Pith. sign in

REVIEW 3 major objections 6 minor 62 references

A review argues that countries can build their own medical foundation models using pooled national data and supercomputing resources.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review

2026-08-04 22:49 UTC pith:ORWLAZ2H

load-bearing objection A competent survey with a policy proposal that is asserted rather than argued; useful as an introduction, not as a research result. the 3 major comments →

arxiv 2608.01686 v1 pith:ORWLAZ2H submitted 2026-08-03 cs.CV

Generative AI and Foundation Models in Medical Image

Masahiro Oda This is my paper
classification cs.CV
keywords generative AIdiffusion modelslarge language modelsfoundation modelsmedical image processingscaling lawsdata augmentationnational medical AI
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

This paper is a survey and a roadmap. It explains how diffusion models, large language models, and foundation models are reshaping medical image analysis, and it argues that these technologies are not necessarily the exclusive province of large technology companies. The central claim is that by pooling national medical imaging data and tapping national supercomputing resources, a country can develop its own foundation models and high-performance diagnostic-support AI, tailored to its own patient population. The stakes are practical: if the claim holds, countries without access to big-tech-scale commercial cloud compute could still build competitive medical AI, and AI tools could be adapted to local demographics rather than imported from elsewhere. The argument rests on scaling laws, the empirical pattern that model performance rises with more training data, parameters, and compute.

Core claim

On its own terms, the central discovery is a feasibility thesis rather than a new algorithm or dataset. The same scaling-law behavior that drove large language models to high performance is treated as applying to vision models and therefore to medical imaging, and the paper assembles evidence that the two ingredients the scaling laws demand, large training corpora and large compute, can be obtained outside big technology companies. The evidence includes existing medical image repositories storing hundreds of millions of images, an annotation workflow that labeled roughly 3.2 million CT slices in three weeks through human-in-the-loop AI correction, and simulation and diffusion-based generatio

What carries the argument

The load-bearing mechanism is the Scaling Laws, the empirically observed relationship in which neural-model performance rises with training-data volume, model parameters, and training compute. The paper takes this relationship from language models and vision transformers and treats it as the justification for scaling up medical models. The second carrying mechanism is the foundation-model paradigm: a large model pre-trained with self-supervised learning on broad, unlabeled data, then adapted to downstream tasks through prompts, few-shot learning, or fine-tuning. A third, enabling mechanism is supply-side infrastructure: human-in-the-loop annotation with model-disagreement triage to label lar

Load-bearing premise

The paper assumes that the scaling laws documented for natural-language and natural-image models also hold for medical images, and that pooled national data plus supercomputers are large enough to push performance to clinically useful levels; if either link breaks, the national-roadmap conclusion does not follow.

What would settle it

A controlled scaling study on a standardized medical-imaging benchmark would settle the central claim: pre-train models at increasing data, parameter, and compute budgets and plot downstream segmentation or detection accuracy. If the curves plateau well before clinically usable accuracy, or if doubling compute and data yields negligible gains for medical modalities, the extrapolation from language-model scaling laws is falsified. A second falsifier would be an attempt to assemble and annotate a national-scale dataset using the described human-in-the-loop and simulation pipeline failing to prod

Watch this falsifier. Get emailed when new claim-graph text bears on it.

Share X Bluesky LinkedIn Reddit HN

If this is right

  • A country with a large medical-image repository and supercomputer access could plausibly produce a medical foundation model without depending on a large technology company.
  • The efficient annotation workflow described in the paper implies that a clinically usable dataset of millions of labeled slices can be assembled in weeks rather than years.
  • Synthetic and simulated images can relieve the data bottleneck, making rare-disease and low-prevalence imaging tasks addressable with AI trained partly on generated examples.
  • Diagnostic AI development would shift from building a niche model per task to adapting a general medical foundation model, lowering the cost of entering new clinical tasks.
  • Medical foundation models built on national data would carry population-specific characteristics, potentially reducing country- or race-related bias compared with models trained on foreign datasets.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: if scaling laws for medical images behave like scaling laws for text, the binding constraint will shift from compute and data volume to data quality, diversity, and consent; countries with large but homogeneous datasets may need deliberate data harmonization rather than simply more images.
  • Editorial inference: the same national-resource playbook could plausibly extend to multimodal medical AI that combines images, text reports, laboratory values, and genomics, since the foundation-model machinery is modality-agnostic, but this paper explicitly stays within imaging.
  • Editorial inference: a direct test of the thesis would be to pre-train a medical foundation model at the described scale and measure whether downstream clinical-task gains track the scaling-law curve; until that is done, the extrapolation from language models remains a hypothesis.
  • Editorial inference: clinical deployment would likely require evaluation against patient outcomes and workflow usability, not just segmentation accuracy, before the claimed high-performance AI systems materialize.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 6 minor

