment between the extracted 2D pre-contrast DCE-MRI slices and the MAMA-MIA expert voxel masks, but the paper does not validate this alignment; it references the extraction procedure of prior work [5,12,14,15]. If the masks were generated on a different contrast phase or at a different resolution/spacing, the volumetric Dice values would be systematically biased. The authors should either provide evidence of alignment (e.g., visual audit, registration metrics, or original spacing information) or treat this as a limitation and test the sensitivity of the results to the extraction parameters.
  4. [Section 3 (overall)] The paper does not compare against any existing supervised or medical-image segmentation method (e.g., nnU-Net, UNETR, or a MedSAM variant) on the same 279-patient subset. Without such a baseline, the phrase 'strong segmentation performance' in the abstract and conclusion is not calibrated, and a reader cannot judge whether the proposed zero-shot workflow is competitive or merely convenient. Adding at least one reference baseline, or explicitly reframing the contribution as a zero-shot feasibility study without a performance claim, would make the claims precise.
minor comments (6)
  1. [Section 2.3 and Fig. 2] The rule for choosing the 'central slice' is not defined operationally; state whether it is the slice with maximum tumor area, the middle of the tumor-positive range, or a manual choice, and whether this uses ground-truth information.
  2. [Section 2.2 and experimental setup] Provide the exact SAM2 checkpoint, whether the video memory/offline mode was used, and any prompt post-processing (e.g., thresholding of logits), so the experiments are reproducible.
  3. [Section 2.1] State how the 279-patient subset was selected from the 922-patient Duke dataset and whether the exclusion affects the generalizability of the failure-mode analysis.
  4. [Section 3, 'Factors Affecting Dice Score'] R^2 values for linear fits are reported but no confidence intervals or correlation coefficients; a Spearman correlation with a confidence interval would be more informative than the linear-regression R^2 for these skewed distributions.
  5. [Abstract and Section 3] The abstract says 'single bounding box annotation on one slice', but the mask-based experiment in Section 3 requires a full mask as input; please restate the abstract and conclusion to distinguish the bounding-box workflow from the mask-prompt analysis.
  6. [Section 3, qualitative results] The supplementary material is referenced in Section 3 but is not included in the arXiv submission; please make it available or remove the reference.

Circularity Check

1 steps flagged · score 6.0 of 10

The mask-based tracking comparison leaks the ground-truth central-slice mask into the scored volumetric Dice, so the reported 0.71 improvement over 0.57 is partly by construction; the main bounding-box evaluation is otherwise externally benchmarked.

  1. self definitional [Section 2.1 and Section 3, 'Comparison to Mask-based Tracking', Eq. (1), Fig. 4]
    "we use the labeled images from MAMA-MIA dataset for 279 patients ... we rerun the center-outward propagation experiment by providing a segmentation mask on a single central slice as input ... the mean Dice score across all patients increasing from 0.57 (bounding box tracking) to 0.71 (segmentation mask tracking). Dice(P, G) = 2 · |P ∩ G| / (|P| + |G|) (Eq. 1)"

    The only mask source in the paper is the MAMA-MIA expert voxel segmentation. In the mask-tracking experiment, that mask on the central slice is given to SAM2 as the prompt, and Eq. (1) computes volumetric Dice over the full volume with no exclusion of the prompted slice. On the prompted central slice, the predicted volume P equals the ground-truth volume G by construction, so that slice contributes a per-slice Dice of exactly 1.0 regardless of SAM2's propagation ability. Since the central slice is selected as the largest and clearest tumor slice, its foreground voxel share is disproportionately large, materially inflating the reported mean from 0.57 to 0.71 and driving the claimed advantage of mask-based over bounding-box tracking.

full rationale

The main evaluation is externally benchmarked: SAM2 is prompted with a single bounding box on one slice, and the resulting volume is compared against MAMA-MIA expert segmentations that are not produced by SAM2 and not fitted in this paper. The center-outward versus bottom-to-top/top-to-bottom comparison is a legitimate empirical finding, and the slice-extraction procedure cited to prior work by the same authors (refs. [5,12]) is a preprocessing choice rather than a load-bearing derived result, so it does not by itself raise the circularity score. The significant circularity is confined to the mask-based tracking experiment: if, as the text implies, the 'segmentation mask on a single central slice' is the MAMA-MIA ground-truth mask, then that slice is both input and scored output in Eq. (1), forcing a per-slice Dice of 1.0 and partially manufacturing the reported 0.57-to-0.71 improvement. The paper neither states that the mask prompt is ground truth nor reports excluding the prompted slice from the Dice computation, and no alternative mask source is described. A clear disclosure or exclusion of the prompted slice would resolve this concern; as written, the mask-based result is partially circular.

