Pith. sign in

REVIEW 2 major objections 8 minor 155 references

A Comprehensive Study on Medical Image Segmentation using Deep Neural Networks

T0 review · 2 major / 8 minor · reviewed 2026-08-07 · deepseek-v4-flash

Pith's one-line read This review claims that deep-learning medical image segmentation is best understood as a three-level intelligent vision system, with explainable AI and early prediction as the bridge from intelligence to wisdom.

desk verdict A useful but not truly comprehensive survey; the DIKIW framing is fresh, but the unexplained omission of nnU-Net from the central comparison undercuts the main claim. read the letter →

arxiv 2506.04121 v1 pith:DOM3SUAI submitted 2025-06-04 eess.IV cs.AIcs.CV

classification eess.IVcs.AIcs.CV
keywords medicalimagesegmentationdeepneuralnetworksexplainableAIearlydiseasepredictionDIKIWhierarchyintelligentvisionsystemsU-Netarchitectureslossfunctions
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 is a review of deep-learning medical image segmentation (MIS) that tries to organize the field and give it a direction. It argues that current state-of-the-art solutions are best understood through three levels of an intelligent vision system: backbone network architectures, specialized segmentation architectures built on them, and clinical applications. It then claims that the field's next step is to move from "intelligence" (accurate segmentation) to "wisdom," by adding explainable AI to open the black box and by using segmentation results for early disease prediction. A sympathetic reader would care because this framing turns a scattered set of architectures, loss functions, and benchmarks into a coherent roadmap for building trustworthy clinical AI.

What carries the argument

The DIKIW hierarchy (Data, Information, Knowledge, Intelligence, Wisdom) is the conceptual ladder: the paper treats accurate segmentation as "intelligence" and treats explainable AI plus early prediction as the bridge to "wisdom." Operationally, the three-level intelligent vision system framework (backbone architectures → specialized MIS architectures → applications) carries the survey, and the comparison of U-Net, U-Net++, and TransUNet+ provides the concrete evidence that hybrid CNN-transformer designs are the current trend. The loss-function taxonomy adapted from the literature (distribution-, region-, boundary-, and association-based) and the standard metrics (Dice, IoU, Hausdorff distance) complete the machinery by giving practitioners categories for choosing and evaluating models.

What would settle it

A straightforward test is to classify every segmentation paper from one recent major medical imaging conference's proceedings into the proposed three levels (backbone, specific architecture, application) and to count how many mention XAI or early prediction. If a large share of papers cannot be assigned to a single level without special pleading, or if XAI and early prediction appear in only a small minority, then the taxonomy and the claimed centrality of those two themes are not comprehensive.

Watch

Extended reading notes

Core claim

The central claim is that the entire field of DNN-based medical image segmentation can be surveyed through the DIKIW hierarchy, with the current state of the art sitting at the "intelligence" level. The paper's organizing device is a three-level framework: Level 1 collects backbone DNNs (CNN, RNN, GNN, GAN, Transformer) and their supporting modules; Level 2 collects architectures specialized for segmentation (U-Net and its variants, transformer hybrids like TransUNet+); Level 3 collects clinical applications, from single-organ to multi-organ and from organ to lesion segmentation. The paper further argues that explainable AI and early prediction are the two mechanisms that raise MIS from intelligence to wisdom, and it supports this with a comparison of three representative networks, an overview of loss functions and evaluation metrics, and a survey of early-prediction applications.

Load-bearing premise

The review's claim to be comprehensive assumes that its chosen three-level taxonomy and selected published systems represent the whole medical image segmentation field.

Editorial extensions

If this is right

  • Using the three-level framing, future work can state exactly which level a new method improves, making it easier to compare contributions across backbone, architecture, and application.
  • XAI becomes a required evaluation dimension for segmentation systems intended for clinical use, not an optional extra, because transparency is tied to trust and ethics.
  • Segmentation-based early prediction becomes a concrete design goal, calling for longitudinal studies that combine past and current segmentation results to forecast disease progression.
  • The loss-function recommendations (Dice for mild imbalance, compound losses for severe imbalance) give a practical starting rule for practitioners facing class-imbalanced medical datasets.
  • The comparison of UNet, UNet++, and TransUNet+ suggests continued gains from hybrid architectures with redesigned skip connections, guiding the next generation of MIS backbones.

Reading between the lines

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

  • The "intelligence to wisdom" framing implies an evaluation rubric that weighs transparency and lead time alongside Dice score; one could extend the review by scoring each surveyed method on those two axes.
  • If early prediction is taken seriously, segmentation benchmarks should add longitudinal and temporal-consistency metrics, since a single static scan cannot measure how early a system detects change.
  • The taxonomy is flexible enough to absorb newer segmentation foundation models as Level 1 backbones, even though the review predates them and does not discuss them.
  • Because the review draws heavily on public leaderboard snapshots, its specific "state of the art" claims are time-bound; a testable extension is to re-run the same three-level organization on a later snapshot and see whether the taxonomy still holds.
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

2 major / 8 minor

Summary. The paper is a survey of deep-learning-based medical image segmentation (MIS). It organizes the field into three levels of an "intelligent vision system" (backbone architectures, task-specific architectures, and applications), and discusses data, loss functions, evaluation metrics, explainable AI (XAI), and early prediction. The authors state as contributions a comprehensive overview of state-of-the-art solutions, a focus on XAI for transparency, and a "new perspective" that links MIS to early disease prediction, framed within a DIKIW (Data–Information–Knowledge–Intelligence–Wisdom) hierarchy. The manuscript contains no new derivations or experiments; it is a literature review with tables summarizing network comparisons, challenge winners, modality comparisons, and leaderboard statistics as of 2022.