Summary. This manuscript is a broad survey of generative AI and foundation models in medical image processing, covering diffusion models, large language models, and medical foundation models, and it concludes with a proposal to develop nationally built medical foundation models by pooling national medical datasets and supercomputing resources. The survey portions provide a useful, clearly organized overview of recent methods and applications, including medical image generation, diffusion-based segmentation, clinical LLMs, and medical vision-language models. The paper's distinctive contribution is the roadmap in Section 6: it argues that the Scaling Laws imply that large data, model, and compute are sufficient to realize high-performance medical AI, and it asserts that the NII platform's 400 million images and national supercomputers make this feasible. The paper does not present new models, experiments, or quantitative analyses; it is a narrative/position paper whose central claim is the feasibility of the national-scale roadmap.

Significance. If the feasibility claim were quantitatively established, the paper would be a valuable roadmap for academic and national initiatives aiming to build medical foundation models outside large commercial labs. The survey is competently assembled and covers the relevant literature with an extensive reference list, and it is honest about domain-gap limitations (e.g., SAM underperforming on medical images in Section 5.1). Strengths include the concrete enumeration of data resources (NII platform, AbdomenAtlas-8K, XCAT, simulation-based data) and the identification of a policy/economic pathway through subsidized supercomputing access. However, the paper ships no code, no dataset, and no quantitative model, which is acceptable for a survey, but it means the roadmap's central assertion must carry its own burden of evidence. As written, that assertion is plausible but unsupported, and the paper's significance depends on closing this gap.

major comments (3)
  1. [§6.1, §6.3, Conclusions] The central feasibility claim—that pooling national medical data and supercomputer time suffices to train competitive foundation models—is not quantitatively supported. §6.1 recalls the Scaling Laws (three factors), and §6.3 concludes that 'it is entirely feasible to develop large-scale AI models,' but no calculation links the cited resources (400M images on the NII platform; unspecified GPU counts) to the parameter/token/compute budget implied by scaling laws. The transfer of language/natural-image scaling laws to 3D/4D medical images is assumed, despite the paper's own note in §5.1 that SAM underperforms task-specific medical models. Provide a concrete budget (e.g., FLOPs, effective token-equivalents, number of volumes), compare with known medical foundation models, and state where the proposed resources fall relative to projected scaling curves.
  2. [§6.2, §6.3] The assumption 'Assuming the implementation of large-scale models is possible through software development' (last paragraph of §6.1) offloads the hardest part. Training at this scale requires cluster orchestration, fault tolerance, memory-efficiency engineering, and stable distributed data pipelines; §6.3 only notes that supercomputers 'support general development environments, including Python.' Similarly, the 400M-image platform is an archive, not a cleaned, deduplicated, de-identified, and task-annotated corpus; the AbdomenAtlas-8K annotation project covers only one modality and body region. These gaps make the feasibility assertion an unsupported premise rather than a demonstrated outcome. Please add a risk/engineering-work assessment and a data-governance plan.
  3. [§6.2] The evidence offered for scaling annotation and data generation is not sufficient. The human-in-the-loop annotation of 8,448 CT volumes 'in three weeks' is impressive, but no person-hours, expert workload, or quality metrics are reported, and it is a single-organ segmentation pipeline. Synthetic data (XCAT, vascular simulation) is useful but restricted to certain modalities; the sim-to-real gap is acknowledged only implicitly. Without a cost model and quality bounds, the statement that 'large-scale medical image datasets ... are realistically available' (§6.3) is an assertion, not a demonstrated result. Please add quantitative evidence or explicitly reframe the claim as a research aspiration rather than a feasibility conclusion.
minor comments (6)
  1. [Fig. 1] The caption appears to be 'Conceptual diagram of task-specific AI development using foundation models,' but the text (§3.2) says Figure 1 illustrates the diffusion and reverse diffusion processes. The caption is likely mismatched with the figure; correct it or move the figure.
  2. [§3.1, §4.1, Tables 1–3] Typos and formatting: 'T able' appears in Tables 1–3; 'V AEs' in §3.1; 'human-likely' in §4.1 should be 'human-like.'
  3. [References] References [2] and [23] are the same paper (Vaswani et al., 'Attention is all you need'); consolidate to avoid duplicate entries.
  4. [§6.2] The description of NII's 'Medical Image Big Data Cloud Platform' (400M images, since 2017) has no citation; add a source or footnote so readers can verify the claim.
  5. [Tables 1–3] The tables use '–' inconsistently for undisclosed values, and entries such as GPT-3.5's 355B parameter count are not sourced in the text. A consistent notation and per-entry references would improve verifiability.
  6. [§5.1] The sentence 'This principle applies not only to language models but also to image processing models' is too broad. The cited Zhai et al. study reports scaling behavior for ViT-based models, so the text should say 'has been reported for ViT-based image models.'