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

The central claim depends on the correctness of the external benchmark annotations and the assumption that SAM2's natural-video tracking transfers to MRI slice sequences. The paper introduces no new entities or fitted parameters for the core evaluation.

assumptions (4)
  • domain assumption The expert voxel-level segmentations in the MAMA-MIA dataset are accurate and correctly aligned with the pre-contrast DCE-MRI volumes.
    Section 2.1 uses MAMA-MIA masks as ground truth without validation of alignment.
  • domain assumption The procedure for extracting 2D horizontal slices from the Duke/MAMA-MIA volumes, following refs. [5,12,14,15], preserves the 3D tumor structure needed for slice-wise propagation.
    Section 2.1 states the procedure is followed from prior work, but no details or validation are given.
  • domain assumption SAM2's video tracking mechanism can be applied to a sequence of MRI axial slices as if they were video frames, with coherent object identity across slices.
    Section 2.3 treats the volume as a sequence and relies on SAM2's tracking functionality, which was trained on natural videos.
  • domain assumption The central slice of the tumor-positive region can be identified in practice; in this study it is chosen as the starting slice.
    Section 2.3 states segmentation begins at the central slice, but the paper does not discuss how this is determined automatically in a real workflow.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Towards Affordable Tumor Segmentation and Visualization for 3D Breast MRI Using SAM2." pith.science (2026). https://pith.science/paper/JC4VQAOO

@misc{pith2026250723272,
  author       = {Pith},
  title        = {Pith review of: Towards Affordable Tumor Segmentation and Visualization for 3D Breast MRI Using SAM2},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JC4VQAOO}},
  note         = {Machine review of arXiv:2507.23272}
}
read the original abstract

Breast MRI provides high-resolution volumetric imaging critical for tumor assessment and treatment planning, yet manual interpretation of 3D scans remains labor-intensive and subjective. While AI-powered tools hold promise for accelerating medical image analysis, adoption of commercial medical AI products remains limited in low- and middle-income countries due to high license costs, proprietary software, and infrastructure demands. In this work, we investigate whether the Segment Anything Model 2 (SAM2) can be adapted for low-cost, minimal-input 3D tumor segmentation in breast MRI. Using a single bounding box annotation on one slice, we propagate segmentation predictions across the 3D volume using three different slice-wise tracking strategies: top-to-bottom, bottom-to-top, and center-outward. We evaluate these strategies across a large cohort of patients and find that center-outward propagation yields the most consistent and accurate segmentations. Despite being a zero-shot model not trained for volumetric medical data, SAM2 achieves strong segmentation performance under minimal supervision. We further analyze how segmentation performance relates to tumor size, location, and shape, identifying key failure modes. Our results suggest that general-purpose foundation models such as SAM2 can support 3D medical image analysis with minimal supervision, offering an accessible and affordable alternative for resource-constrained settings.

Figures

Figures reproduced from arXiv: 2507.23272 by the authors.