Significance. If taken at face value, the paper would offer a useful, organized entry point to the MIS literature, with accessible tables of MICCAI challenge winners, per-task counts from Papers with Code, and a taxonomy of loss functions and XAI methods. The explicit linkage of segmentation to XAI and early prediction is a reasonable framing that could help orient new researchers. The paper does not present new algorithms, code, or falsifiable predictions; its value is entirely in synthesis and organization. However, the strength of that synthesis depends on the representativeness and accuracy of its selection of methods, and that is where the main concern arises.

major comments (2)
  1. [Section II.A (Level 2), Table II] Table II compares only UNet, UNet++, and TransUNet+ as "three specific networks for the medical imaging segment." This selection is not justified, and it is contradicted by the paper's own evidence: Table IV lists nnU-Net (Isensee et al., refs. [53] and [55]) as the winning method in two of the six reported MICCAI challenges (KiTS19 in 2019 and AMOS22 in 2022), and Section II.B.2 explicitly names nnU-Net among the state-of-the-art 3D models. Since the paper's abstract and introduction claim a "comprehensive study" and "state-of-the-art solutions," the omission of nnU-Net from the central architecture comparison is an internal inconsistency. The authors should either include nnU-Net in Table II with a proper comparison or state an explicit inclusion criterion that explains why TransUNet+ was chosen instead of the most consistently successful method in the paper's own challenge data.
  2. [Section IV (Early Prediction with MIS)] The introduction lists as a major contribution "a new perspective on MIS by incorporating the capability of making early predictions based on the results." Section IV, however, is only a list of example applications (FCM for hematoma, ESFPNet for bronchial lesions, and a series of brief citations) without any synthesis: there is no definition of what "early prediction" means in the segmentation context, no analysis of how segmentation outputs are converted into temporal predictions, no discussion of evaluation protocols for early prediction, and no critical assessment of the cited methods. The claim that segmentation-based early prediction elevates the system from "intelligence" to "wisdom" is asserted, not argued. Since this is one of the paper's three advertised contributions, the section needs to be substantially strengthened to substantiate that claim.
minor comments (8)
  1. [Table II] The performance numbers are drawn from different datasets (LiTS for UNet/UNet++ and Synapse for TransUNet+), yet they are presented in a single comparison table without a caveat. The authors should note that cross-dataset comparisons are not meaningful without matched evaluation settings.
  2. [Section II.A.1 (RNN paragraph)] The description attributes the recurrent neural network concept to Pearlmutter (1989, ref. [10]), but that paper addresses learning state-space trajectories in recurrent networks, not the introduction of RNNs. The historical attribution should be corrected.
  3. [Section II.A.1 (CNN paragraph)] Calling Neocognitron (1982, ref. [3]) a "CNN" is imprecise; it is a biologically inspired predecessor. The text should distinguish between Neocognitron and the modern convolutional network formulation.
  4. [Section II.D, Eq. (7)] The Hausdorff distance formula contains an incomplete expression (the superscripts for the max/min are not properly rendered) and the sentence "take d (gt, ms) is taken as" is a typo. The formula and text should be cleaned up.
  5. [Section V.A.1] The reference to "S ection IV4" appears to be a typo; it should point to the correct subsection on learning paradigms (likely Section II.B.4).
  6. [Section V.B.3] The phrase "as back propagation be activated" is ungrammatical and should be rewritten (e.g., "as backpropagation proceeds through a deep network").
  7. [References] Several references are non-archival web sources or blog posts (e.g., [58], [60], [121], [123], [146], [147], [150]). For a scholarly survey, the authors should prefer peer-reviewed papers or official datasets, or at least mark such sources clearly as web resources.
  8. [Section III] The claim that linear regression is "unexplainable" because it does not justify the linearity assumption is debatable and would benefit from a more careful definition of explainability versus interpretability, or a citation to a standard reference.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the paper is a literature survey with no derived predictions, fitted parameters, or load-bearing self-citation chain.

full rationale

The paper is a survey of deep neural network medical image segmentation. It does not present an original derivation, model, experiment, or quantitative prediction of its own. Its central claims are organizational: it categorizes state-of-the-art solutions into three levels of an intelligent vision system, reviews network architectures, data modalities, loss functions, evaluation metrics, XAI methods, and early prediction applications, and summarizes existing challenge results. Every technical statement is attributed to external published work or public leaderboards. For example, Table II compares UNet, UNet++, and TransUNet+ using their reported scores on the LiTS and Synapse datasets, and Figures 3 and 5 report Dice scores from paperswithcode.com. There are no fitted parameters that are later renamed as predictions, and no equation in the paper reduces to its own inputs by construction. The 'comprehensive' and 'new perspective' claims are rhetorical and could be challenged on completeness or representativeness, but that is a correctness and scope concern, not a circularity concern. The authors do not invoke their own previous results to justify the survey's framework; no self-citation is load-bearing. Under the stated rules, a survey with no derivation chain should receive a non-finding of 0.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

The paper introduces no new entities, parameters, or postulates. Its only assumptions are the validity of the DIKIW framing, the reliability of external leaderboard data, and the representativeness of the selected literature. These are all domain-level assumptions rather than ad hoc constructs, but the comprehensiveness assumption is not rigorously supported.

assumptions (3)
  • domain assumption The DIKIW hierarchy (Data, Information, Knowledge, Intelligence, Wisdom) is a meaningful framework for evaluating intelligent vision systems.
    The entire organizational structure of the paper (three levels of IVS, intelligence-to-wisdom) relies on this framework, cited from reference [1].
  • domain assumption Leaderboard statistics from paperswithcode.com as of November 2022 are accurate and representative of state-of-the-art performance.
    The paper uses these rankings to identify top models such as SEP and ESFPNet, and to support claims about the current trend. This is an external, time-sensitive data source.
  • domain assumption The selection of cited works and topics is sufficient to support the claim of being 'comprehensive'.
    The paper asserts comprehensiveness but does not justify its literature coverage methodologically, and it omits notable areas that emerged later.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A Comprehensive Study on Medical Image Segmentation using Deep Neural Networks." pith.science (2026). https://pith.science/paper/DOM3SUAI

@misc{pith2026250604121,
  author       = {Pith},
  title        = {Pith review of: A Comprehensive Study on Medical Image Segmentation using Deep Neural Networks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DOM3SUAI}},
  note         = {Machine review of arXiv:2506.04121}
}
read the original abstract

Over the past decade, Medical Image Segmentation (MIS) using Deep Neural Networks (DNNs) has achieved significant performance improvements and holds great promise for future developments. This paper presents a comprehensive study on MIS based on DNNs. Intelligent Vision Systems are often evaluated based on their output levels, such as Data, Information, Knowledge, Intelligence, and Wisdom (DIKIW),and the state-of-the-art solutions in MIS at these levels are the focus of research. Additionally, Explainable Artificial Intelligence (XAI) has become an important research direction, as it aims to uncover the "black box" nature of previous DNN architectures to meet the requirements of transparency and ethics. The study emphasizes the importance of MIS in disease diagnosis and early detection, particularly for increasing the survival rate of cancer patients through timely diagnosis. XAI and early prediction are considered two important steps in the journey from "intelligence" to "wisdom." Additionally, the paper addresses existing challenges and proposes potential solutions to enhance the efficiency of implementing DNN-based MIS.