Circularity Check

0 steps flagged

No circularity found: the paper is an overview that makes no fitted predictions and contains no self-cited load-bearing derivation.

full rationale

This manuscript is a narrative review and position paper. It does not present a new model, a derivation, a fitted parameter, or a quantitative prediction that could reduce to its inputs. The central feasibility claim in Section 6.3 ('By maximizing the use of national data and computing resources, we can anticipate the emergence of nationally developed foundation models...') is an extrapolation from external scaling-law results, not a result derived from any equation in the paper. The scaling laws cited (Kaplan et al. [25] and Zhai et al. [33]) are external, peer-reviewed/arXiv results, and the paper does not claim to derive them or fit them to medical data. The discussion of medical datasets (NII platform, AbdomenAtlas-8K, synthetic data) and supercomputers is presented as evidence of availability, not as a calculation that proves sufficiency. There are no self-citations by the author, no imported uniqueness theorems, and no ansatz smuggled via citation. The skeptical concern that scaling laws may not transfer to medical imaging is a correctness or validity concern, not a circularity concern, and the paper itself acknowledges domain gaps (e.g., SAM underperforms task-specific models on medical images). Therefore, no circular step is present, and the score is 0.

Axiom & Free-Parameter Ledger

0 free parameters · 4 axioms · 0 invented entities

The paper's roadmap rests on assumptions about scaling laws, dataset sufficiency, annotation feasibility, and compute access. None of these are demonstrated in the paper, so the central claim relies on unverified premises.

axioms (4)
  • domain assumption Scaling Laws for language models apply to medical image models.
    Section 6.1 invokes the Scaling Laws to argue that increasing data, model size, and compute will improve medical AI performance, without evidence from medical imaging.
  • domain assumption National medical image datasets, e.g., NII Medical Image Big Data Cloud Platform, can support foundation model training.
    Section 6.2 relies on this dataset's existence and size, but includes no analysis of its diversity, label quality, or suitability for pretraining.
  • domain assumption Human-in-the-loop annotation and synthetic data can produce enough annotated data at feasible cost.
    Section 6.2 cites prior work and assumes these methods scale to the national level, with no cost or quality assessment.
  • domain assumption Academic researchers can access national supercomputers with sufficient GPU capacity at feasible cost.
    Section 6.3 argues that usage fees can be subsidized by support programs, but provides no evidence of capacity or cost.

reviewed 2026-08-04 · how reviews work

0 comments
Cite this review

Pith. "Pith review of Generative AI and Foundation Models in Medical Image." pith.science (2026). https://pith.science/paper/ORWLAZ2H

@misc{pith2026260801686,
  author       = {Pith},
  title        = {Pith review of: Generative AI and Foundation Models in Medical Image},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ORWLAZ2H}},
  note         = {Machine review of arXiv:2608.01686}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