Figure 1
Figure 1. (a) A side profile diagram of the breast, highlighting the imaging region [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. (a) A side profile diagram of the breast, highlighting tumor-positive slices [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Evaluation of 3D propagation strategies. (a) Dice score histograms per [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Evaluation of tumor tracking methods for SAM2. The histogram (left) [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Sequential MRI slices for (a) Patient 350 and (b) Patient 400. The left [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Scatter plots showing relationships between Dice score and tumor prop [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: Example cases where SAM2 yields low Dice scores due to the presence of [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

26 extracted references · 17 canonical work pages

  1. [1]

    BMC medical education23(1), 689 (2023)

    Alowais, S.A., Alghamdi, S.S., Alsuhebany, N., Alqahtani, T., Alshaya, A.I., Al- mohareb, S.N., Aldairem, A., Alrashed, M., Bin Saleh, K., Badreldin, H.A., et al.: Revolutionizing healthcare: the role of artificial intelligence in clinical practice. BMC medical education23(1), 689 (2023)

  2. [2]

    arXiv preprint arXiv:2307.13721 (2023)

    Awais, M., Naseer, M., Khan, S., Anwer, R.M., Cholakkal, H., Shah, M., Yang, M.H., Khan, F.S.: Foundational models defining a new era in vision: A survey and outlook. arXiv preprint arXiv:2307.13721 (2023)

  3. [3]

    arXiv preprint arXiv:2102.04306 (2021)

    Chen, J., Lu, Y., Yu, Q., et al.: Transunet: Transformers make strong encoders for medical image segmentation. arXiv preprint arXiv:2102.04306 (2021)

  4. [4]

    In: European Conference on Computer Vision

    Cheng, H.K., Schwing, A.G.: Xmem: Long-term video object segmentation with an atkinson-shiffrin memory model. In: European Conference on Computer Vision. pp. 640–658. Springer (2022)

  5. [5]

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

    Chung, M., Won, J.B., Kim, G., Kim, Y., Ozbulak, U.: Evaluating visual explana- tions of attention maps for transformer-based medical imaging. In: International Conference on Medical Image Computing and Computer-Assisted Intervention. pp. 110–120. Springer (2024)

  6. [6]

    British journal of cancer 126(4), 533–550 (2022)

    Clift, A.K., Dodwell, D., Lord, S., Petrou, S., Brady, S.M., Collins, G.S., Hippisley- Cox, J.: The current status of risk-stratified breast screening. British journal of cancer 126(4), 533–550 (2022)

  7. [7]

    Public Health Nursing 42(2), 1017–1030 (2025)

    Dangi, R.R., Sharma, A., Vageriya, V.: Transforming healthcare in low-resource settings with artificial intelligence: Recent developments and outcomes. Public Health Nursing 42(2), 1017–1030 (2025)

  8. [8]

    Scientific data 12(1), 453 (2025) 10 Kang et al

    Garrucho, L., Kushibar, K., Reidel, C.A., Joshi, S., Osuala, R., Tsirikoglou, A., Bobowicz, M., Del Riego, J., Catanese, A., Gwoździewicz, K., et al.: A large-scale multicenter breast cancer dce-mri benchmark dataset with expert segmentations. Scientific data 12(1), 453 (2025) 10 Kang et al

Show all 26 references
  1. [9]

    AI Open2, 225–250 (2021)

    Han, X., Zhang, Z., Ding, N., Gu, Y., Liu, X., Huo, Y., Qiu, J., Yao, Y., Zhang, A., Zhang, L., et al.: Pre-trained models: Past, present and future. AI Open2, 225–250 (2021)

  2. [10]

    Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) pp

    Hatamizadeh, A., Yin, H., Molchanov, P., et al.: Unetr: Transformers for 3d med- ical image segmentation. Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) pp. 574–584 (2022)

  3. [11]

    Nature Methods 18(2), 203–211 (2021)

    Isensee, F., Jaeger, P.F., Kohl, S.A.A., Petersen, J.P., Maier-Hein, K.H.: nnu- net: Self-adapting framework for u-net-based medical image segmentation. Nature Methods 18(2), 203–211 (2021)

  4. [12]

    In: Deep Breast Workshop on AI and Imaging for Diagnostic and Treatment Chal- lenges in Breast Care

    Kang, S., De Neve, W., Rameau, F., Ozbulak, U.: Exploring patient data require- ments in training effective ai models for mri-based breast cancer classification. In: Deep Breast Workshop on AI and Imaging for Diagnostic and Treatment Chal- lenges in Breast Care. pp. 75–84. Spr...