Figures

Figures reproduced from arXiv: 2506.04121 by the authors.

Figure 1
Figure 1. The pipeline of SOTA solutions in MIS. A. Network Architectures This study conducts a survey of state-of-the-art solutions in MIS based on the standard framework of intelligent vision systems (IVS). An IVS encompasses three main levels. Level 1 encompasses the backbone deep neural network (DNN) architectures used in image segmentation. Level 2 builds on Level 1 to create specialized image segmentation models. Level … view at source ↗
Figure 2
Figure 2. illustrates the procedure of SEP technique. The core concept of this technique is to copy the polyp region and paste it to other locations in order to generate a large number of new images. The augmentation process is divided into three modules: (1) a Potential Map Generation Module that generates a potential value for each coordinate, (2) a Pasting Module, and (3) an Update Module that updates the potential values … view at source ↗
Figure 3
Figure 3. Leaderboard the models with the highest mean Dice on the Kvasir￾SEG. ESFPNet (2022, [52]) is a deep learning architecture designed for real-time accurate segmentation and robust detection of bronchial lesions in autofluorescent bronchoscopy (AFB) video streams [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: ESFPNet architecture [52] [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Leaderboard the models with the highest mean Dice on the CVC-ClinicDB [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: An overview of 20 loss functions for MIS and their relationships [119]. D. Evaluation Metrics Segmentation performance evaluation involves comparing the similarity between a manually generated and a DNN￾generated segmentation. Many metrics are used for evaluation, but …
Figure 7
Figure 7. Figure 7: Illustrate GT, MS, TP, TN, FP, and FN modified from the source [121]. 1) Jaccard index (JAC - IoU) evaluates the overlap between GT and MS regions [PITH_FULL_IMAGE:figures/full_fig_p011_7.png]
Figure 8
Figure 8. Figure 8: Illustrate the Hausdorff distance [122]. 4) Other evaluation metrics  Sensitivity / Recall / True Positive Rate (TPR) [PITH_FULL_IMAGE:figures/full_fig_p012_8.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

155 extracted references · 61 canonical work pages

  1. [53]

    F., Kohl, S

    Isensee, F., Jäger, P. F., Kohl, S. A. A., Petersen, J., & Maier-Hein, K. H. (2019). Automated design of deep learning methods for biomedical image segmentation. In arXiv [cs.CV]. http://arxiv.org/abs/1904.08128

  2. [55]

    Isensee, F., Ulrich, C., Wald, T., & Maier-Hein, K. H. (2022). Extending nnU-Net is all you need. In arXiv [eess.IV]. http://arxiv.org/abs/2208.10791

  3. [1]

    Liew, ―DIKIW: Data, information, knowledge, Intelligence, wisdom and their interrelationships,‖ 2013

    A. Liew, ―DIKIW: Data, information, knowledge, Intelligence, wisdom and their interrelationships,‖ 2013

  4. [2]

    COVID -19 infection localization and severity grading from chest X -ray images,

    A. M. Tahir et al., “COVID -19 infection localization and severity grading from chest X -ray images,” Comput. Biol. Med., vol. 139, no. 105002, p. 105002, 2021

  5. [3]

    Fukushima and S

    K. Fukushima and S. Miyake, ―Neocognitr on: A new algorithm for pattern recognition tolerant of deformations and shifts in position,‖ Pattern Recognit., vol. 15, no. 6, pp. 455–469, 1982

  6. [4]

    Simonyan, K., & Zisserman, A. (2014). Very deep convolutional networks for large -scale image recognition. Ar Xiv [Cs.CV]. https://doi.org/10.48550/ARXIV.1409.1556

  7. [5]

    K. He, X. Zhang, S. Ren, and J. Sun, ―Deep residual learning for image recognition,‖ arXiv [cs.CV], 2015

  8. [6]

    van der, & Weinberger, K

    Huang, G., Liu, Z., Maaten, L. van der, & Weinberger, K. Q. (2017). Densely connected convolutional networks. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

Show all 155 references
  1. [7]

    -C., Papandreou, G., Schroff, F., & Adam, H

    Chen, L. -C., Papandreou, G., Schroff, F., & Adam, H. (2017). Rethinking Atrous Convolution for Semantic Image Seg mentation. ArXiv [Cs.CV]. https://doi.org/10.48550/ARXIV.1706.05587

  2. [8]

    G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., & Adam, H

    Howard, A. G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., & Adam, H. (2017). MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications. ArXiv [Cs.CV]. https://doi.org/10.48550/ARXIV.1704.04861 (IJACSA) International Journ...

  3. [9]

    Tan, M., & Le, Q. V. (2019). EfficientNet: Rethinking model scaling for convolutional Neural Networks. ArXiv [Cs.LG]. https://doi.org/10.48550/ARXIV.1905.11946

  4. [10]

    Pearlmutter, B. A. (1989). Learning state space trajectories in recurrent neural networks. Neural Computation, 1(2), 263–269. https://doi.org/10.1162/neco.1989.1.2.263

  5. [11]

    T. Lei, R. Wang, Y. Wan, X. Du, H. Meng, και A. K. Nandi, ‗Medical Image Segmentation Using Deep Learning: A Survey‘, 2020

  6. [12]

    C., Hagenbuchner, M., & Monfardini, G

    Scarselli, F., Gori, M., Tsoi, A. C., Hagenbuchner, M., & Monfardini, G. (2009). The graph neural network model. IEEE Transactions on Neural Networks, 20(1), 61–80. https://doi.org/10.1109/TNN.2008.2005605

  7. [13]

    Ahmedt-Aristizabal, M

    D. Ahmedt-Aristizabal, M. A. Armin, S. Denman, C. Fookes, and L. Petersson, ―Graph -based deep learning for medical diagnosis and analysis: Past, present and future,‖ Sensors (Basel), vol. 21, no. 14, p. 4758, 2021

  8. [14]

    I. J. Goodfellow et al., ―Generative Adversa rial Networks,‖ arXiv [stat.ML], 2014

  9. [15]

    Xun et al

    S. Xun et al. , ―Generative adversarial networks in medical image segmentation: A review,‖ Comput. Biol. Med. , vol. 140, no. 105063, p. 105063, 2021

  10. [16]