In recent years, generative AI has attracted significant public attention, and its use has been rapidly expanding across a wide range of domains. From creative tasks such as text summarization, idea generation, and source code generation, to the streamlining of medical support tasks like diagnostic report generation and summarization, AI is now deeply involved in many areas. Today's breadth of AI applications is clearly distinct from what was seen before generative AI gained widespread recognition. Representative generative AI services include DALL-E 3 (OpenAI, California, USA) and Stable Diffusion (Stability AI, London, England, UK) for image generation, ChatGPT (OpenAI, California, USA), and Gemini (Google, California, USA) for text generation. The rise of generative AI has been influenced by advances in deep learning models and the scaling up of data, models, and computational resources based on the scaling laws. Moreover, the emergence of foundation models, which are trained on large-scale datasets and possess general-purpose knowledge applicable to various downstream tasks, is creating a new paradigm in AI development. These shifts brought about by generative AI and foundation models also profoundly impact medical image processing, fundamentally changing the framework for AI development in healthcare. This paper provides an overview of diffusion models used in image generation AI and large language models (LLMs) used in text generation AI, and introduces their applications in medical support. This paper also discusses foundation models, which are gaining attention alongside generative AI, including their construction methods and applications in the medical field. Finally, the paper explores how to develop foundation models and high-performance AI for medical support by fully utilizing national data and computational resources.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

62 extracted references · 62 canonical work pages

  1. [1]

    ImageNet Classification with Deep Convo- lutional Neural Networks

    Krizhevsky A, Sutskever I, Hinton GE. ImageNet Classification with Deep Convo- lutional Neural Networks. Proceedings of the 26th International Conference on Neural Information Processing Systems (NIPS 12). 2012;1:1097–1105

    work page 2012

  2. [2]

    Attention Is All You Need

    Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, et al. Attention Is All You Need. arXiv:170603762. 2017

    work page 2017

  3. [3]

    Auto-Encoding Variational Bayes

    Kingma DP, Welling M. Auto-Encoding Variational Bayes. International Conference on Learning Representations (ICLR) 2014. 2014

    work page 2014

  4. [4]

    Generative adversarial networks

    Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, et al. Generative adversarial networks. Communications of the ACM. 2020;63(11):139–144

    work page 2020

  5. [5]

    Analyzing and Improv- ing the Image Quality of StyleGAN

    Karras T, Laine S, Aittala M, Hellsten J, Lehtinen J, Aila T. Analyzing and Improv- ing the Image Quality of StyleGAN. 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2020;p. 8107–8116. 12

    work page 2020

  6. [6]

    Deep unsupervised learning using nonequilibrium thermodynamics

    Sohl-Dickstein J, Weiss EA, Mah- eswaranathan N, Ganguli S. Deep unsupervised learning using nonequilibrium thermodynamics. Proceedings of the 32nd International Conference on International Conference on Machine Learning (ICML 15). 2015;37:2256–2265

    work page 2015

  7. [7]

    Denoising Diffusion Probabilistic Models

    Ho J, Jain A, Abbeel P. Denoising Diffusion Probabilistic Models. Proceedings of the 34th International Conference on Neural Informa- tion Processing Systems (NIPS 20). 2020;p. 6840–6851

    work page 2020

  8. [8]

    U- Net: Convolutional Networks for Biomedical Image Segmentation

    Ronneberger O, Fischer P, Brox T. U- Net: Convolutional Networks for Biomedical Image Segmentation. Medical Image Com- puting and Computer Assisted Intervention (MICCAI 2015). 2015;9351:234–241

    work page 2015

  9. [9]

    Brain Imaging Generation with Latent Diffusion Models

    Pinaya WHL, Tudosiu PD, Dafflon J, Costa PFD, Fernandez V, Nachev P, et al. Brain Imaging Generation with Latent Diffusion Models. DGM4MICCAI 2022. 2022;13609:117–126

    work page 2022

  10. [10]

    Three- Dimensional Medical Image Synthesis with Denoising Diffusion Probabilistic Models

    Dorjsembe Z, Odonchimed S, Xiao F. Three- Dimensional Medical Image Synthesis with Denoising Diffusion Probabilistic Models. MIDL 2022. 2022

    work page 2022

  11. [11]