  5. [13]

    BMC medicine17, 1–9 (2019)

    Kelly, C.J., Karthikesalingam, A., Suleyman, M., Corrado, G., King, D.: Key chal- lenges for delivering clinical impact with artificial intelligence. BMC medicine17, 1–9 (2019)

  6. [14]

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

    Konz, N., Gu, H., Dong, H., Mazurowski, M.A.: The intrinsic manifolds of radio- logical images and their role in deep learning. In: International Conference on Med- ical Image Computing and Computer-Assisted Intervention. pp. 684–694. Springer (2022)

  7. [15]

    arXiv preprint arXiv:2401.08865 (2024)

    Konz, N., Mazurowski, M.A.: The effect of intrinsic dataset properties on general- ization: Unraveling learning differences between natural and medical images. arXiv preprint arXiv:2401.08865 (2024)

  8. [16]

    Cancers13(17), 4287 (2021)

    Łukasiewicz, S., Czeczelewski, M., Forma, A., Baj, J., Sitarz, R., Stanisławek, A.: Breast cancer—epidemiology, risk factors, classification, prognostic markers, and current treatment strategies—an updated review. Cancers13(17), 4287 (2021)

  9. [17]

    arXiv preprint arXiv:2504.03600 (2025)

    Ma, J., Yang, Z., Kim, S., Chen, B., Baharoon, M., Fallahpour, A., Asakereh, R., Lyu, H., Wang, B.: Medsam2: Segment anything in 3d medical images and videos. arXiv preprint arXiv:2504.03600 (2025)

  10. [18]

    Mango, V.L., Morris, E.A., Dershaw, D.D., Abramson, A., Fry, C., Moskowitz, C.S., Hughes, M., Kaplan, J., Jochelson, M.S.: Abbreviated protocol for breast mri: are multiple sequences needed for cancer detection? European journal of radiology 84(1), 65–70 (2015)

  11. [19]

    European journal of breast health15(1), 1 (2019)

    Mootz, A.R., Madhuranthakam, A.J., Doğan, B.: Changing paradigms in breast cancer screening: abbreviated breast mri. European journal of breast health15(1), 1 (2019)

  12. [20]

    inconvenient truth

    Panch, T., Mattie, H., Celi, L.A.: The “inconvenient truth” about ai in healthcare. NPJ digital medicine2(1), 1–3 (2019)

  13. [21]

    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)

  14. [22]

    British journal of cancer 119(4), 508–516 (2018)

    Saha, A., Harowicz, M.R., Grimm, L.J., Kim, C.E., Ghate, S.V., Walsh, R., Mazurowski, M.A.: A machine learning approach to radiogenomics of breast can- cer: a study of 922 subjects and 529 dce-mri features. British journal of cancer 119(4), 508–516 (2018)

  15. [23]

    CA: a cancer journal for clinicians74(1), 12–49 (2024)

    Siegel, R.L., Giaquinto, A.N., Jemal, A.: Cancer statistics, 2024. CA: a cancer journal for clinicians74(1), 12–49 (2024)

  16. [24]

    IEEE Journal of Biomedical and Health Informatics (2024) Affordable Tumor Segmentation and Visualization for 3D Breast MRI 11

    Zhang, Z., Han, L., Zhang, T., Lin, Z., Gao, Q., Tong, T., Sun, Y., Tan, T.: Unimrisegnet: Universal 3d network for various organs and cancers segmentation on multi-sequence mri. IEEE Journal of Biomedical and Health Informatics (2024) Affordable Tumor Segmentation and Visuali...

  17. [25]

    International Journal of Machine Learning and Cybernetics pp

    Zhou, C., Li, Q., Li, C., Yu, J., Liu, Y., Wang, G., Zhang, K., Ji, C., Yan, Q., He, L., et al.: A comprehensive survey on pretrained foundation models: A history from bert to chatgpt. International Journal of Machine Learning and Cybernetics pp. 1–65 (2024)

  18. [26]

    arXiv preprint arXiv:2408.00874 (2024)

    Zhu, J., Hamdi, A., Qi, Y., Jin, Y., Wu, J.: Medical sam 2: Segment medical images as video via segment anything model 2. arXiv preprint arXiv:2408.00874 (2024)

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.