    N., Kaiser, L., & Polosukhin, I

    Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gom ez, A. N., Kaiser, L., & Polosukhin, I. (2017). Attention is all you need. In arXiv [cs.CL]. https://proceedings.neurips.cc/paper/2017/file/3f5ee243547dee91fbd053 c1c4a845aa-Paper.pdf

  11. [17]

    Chen et al

    J. Chen et al. , ―TransUNet: Transformers make strong encoders for medical image segmentation,‖ arXiv [cs.CV], 2021

  12. [18]

    Shamshad et al., ―Transformers in medical imaging: A survey,‖ arXiv [eess.IV], 2022

    F. Shamshad et al., ―Transformers in medical imaging: A survey,‖ arXiv [eess.IV], 2022

  13. [19]

    Szegedy, V

    C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, and Z. Wojna, ―Rethinking the inception architecture for computer vision,‖ arXiv [cs.CV], 2015

  14. [20]

    X. Ren, X. Guo, and S. Xu, ―A medical image segmentation method combining inception module with U-net model,‖ Research Square, 2022

  15. [21]

    Multi -scale context aggregation by dilated convolutions

    Yu F, Koltun V. Multi -scale context aggregation by dilated convolutions. In: International Conference on Learning Representations (2016). p. 1–13

  16. [22]

    Wang κ.ά., ‗Stacked dilated convolutions and asymmetric architecture for U -Net-based medical image segme ntation‘, Computers in Biology and Medicine, τ. 148, σ. 105891, 2022

  17. [23]

    Liu, W., Lei, H., Xie, H., Zhao, B., Yue, G., & Lei, B. (2020). Multi - level light U -net and atrous spatial pyramid pooling for optic disc segmentation on fundus image. In Ophthalmic Medi cal Image Analysis (pp. 104–113). Springer International Publishing

  18. [24]

    O., Alghamdi, R., & Mazher, M

    Qayyum, A., Ahmad, I., Mumtaz, W., Alassafi, M. O., Alghamdi, R., & Mazher, M. (2020). Automatic segmentation using a hybrid dense network integrated with an 3D -atrous spatial pyramid poo ling module for computed tomography (CT) imaging. IEEE Access: Practical Innovations, Op...

  19. [25]

    Wang, J., Lv, P., Wang, H., & Shi, C. (2021). SAR-U-Net: Squeeze-and- excitation block and atrous spatial pyramid pooling based residual U-Net for automatic liver segmentation in Computed Tomography. Computer Methods and Programs in Biomedicine, 208(106268), 106268. https://do...

  20. [26]

    Qiu, ―U -Net-ASPP: U-Net based on atrous spatial pyramid pooling model for medical image segmentation in COVID -19,‖ Tamkang j

    X. Qiu, ―U -Net-ASPP: U-Net based on atrous spatial pyramid pooling model for medical image segmentation in COVID -19,‖ Tamkang j. sci. eng., vol. 25, no. 6, pp. 1167–1176, 2022

  21. [27]

    M. M. Stofa, M. A. Zulkifley, M. A. A. M. Zainuri, and A. A. Ibrahim, ―U-net with atrous spatial pyramid pooling for skin lesion segmentation,‖ in Lecture Notes in Electrical Engineering, Singapore: Springer Singapore, 2022, pp. 1025–1033

  22. [28]

    Z. Yang, L. Chen, T. Fu, Z. Yin, and F. Yang, ―Spine Image Segmentation Based on U -Net and Atrous spatial pyramid pooling,‖ J. Phys. Conf. Ser., vol. 2209, no. 1, p. 012020, 2022

  23. [29]

    Xu et al., ―Show, attend and tell: Neural image caption generation with visual attention,‖ arXiv [cs.LG], pp

    K. Xu et al., ―Show, attend and tell: Neural image caption generation with visual attention,‖ arXiv [cs.LG], pp. 2048–2057, 07--09 Jul 2015

  24. [30]

    Gu et al., ―CA -Net: Comprehensive attention convolutional Neural Networks for explainable medical image segmentation,‖ IEEE Trans

    R. Gu et al., ―CA -Net: Comprehensive attention convolutional Neural Networks for explainable medical image segmentation,‖ IEEE Trans. Med. Imaging, vol. 40, no. 2, pp. 699–711, 2021

  25. [31]

    J. Hu, H. Wang, J. Wang, Y. Wang, F. He, and J. Zhang, ―SA -Net: A scale-attention network for medical image segmentation,‖ PLoS One, vol. 16, no. 4, p. e0247388, 2021

  26. [32]

    D. Nie, Y. Gao, L. Wang, and D. Shen, ―ASDNet: Attention based semi- supervised deep networks for medical im age segmentation,‖ in Medical Image Computing and Computer Assisted Intervention – MICCAI 2018, Cham: Springer International Publishing, 2018, pp. 370–378

  27. [33]

    J. Hu, L. Shen, and G. Sun, ―Squeeze -and-Excitation Networks,‖ in 2018 IEEE/CVF Conference on Comput er Vision and Pattern Recognition, 2018, pp. 7132–7141

  28. [34]

    A. G. Roy, N. Navab, and C. Wachinger, ―Recalibrating fully convolutional networks with spatial and channel ‗squeeze and excitation‘ blocks,‖ IEEE Trans. Med. Imaging, vol. 38, no. 2, pp. 540 – 549, 2019

  29. [35]

    Chen et al., ―An efficient memristor -based circuit implementation of squeeze-and-excitation fully convolutional neural networks,‖ IEEE Trans

    J. Chen et al., ―An efficient memristor -based circuit implementation of squeeze-and-excitation fully convolutional neural networks,‖ IEEE Trans. Neural Netw. Learn. Syst., vol. 33, no. 4, pp. 1779–1790, 2022

  30. [36]

    Iantsen, D

    A. Iantsen, D. Visvikis, and M. Hatt, ―Squ eeze-and-excitation normalization for automated delineation of head and neck primary tumors in combined PET and CT images,‖ in Head and Neck Tumor Segmentation, Cham: Springer International Publishing, 2021, pp. 37 – 43

  31. [37]

    Zhang, Y

    Y. Zhang, Y. Tian, Y. Kong, B. Zhong, and Y. Fu, ―Residual Dense Network for Image Super -Resolution,‖ in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018, pp. 2472–2481

  32. [38]

    Ronneberger, O., Fischer, P., & Brox, T. (2015). U -Net: Convolutional Networks for Biomedical Imag e Segmentation. In arXiv [cs.CV]. http://arxiv.org/abs/1505.04597