    Diffusion Deformable Model for 4D Temporal Medical Image Genera- tion

    Kim B, Ye JC. Diffusion Deformable Model for 4D Temporal Medical Image Genera- tion. Medical Image Computing and Com- puter Assisted Intervention (MICCAI 2022). 2022;13431:539–548

    work page 2022

  12. [12]

    Image synthesis with class-aware seman- tic diffusion models for surgical scene seg- mentation

    Zhou Y, Towning R, Awad Z, Giannarou S. Image synthesis with class-aware seman- tic diffusion models for surgical scene seg- mentation. Healthcare Technology Letters. 2025;12(1)

    work page 2025

  13. [13]

    Synthesizing Multi-class Surgical Datasets with Anatomy-Aware Dif- fusion Models

    Venkatesh DK, Rivoir D, Pfeiffer M, Kol- binger F, Speidel S. Synthesizing Multi-class Surgical Datasets with Anatomy-Aware Dif- fusion Models. arXiv:241007753. 2024

    work page 2024

  14. [14]

    mixup: Beyond Empirical Risk Minimiza- tion

    Zhang H, Cisse M, Dauphin YN, Lopez-Paz D. mixup: Beyond Empirical Risk Minimiza- tion. International Conference on Learning Representations (ICLR) 2018. 2018

    work page 2018

  15. [15]

    Artificial Case Image Generation of Breast Cancer Mass by Stable Diffusion and Its Application to Dif- ferentiation between Benign and Malignant

    Abe K, Takeo H, Nawano S. Artificial Case Image Generation of Breast Cancer Mass by Stable Diffusion and Its Application to Dif- ferentiation between Benign and Malignant. Proceedings of JAMIT 2023. 2023;p. 142–148

    work page 2023

  16. [16]

    MedSegDiff: Medical Image Segmentation with Diffusion Proba- bilistic Model

    Junde Wu RF, Fang H, Zhang Y, Yang Y, Xiong H, Liu H, et al. MedSegDiff: Medical Image Segmentation with Diffusion Proba- bilistic Model. Medical Imaging with Deep Learning. 2024;227:1623–1639

    work page 2024

  17. [17]

    MedSegDiff-V2: diffusion-based medical image segmentation with trans- former

    Wu J, Ji W, Fu H, Xu M, Jin Y, Xu Y. MedSegDiff-V2: diffusion-based medical image segmentation with trans- former. Proceedings of the Thirty-Eighth AAAI Conference on Artificial Intelligence and Thirty-Sixth Conference on Innova- tive Applications of Artificial Intelligence and Fourteenth Symposium on Educa- tional Advances in Artificial Intelligence (AAA...

    work page 2024

  18. [18]

    Diffusion Trans- former U-Net for Medical Image Segmenta- tion

    Chowdary GJ, Yin Z. Diffusion Trans- former U-Net for Medical Image Segmenta- tion. Medical Image Computing and Com- puter Assisted Intervention (MICCAI 2023). 2023;14223:622–631

    work page 2023

  19. [19]

    BerDiff: Condi- tional Bernoulli Diffusion Model for Medical Image Segmentation

    Chen T, Wang C, Shan H. BerDiff: Condi- tional Bernoulli Diffusion Model for Medical Image Segmentation. Medical Image Com- puting and Computer Assisted Intervention (MICCAI 2023). 2023;14223:491–501

    work page 2023

  20. [20]

    Enhancing Label-Efficient Medical Image Segmentation with Text-Guided Diffu- sion Models

    Feng CM. Enhancing Label-Efficient Medical Image Segmentation with Text-Guided Diffu- sion Models. Medical Image Computing and Computer Assisted Intervention (MICCAI 2024). 2024;15008:253–262

    work page 2024

  21. [21]

    Diff-VPS: Video Polyp Segmentation via a Multi-task Diffusion Network with Adversar- ial Temporal Reasoning

    Lu Y, Yang Y, Xing Z, Wang Q, Zhu L. Diff-VPS: Video Polyp Segmentation via a Multi-task Diffusion Network with Adversar- ial Temporal Reasoning. Medical Image Com- puting and Computer Assisted Intervention (MICCAI 2024). 2024;15006:165–175

    work page 2024

  22. [22]