  33. [39]

    Milletari, F., Navab, N., & Ahmadi, S. -A. (2016). V -net: Fully convolutional neural networks for volumetric medical image segmentation. 2016 F ourth International Conference on 3D Vision (3DV), 565–571

  34. [40]

    S., Brox, T., & Ronneberger, O

    Çiçek, Ö., Abdulkadir, A., Lienkamp, S. S., Brox, T., & Ronneberger, O. (2016). 3D U -net: Learning dense volumetric segmentation from sparse annotation. In Medical Image Computing and Computer -Assisted Intervention – MICCAI 2016 (pp. 424 –432). Springer International Publishing

  35. [41]

    He, K., Gkioxari, G., Dollar, P., & Girshick, R. (2017). Mask R -CNN. 2017 IEEE International Conference on Computer Vision (ICCV)

  36. [42]

    Zhou, Z., Siddiquee, M. M. R., Tajbakhsh, N., & Liang, J. (2018). UNet++: A Nested U-Net Architecture for Medical Image Segmentation. ArXiv [Cs.CV]. https://doi.org/10.48550/ARXIV.1807.10165

  37. [43]

    Huang, H., Lin, L., Tong, R., Hu, H., Zhang, Q., Iwamoto, Y., Han, X., Chen, Y.-W., & Wu, J. (2020). UNet 3+: A full -scale connected UNet for medical image segmentation. In arXiv [eess.IV]. http://arxiv.org/abs/2004.08790

  38. [44]

    Jafari, M., Auer, D., Francis, S., Garibaldi, J., & Chen, X. (2020). DRU - Net: An Efficient Deep Convolutional Neural Network for Medical Image Segmentation. 2020 IEEE 17th International Symposium on Biomedical Imaging (ISBI), 1144–1148

  39. [45]

    A., Johansen, D., Halvorsen, P., & Johansen, H

    Jha, D., Riegler, M. A., Johansen, D., Halvorsen, P., & Johansen, H. D. (2020). DoubleU-Net: A deep convolutional neural network for medical image segmentation. ArXiv [Eess.IV]. https://doi.org/10.48550/ARXIV.2006.04868

  40. [46]

    L., & Zhou, Y

    Chen, J., Lu, Y., Yu, Q., Luo, X., Adeli, E., Wang, Y., Lu, L., Yuille, A. L., & Zhou, Y. (2021). TransUNet: Transformers make strong encoders for medical image segmentation. ArXiv [Cs.CV]. https://doi.org/10.48550/ARXIV.2102.04306

  41. [47]

    Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., & Guo, B. (2021). Swin Transformer: Hierarchical vision Transformer using shifted windows. ArXiv [Cs.CV]. https://doi.org/10.48550/ARXIV.2103.14030

  42. [48]

    -H., & Khan, F

    Shaker, A., Maaz, M., Rasheed, H., Khan, S., Yang, M. -H., & Khan, F. S. (2022). UNETR++: Delving into efficient and accurate 3D medical image segmentation. In arXiv [cs.CV]. http://arxiv.org/abs/2212.04497 (IJACSA) International Journal of Advanced Computer Science and Applic...

  43. [49]

    Zhao, Z., Zhou, F., Xu, K., Zeng, Z., Guan, C., & Zhou, S. K. (2022). LE-UDA: Label -efficient unsupervised domain adaptation for medical image segmentation. In arXiv [eess.IV]. http://arxiv.org/abs/2212.02078

  44. [50]

    Y. Liu, H. Wang, Z. Chen, K. Huangliang, and H. Zhang, ―TransUNet +: Redesigning the skip connection to enhance features in medical image segmentation,‖ Knowl. Based Syst., vol. 256, no. 109859, p. 109859, 2022

  45. [51]

    Zhou, L. (2022). Spatially exclusive pa sting: A general data augmentation for the polyp segmentation. In arXiv [eess.IV]. http://arxiv.org/abs/2211.08284

  46. [52]

    Chang, Q., Ahmad, D., Toth, J., Bascom, R., & Higgins, W. E. (2022). ESFPNet: efficient deep learning architecture for real -time lesion segmentation in autofluorescence bronchoscopic video. In arXiv [eess.IV]. http://arxiv.org/abs/2207.07759

  47. [54]

    Li, S., Sui, X., Luo, X., Xu, X., Liu, Y., & Goh, R. (2021). Medical Image Segmentation using Squeeze -and-Expansion Transformers. Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence

  48. [56]

    R. L. Siegel, K. D. Miller, H. E. Fuchs, and A. Jemal, ―Cancer statistics, 2022,‖ CA Cancer J. Clin., vol. 72, no. 1, pp. 7–33, 2022

  49. [57]

    Jun, SOTA-MedSeg: SOTA medical image segmentation methods based on various challenges

  50. [58]

    [Online]

    Wikipedia contributors, ―Medical imaging,‖ Wikipedia, The Free Encyclopedia, 22 -Aug-2022. [Online]. Available: https://en.wikipedia.org/w/index.php?title=Medical_imaging&oldid=110 5973862

  51. [59]

    KASBAN, Hany; EL -BENDARY, M. A. M.; SALAMA, D. H. A comparative study of medical imaging techniques. International Journal of Information Science and Intelligent System, 2015, 4.2: 37-58

  52. [60]

    [Online]

    International Journal of Radiology Sciences, ―A comparative study of medical imaging modalities,‖ Radiologyjournals.com. [Online]. Available: https://www.radiologyjournals.com/archives/2021.v3.i1.11. [Accessed: 03-Nov-2022]

  53. [61]

    Olubusola Isinkaye, A

    F. Olubusola Isinkaye, A. Gabriel Aluko, and O. Ayodele Jongbo, ―Segmentation of medical X -ray bone image using different image processing techniques,‖ Int. J. Image Graph. Signal Process., vol. 13, no. 5, pp. 27–40, 2021

  54. [62]

    Azimi et al., ―Improving clas sification model performance on chest X-rays through lung segmentation,‖ arXiv [eess.IV], 2022

    H. Azimi et al., ―Improving clas sification model performance on chest X-rays through lung segmentation,‖ arXiv [eess.IV], 2022

  55. [63]

    H. Zhu, Z. Cao, L. Lian, G. Ye, H. Gao, and J. Wu, ―CariesNet: a deep learning approach for segmentation of multi-stage caries lesion from oral panoramic X-ray image,‖ Neural Comput. Appl., pp. 1–9, 2022