    Long Short-Term Memory

    Hochreiter S, Schmidhuber J. Long Short-Term Memory. Neural Computation. 1997;9(8):1735–1780. 13

    work page 1997

  23. [23]

    Attention is all you need

    Vaswani A, Shazeer N, Parmar N, Uszkor- eit J, Jones L, Gomez AN, et al. Attention is all you need. Proceedings of the 31st International Conference on Neural Informa- tion Processing Systems (NIPS 17). 2017;p. 6000–6010

    work page 2017

  24. [24]

    An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale

    Dosovitskiy A, Beyer L, Kolesnikov A, Weis- senborn D, Zhai X, Unterthiner T, et al. An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale. ICLR 2021. 2021

    work page 2021

  25. [25]

    Scaling Laws for Neural Language Models

    Kaplan J, McCandlish S, Henighan T, Brown TB, Chess B, Child R, et al. Scaling Laws for Neural Language Models. arXiv:200108361. 2020

    work page 2020

  26. [26]

    Available from: https://openai.com/index/ language-unsupervised/

    Radford A, Narasimhan K, Salimans T, Sutskever I.: Improving language under- standing with unsupervised learning. Available from: https://openai.com/index/ language-unsupervised/

    work page

  27. [27]

    The Car- bon Footprint of Machine Learning Train- ing Will Plateau, Then Shrink

    Patterson D, Gonzalez J, H¨ olzle U, Le Q, Liang C, Munguia LM, et al. The Car- bon Footprint of Machine Learning Train- ing Will Plateau, Then Shrink. Computer. 2022;55(7):18–28

    work page 2022

  28. [28]

    BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

    Devlin J, Chang MW, Lee K, Toutanova K. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. NAACL-HLT 2019. 2019;1:4171–4186

    work page 2019

  29. [29]

    A Survey of Large Language Mod- els in Medicine: Progress, Application, and Challenge

    Zhou H, Liu F, Gu B, Zou X, Huang J, Wu J, et al. A Survey of Large Language Mod- els in Medicine: Progress, Application, and Challenge. arXiv:231105112. 2024

    work page 2024

  30. [30]

    MammoScreen.: Available from: https:// www.mammoscreen.com/

    work page

  31. [31]

    radpair.com/

    RADPAIR.: Available from: https://www. radpair.com/

    work page

  32. [32]

    Interactive computer-aided diagnosis on medical image using large language mod- els

    Wang S, Zhao Z, Ouyang X, Liu T, Wang Q, Shen D. Interactive computer-aided diagnosis on medical image using large language mod- els. Communications Engineering. 2024;3

    work page 2024

  33. [33]

    Scaling Vision Transformers

    Zhai X, Kolesnikov A, Houlsby N, Beyer L. Scaling Vision Transformers. 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2022;p. 1204–1213

    work page 2022

  34. [34]

    Segment Any- thing

    Kirillov A, Mintun E, Ravi N, Mao H, Rol- land C, Gustafson L, et al. Segment Any- thing. arXiv:230402643. 2023

    work page 2023

  35. [35]

    Flamingo: a visual language model for few-shot learning

    Alayrac JB, Donahue J, Luc P, Miech A, Barr I, Hasson Y, et al. Flamingo: a visual language model for few-shot learning. Pro- ceedings of the 36th International Conference on Neural Information Processing Systems (NIPS 22). 2022;p. 23716–23736

    work page 2022

  36. [36]

    SAM 2: Segment Anything in Images and Videos

    Ravi N, Gabeur V, Hu YT, Hu R, Ryali C, Ma T, et al. SAM 2: Segment Anything in Images and Videos. arXiv:240800714. 2024

    work page 2024

  37. [37]

    Segment Anything Model (SAM) for Digital Pathology: Assess Zero- shot Segmentation on Whole Slide Imaging

    Deng R, Cui C, Liu Q, Yao T, Remedios L W, Bao S, et al. Segment Anything Model (SAM) for Digital Pathology: Assess Zero- shot Segmentation on Whole Slide Imaging. arXiv:230404155. 2023

    work page 2023

  38. [38]

    The ability of Segmenting Anything Model (SAM) to segment ultrasound images

    Chen F, Chen L, Han H, Zhang S, Zhang D, Liao H. The ability of Segmenting Anything Model (SAM) to segment ultrasound images. BioScience Trends. 2023;17(3):211–218

    work page 2023

  39. [39]

    Segment anything in medical images

    Ma J, He Y, Li F, Han L, You C, Wang B. Segment anything in medical images. Nature Communications. 2024;15

    work page 2024

  40. [40]

    Integrating vision and language: revo- lutionary foundations in medical imaging AI

    Li S. Integrating vision and language: revo- lutionary foundations in medical imaging AI. Proceedings of SPIE Medical Imaging 2024. 2024;12926

    work page 2024

  41. [41]

    Self-Supervised Visual Feature Learning With Deep Neural Net- works: A Survey

    Jing L, Tian Y. Self-Supervised Visual Feature Learning With Deep Neural Net- works: A Survey. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2020;43(11):4037–4058

    work page 2020

  42. [42]

    A simple framework for contrastive learning of visual representations

    Chen T, Kornblith S, Norouzi M, Hinton G. A simple framework for contrastive learning of visual representations. Proceedings of the 37th International Conference on Machine Learning (ICML 20). 2020;p. 1597–1607. 14

    work page 2020

  43. [43]

    Momentum Contrast for Unsupervised Visual Representation Learning

    He K, Fan H, Wu Y, Xie S, Girshick R. Momentum Contrast for Unsupervised Visual Representation Learning. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2020;p. 9729–9738

    work page 2020

  44. [44]

    Exploring Simple Siamese Representation Learning

    Chen X, He K. Exploring Simple Siamese Representation Learning. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2021;p. 15750–15758

    work page 2021

  45. [45]

    Emerging Prop- erties in Self-Supervised Vision Transformers

    Caron M, Mairal J, Touvron H, Misra I, Bojanowski P, J´ egou H, et al. Emerging Prop- erties in Self-Supervised Vision Transformers. Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV). 2021;p. 9650–9660

    work page 2021

  46. [46]

    Big Self-Supervised Models are Strong Semi-Supervised Learners

    Chen T, Kornblith S, Swersky K, Norouzi M, Hinton G. Big Self-Supervised Models are Strong Semi-Supervised Learners. Pro- ceedings of the 34th International Conference on Neural Information Processing Systems (NIPS 20). 2020;p. 22243–22255

    work page 2020

  47. [47]

    Improved Baselines with Momentum Con- trastive Learning

    Chen X, Fan H, Girshick R, He K. Improved Baselines with Momentum Con- trastive Learning. arXiv:200304297. 2020

    work page 2020

  48. [48]

    Learning Transfer- able Visual Models From Natural Language Supervision

    Radford A, Kim JW, Hallacy C, Ramesh A, Goh G, Agarwal S, et al. Learning Transfer- able Visual Models From Natural Language Supervision. Proceedings of the 38th Inter- national Conference on Machine Learning. 2021;p. 8748–8763

    work page 2021

  49. [49]

    Masked Autoencoders Are Scalable Vision Learners

    He K, Chen X, Xie S, Li Y, Doll´ ar P, Gir- shick R. Masked Autoencoders Are Scalable Vision Learners. 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2022

    work page 2022

  50. [50]

    What to Hide from Your Students: Attention- Guided Masked Image Modeling

    Kakogeorgiou I, Gidaris S, Psomas B, Avrithis Y, Bursuc A, Karantzalos K, et al. What to Hide from Your Students: Attention- Guided Masked Image Modeling. Proceed- ings of the European Conference on Com- puter Vision (ECCV). 2022;13690:300–318

    work page 2022

  51. [51]

    Contrastive Masked Autoencoders are Stronger Vision Learners

    Huang Z, Jin X, Lu C, Hou Q, Cheng MM, Fu D, et al. Contrastive Masked Autoencoders are Stronger Vision Learners. IEEE Trans- actions on Pattern Analysis and Machine Intelligence. 2023;46(4):2506–2517

    work page 2023

  52. [52]