  56. [64]

    S. P. Primakov et al., ―Automated detection and segmentation of non - small cell lung cancer computed tomography images,‖ Nat. Commun., vol. 13, no. 1, p. 3423, 2022

  57. [65]

    P. Tang, P. Yang, D. Nie, X. Wu, J. Zhou, and Y. Wang, ―Unified medical image segmentation by learning from uncertainty in an end -to- end manner,‖ Knowl. Based Syst., vol. 241, no. 108215, p. 108215, 2022

  58. [66]

    Isensee et al., ―nnU -Net: Self-adapting framework for U -Net-based medical image segmentation,‖ arXiv [cs.CV], 2018

    F. Isensee et al., ―nnU -Net: Self-adapting framework for U -Net-based medical image segmentation,‖ arXiv [cs.CV], 2018

  59. [67]

    Deng et al., ―A deep learning -based approach to automatic proximal femur segmentation in quantitative CT images,‖ Med

    Y. Deng et al., ―A deep learning -based approach to automatic proximal femur segmentation in quantitative CT images,‖ Med. Biol. Eng. Comput., vol. 60, no. 5, pp. 1417–1429, 2022

  60. [68]

    L. B. da Cruz et al., ―Kidney tumor segmentation from c omputed tomography images using DeepLabv3+ 2.5D model,‖ Expert Syst. Appl., vol. 192, no. 116270, p. 116270, 2022

  61. [69]

    Cui et al., ―A fully automatic AI system for tooth and alveolar bone segmentation from cone-beam CT images,‖ Nat

    Z. Cui et al., ―A fully automatic AI system for tooth and alveolar bone segmentation from cone-beam CT images,‖ Nat. Commun., vol. 13, no. 1, p. 2096, 2022

  62. [70]

    Hatamizadeh, V

    A. Hatamizadeh, V. Nath, Y. Tang, D. Yang, H. R. Roth, and D. Xu, ―Swin UNETR: Swin transformers for semantic segmentation of brain tumors in MRI images,‖ in Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries, Cham : Springer International Publishing,...

  63. [71]

    Karimi, H

    D. Karimi, H. Dou, and A. Gholipour, ―Medical image segmentation using transformer networks,‖ IEEE Access, vol. 10, pp. 29322 –29332, 2022

  64. [72]

    Boone et al., ―ROOD -MRI: Benchmarking the robustness of deep learning segmentation models to out -of-distribution and corrupted data in MRI,‖ arXiv [eess.IV], 2022

    L. Boone et al., ―ROOD -MRI: Benchmarking the robustness of deep learning segmentation models to out -of-distribution and corrupted data in MRI,‖ arXiv [eess.IV], 2022

  65. [73]

    Wu et al., ―Vessel -net: Retinal vessel segmentation under multi -path supervision,‖ in Lecture Notes in Computer Science, Cham: Springer International Publishing, 2019, pp

    Y. Wu et al., ―Vessel -net: Retinal vessel segmentation under multi -path supervision,‖ in Lecture Notes in Computer Science, Cham: Springer International Publishing, 2019, pp. 264–272

  66. [74]

    E. S. Uysal, M. Ş. Bilici, B. S. Zaza, M. Y. Özgenç, and O. Boyar, ―Exploring the limits of data augmentation for retinal Vessel Segmentation,‖ arXiv [eess.IV], 2021

  67. [75]

    C. You, Y. Zhou, R. Zhao, L. Staib, and J. S. Duncan, ―SimCVD: Simple contrastive voxel -wise representation distillation for semi - supervised medical image segmentation,‖ arXiv [cs.CV], 2021

  68. [76]

    Z. Xiao, Y. Su, Z. Deng, and W. Zhang, ―Efficient combination of CNN and transformer for Dual -teacher uncertainty -guided semi -supervised medical image segmentation,‖ Comput. Methods Programs Biomed., vol. 226, no. 107099, p. 107099, 2022

  69. [77]

    Q. Wang, Y. Du, H. Fan, and C. Ma, ―Towards collaborative appearance and semantic adaptation for medical image segmentation,‖ Neurocomputing, vol. 491, pp. 633–643, 2022

  70. [78]

    Ma and L

    L. Ma and L. Liang, ―Adaptive adversarial training to improve adversarial robustness of DNNs for medical image segmentation and detection,‖ arXiv [eess.IV], 2022

  71. [79]

    B. Lv, F. Liu, F. Gou, and J. Wu, ―Multi -scale tumor localization based on Priori guidance -based segmentation method for osteosarcoma MRI images,‖ Mathematics, vol. 10, no. 12, p. 2099, 2022

  72. [80]

    G. Chen, Y. Dai, and J. Zhang, ―C -Net: Cascaded convolutional neural network with global guidance and refinement residual s for breast ultrasound images segmentation,‖ Comput. Methods Programs Biomed., vol. 225, no. 107086, p. 107086, 2022

  73. [81]

    G. Chen, J. Yin, Y. Dai, J. Zhang, X. Yin, and L. Cui, ―A novel convolutional neural network for kidney ultrasound images segmentation,‖ Comput. Methods Programs Biomed., vol. 218, no. 106712, p. 106712, 2022

  74. [82]

    Jiang, Y

    J. Jiang, Y. Guo, Z. Bi, Z. Huang, G. Yu, and J. Wang, ―Segmentation of prostate ultrasound images: the state of the art and the future directions of segmentation algorithms,‖ Artif. Intell. Rev., vol. 56, no. 1, pp. 615 – 651, 2023

  75. [83]

    Q. Zhou, Q. Wang, Y. Bao, L. Kong, X. Jin, and W. Ou, ―LAEDNet: A Lightweight Attention Encoder –Decoder Network for ultrasound medical image segmentation,‖ Comput. Electr. Eng., vol. 99, no. 107777, p. 107777, 2022

  76. [84]

    Isensee, P

    F. Isensee, P. F. Jäger, S. A. A. Kohl, J. Petersen, and K. H. Maier-Hein, ―Automated design of deep learning methods for biomedical image segmentation,‖ arXiv [cs.CV], 2019

  77. [85]

    W. Li, G. Wang, L. Fidon, S. Ourselin, M. J. Cardoso, and T. Vercauteren, ―On the compactness, efficiency, and representation of 3D convolutional networks: Brain parcellation as a pretext task,‖ in Lecture Notes in Computer Science, Cham: Springer International Publishing, 201...

  78. [86]

    Mathlouthi, A

    S. Mathlouthi, A. G. Blaiech, M. Sa id, A. B. Abdallah, and M. Hedi Bedoui, ―A novel deep learning model for knee cartilage 3D segmentation,‖ in 2021 IEEE/ACS 18th International Conference on Computer Systems and Applications (AICCSA), 2021, pp. 1–7

  79. [87]

    Gibson et al., ―Automatic multi -organ segmentation on abdominal CT with dense V -networks,‖ IEEE Trans

    E. Gibson et al., ―Automatic multi -organ segmentation on abdominal CT with dense V -networks,‖ IEEE Trans. Med. Imaging, vol. 37, no. 8, pp. 1822–1834, 2018

  80. [88]

    R., & Xu, D

    Hatamizadeh, A., Tang, Y., Nath, V., Yang, D., Myronenko, A., Landman, B., Roth, H. R., & Xu, D. (2022). UNETR: Transformers for 3D Medical Image Segmentation. 2022 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 574–584

  81. [89]

    C. Hsu, C. Chang, T. W. Chen, H. Tsai, S. Ma, and W. Wang, ―Brain tumor segmentation (BraTS) challenge short paper: Improving three - (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 14, No. 3, 2023 183 | P a g e www.ijacsa.thesai.org dimensio...

  82. [90]

    N.-V. Ho, T. Nguyen, G.-H. Diep, N. Le, and B. -S. Hua, ―Point-unet: A context-aware point-based neural network for volumetric segmentation,‖ in Medical Image Computing and Computer Assisted Intervention – MICCAI 2021, Cham: Springer International Publishing, 2021, pp. 644– 655

  83. [91]

    McInerney and D

    T. McInerney and D. Terzo poulos, ―A dynamic finite element surface model for segmentation and tracking in multidimensional medical images with application to cardiac 4D image analysis,‖ Comput. Med. Imaging Graph., vol. 19, no. 1, pp. 69–83, 1995

  84. [92]

    M. C. Aznar, G. F. Persson, I. M . Kofoed, D. E. Nygaard, and S. S. Korreman, ―Irregular breathing during 4DCT scanning of lung cancer patients: is the midventilation approach robust?,‖ Phys. Med. , vol. 30, no. 1, pp. 69–75, 2014

  85. [93]

    Y. Gao, J. M. Phillips, Y. Zheng, R. Min, P. T. Fletcher, and G. Gerig, ―Fully convolutional structured LSTM networks for joint 4D medical image segmentation,‖ in 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018), 2018, pp. 1104–1108

  86. [94]

    W. P. Segars, B. M. W. Tsui, J. Cai, F. -F. Yin, G. S. K. Fung, and E. Samei, ―Application of the 4-D XCAT phantoms in biomedical imaging and beyond,‖ IEEE Trans. Med. Imaging , vol. 37, no. 3, pp. 680 –692, 2018

  87. [95]

    Deo, R. C. (2015). Machine learning in medicine. Circulation, 132(20), 1920–1930. https://doi.org/10.1161/CIRCULATIONAHA.115.001593

  88. [96]

    G. S. Handelman, H. K. Kok, R. V. Chandra, A. H. Razavi, M. J. Lee, and H. Asadi, ―eDoctor: machine learning and the future of medicine,‖ J. Intern. Med., vol. 284, no. 6, pp. 603–619, 2018

  89. [97]

    Militello et al

    C. Militello et al. , ―On unsupervised methods for medical image segmentation: Investigating classic approaches in breast cancer DCE - MRI,‖ Appl. Sci. (Basel), vol. 12, no. 1, p. 162, 2021

  90. [98]

    Gholamiankhah, S

    F. Gholamiankhah, S. Mostafapour, N. A. Goushbolagh, S. Shojaerazavi, H. Arabi, and H. Zaidi, ―A novel unsupervised COVID-19 lesion segmentation from CT images based -on the lung tissue detection,‖ in 2021 IEEE Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC)...

  91. [99]

    X. Wang, C. Guo, and X. Zhou, ―Robust segmentation of 3D brain MRI images in cross datasets by integrating supervised and unsupervised learning,‖ in 2020 10th International Conference on Information Science and Technology (ICIST), 2020, pp. 194–201

  92. [100]

    Aganj, M

    I. Aganj, M. G. Harisinghani, R. Weissleder, and B. Fischl, ―Unsupervised medical image segmentation based on the local center of mass,‖ Sci. Rep., vol. 8, no. 1, p. 13012, 2018

  93. [101]

    C. Chen, Q. Dou, H. Chen, J. Qin, and P. A. Heng, ―Unsupervised bidirectional cross-modality adaptation via deeply Synergistic Image and Feature Alignment for medical image segmentation,‖ IEEE Trans. Med. Imaging, vol. 39, no. 7, pp. 2494–2505, 2020

  94. [102]

    Chebli, A

    A. Chebli, A. Djebbar, and H. F. Marouani, ―Semi -supervised learning for medical application: A survey,‖ in 2018 International Conference on Applied Smart Systems (ICASS), 2018, pp. 1–9

  95. [103]

    Y. Li, J. Chen, X. Xie, K. Ma, and Y. Zheng, ―Self -loop uncertainty: A novel pseudo-label for semi-supervised medical image segmentation,‖ in Medical Image Computing and Computer Assisted Intervention – MICCAI 2020, Cham: Springer International Publishing, 2020, pp. 614 – 623

  96. [104]

    C. You, Y. Zhou, R. Zhao, L. Staib, and J. S. Duncan, ―SimCVD: Simple contrastive voxel -wise representation distillati on for semi - supervised medical image segmentation,‖ IEEE Trans. Med. Imaging , vol. 41, no. 9, pp. 2228–2237, 2022

  97. [105]

    Zhang, Y., & Zhang, J. (2021). Dual -task mutual learning for semi - supervised medical image segmentation. In Pattern Recognition and Computer Vision (pp. 548–559). Springer International Publishing

  98. [106]

    Ouali, C

    Y. Ouali, C. Hudelot, and M. Tami, ―An overview of deep semi - supervised learning,‖ arXiv [cs.LG], 2020

  99. [107]

    Yang, H., Wu, G., Shen, D., & Liao, S. (2021). Automatic prostate cancer detection on multi -parametric MRI with hierarchical weakly supervised learning. 2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI), 316–319

  100. [108]

    J., Yip, T

    Lyu, F., Ma, A. J., Yip, T. C.-F., Wong, G. L.-H., & Yuen, P. C. (2022). Weakly supervised liver tumor segmentation usin g Couinaud segment annotation. IEEE Transactions on Medical Imaging , 41(5), 1138–1149. https://doi.org/10.1109/TMI.2021.3132905

  101. [109]

    B., Mccaffrey, L., & Granger, E

    Belharbi, S., Rony, J., Dolz, J., Ayed, I. B., Mccaffrey, L., & Granger, E. (2022). Deep interpretable classification and weakly -supervised segmentation of histology images via max -min uncertainty. IEEE Transactions on Medical Imaging , 41(3), 702 –714. https://doi.org/10.11...