    What Do Self-Supervised Vision Transform- ers Learn? arXiv:230500729

    Park N, Kim W, Heo B, Kim T, Yun S. What Do Self-Supervised Vision Transform- ers Learn? arXiv:230500729. 2023

    work page 2023

  53. [53]

    Medical SAM 2: Segment medical images as video via Segment Anything Model 2

    Zhu J, Hamdi A, Qi Y, Jin Y, Wu J. Medical SAM 2: Segment medical images as video via Segment Anything Model 2. arXiv:240800874. 2024

    work page 2024

  54. [54]

    BiomedCLIP: a multimodal biomedical foundation model pretrained from fifteen million scientific image-text pairs

    Zhang S, Xu Y, Usuyama N, Xu H, Bagga J, Tinn R, et al. BiomedCLIP: a multimodal biomedical foundation model pretrained from fifteen million scientific image-text pairs. arXiv:230300915. 2023

    work page 2023

  55. [55]

    Domain-specific lan- guage model pretraining for biomedical nat- ural language processing

    Gu Y, Tinn R, Cheng H, Lucas M, Usuyama N, Liu X, et al. Domain-specific lan- guage model pretraining for biomedical nat- ural language processing. ACM Transactions on Computing for Healthcare (HEALTH). 2021;3(1):1–23

    work page 2021

  56. [56]

    Learning to Exploit Temporal Structure for Biomedical Vision- Language Processing

    Bannur S, Hyland S, Liu Q, P´ erez-Garc ´ ıa F, Ilse M, Castro DC, et al. Learning to Exploit Temporal Structure for Biomedical Vision- Language Processing. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2023;p. 15016–15027

    work page 2023

  57. [57]

    PathAsst: A Generative Founda- tion AI Assistant Towards Artificial General Intelligence of Pathology

    Sun Y, Zhu C, Zheng S, Zhang K, Sun L, Shui Z, et al. PathAsst: A Generative Founda- tion AI Assistant Towards Artificial General Intelligence of Pathology. Proceedings of the Thirty-Eighth AAAI Conference on Artificial Intelligence (AAAI-24). 2024;p. 5034–5042

    work page 2024

  58. [58]

    A whole-slide foundation model for digital pathology from real-world data

    Xu H, Usuyama N, Bagga J, Zhang S, Rao R, Naumann T, et al. A whole-slide foundation model for digital pathology from real-world data. Nature. 2024;630:181–188

    work page 2024

  59. [59]

    A foundation model for generalizable dis- ease detection from retinal images

    Zhou Y, Chia MA, Wagner SK, Ayhan MS, Williamson DJ, Struyven RR, et al. A foundation model for generalizable dis- ease detection from retinal images. Nature. 15 2023;622:156–163

    work page 2023

  60. [60]

    AbdomenAtlas-8K: Anno- tating 8,000 CT Volumes for Multi-Organ Segmentation in Three Weeks

    Qu C, Zhang T, Qiao H, Liu J, Tang Y, Yuille A, et al. AbdomenAtlas-8K: Anno- tating 8,000 CT Volumes for Multi-Organ Segmentation in Three Weeks. Proceedings of the 37th International Conference on Neural Information Processing Systems (NIPS 23). 2023;p. 36620–36636

    work page 2023

  61. [61]

    Application of the 4D XCAT Phantoms in Biomedical Imaging and Beyond

    Segars WP, Tsui B, Cai J, Yin FF, Fung GS, Samei E. Application of the 4D XCAT Phantoms in Biomedical Imaging and Beyond. IEEE Transactions on Medical Imaging. 2017;37(3):680–692

    work page 2017

  62. [62]

    Physiology- based simulation of the retinal vascula- ture enables annotation-free segmentation of OCT angiographs

    Menten MJ, Paetzold JC, Dima A, Menze BH, Knier B, Rueckert D. Physiology- based simulation of the retinal vascula- ture enables annotation-free segmentation of OCT angiographs. International Con- ference on Medical Image Computing and Computer-Assisted Intervention (MICCAI). 2022;13438:330–340. 16

    work page 2022

This paper was first reviewed by deepseek-v4-flash on August 4, 2026.