  102. [110]

    Weiss, T

    K. Weiss, T. M. Khoshgoftaar, and D. Wang, ―A survey of transfer learning,‖ J. Big Data, vol. 3, no. 1, 2016

  103. [111]

    J. Deng, W. Dong, R. Socher, L.-J. Li, K. Li, and L. Fei-Fei, ―ImageNet: A large-scale hierarchical image database,‖ in 2009 IEEE Conference on Computer Vision and Pattern Recognition, 2009, pp. 248–255

  104. [112]

    A., Vernooij, M

    van Opbroek, A., Ikram, M. A., Vernooij, M. W., & de Bruijne, M. (2015). Transfer learning improves supervised image segmentation across imaging protocols. IEEE Transactions on Medical Imaging , 34(5), 1018–1030. https://doi.org/10.1109/TMI.2014.2366792

  105. [113]

    C., Vernooij, M

    Van Opbroek, A., Achterberg, H. C., Vernooij, M. W., & De Bruijne, M. (2019). Transfer learning for image segmentation by combining image weighting and kernel learning. IEEE Transactions on Medical Imaging , 38(1), 213–224. https://doi.org/10.1109/TMI.2018.2859478

  106. [114]

    Miloslavskaya and A

    N. Miloslavskaya and A. Tolstoy, ―Big data, fa st data and data lake concepts,‖ Procedia Comput. Sci., vol. 88, pp. 300–305, 2016

  107. [115]

    A., Xu, D., & Roth, H

    Nath, V., Yang, D., Landman, B. A., Xu, D., & Roth, H. R. (2021). Diminishing uncertainty within the training pool: Active learning for medical image segmentation. IEEE Tra nsactions on Medical Imaging , 40(10), 2534–2547. https://doi.org/10.1109/TMI.2020.3048055

  108. [116]

    E., Sajed, D

    Li, W., Li, J., Wang, Z., Polson, J., Sisk, A. E., Sajed, D. P., Speier, W., & Arnold, C. W. (2022). PathAL: An active learning framework for histopathology image analysis. IEEE Transactions on Medical Imaging , 41(5), 1176–1187. https://doi.org/10.1109/TMI.2021.3135002

  109. [117]

    Wang, J., Chen, Z., Wang, L., & Zhou, Q. (2019). An active learning with two-step query for medical image segmentation. 2019 International Conference on Medical Imaging Physics and Engineering (ICMIPE), 1– 5

  110. [118]

    Li, W., Zhang, M., & Chen, D. (2020). Fundus retinal blood vessel segmentation based o n active learning. 2020 International Conference on Computer Information and Big Data Applications (CIBDA) , 264 – 268

  111. [119]

    Ma, J., Chen, J., Ng, M., Huang, R., Li, Y., Li, C., Yang, X., & Martel, A. L. (2021). Loss odyssey in medical image segmentation. Medical Image Analysis, 71(102035), 102035. https://doi.org/10.1016/j.media.2021.102035

  112. [120]

    S. A. Taghanaki et al., ―Combo loss: Handling input and output imbalance in multi -organ segmentation,‖ Comput. Med. Imaging Graph., vol. 75, pp. 24–33, 2019

  113. [121]

    X. P. Burgos -Artizzu, ―Computer -aided covid -19 patient screening using chest images (X-Ray and CT scans),‖ bioRxiv, 2020

  114. [122]

    Kaspar et al., ―Figure 3.4: The Hausdorff distance is defined as the maximum of all,‖ R esearchGate

    D. Kaspar et al., ―Figure 3.4: The Hausdorff distance is defined as the maximum of all,‖ R esearchGate. [Online]. Available: https://www.researchgate.net/figure/The-Hausdorff-distance-is-defined- as-the-maximum-of-all-minimum-distances-between- two_fig20_242202827. [Accessed: ...

  115. [123]

    A. C. N. Matcha, ―Class activation maps: Visualizing neura l network decision-making,‖ Heartbeat, 28 -Oct-2019. [Online]. Available: https://heartbeat.comet.ml/class-activation-maps-visualizing-neural- network-decision-making-92efa5af9a33. [Accessed: 11-Nov-2022]

  116. [124]

    B. Zhou, A. Khosla, A. Lapedriza, A. Oliva, and A. T orralba, ―Learning deep features for discriminative localization,‖ in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016

  117. [125]

    R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D

    Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2016). Grad -CAM: Visual expl anations from deep networks via Gradient-based localization. In arXiv [cs.CV]

  118. [126]

    Chattopadhay, A., Sarkar, A., Howlader, P., & Balasubramanian, V. N. (2018). Grad -CAM++: Generalized gradient -based visual explanations (IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 14, No. 3, 2023 184 | P a g e www.ijacsa.thesai.org for ...

  119. [127]

    M. T. Ribeiro, S. Singh, and C. Guestrin, ―Why should I trust you?: Explaining the predictions of any classifier,‖ in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016

  120. [128]

    Lundberg and S

    S. Lundberg and S. -I. Lee, ―A unified approach to interpreting model predictions,‖ arXiv [cs.AI], 2017

  121. [129]

    W. Xu et al., ―Early prediction of cerebral computed tomography under intelligent segmentation algorithm co mbined with serological indexes for hematoma enlargement after intracerebral hemorrhage,‖ Comput. Math. Methods Med., vol. 2022, p. 5863082, 2022

  122. [130]

    Shekhar, D

    S. Shekhar, D. Eswaran, B. Hooi, J. Elmer, C. Faloutsos, and L. Akoglu, ―Benefit-aware early prediction of health outcomes on multivariate EEG time series,‖ arXiv [cs.LG], 2021

  123. [131]

    Ali et al., ―A self-training deep neural network for early prediction of cognitive deficits in very preterm infants using brain functional connectome data,‖ Pediatr

    R. Ali et al., ―A self-training deep neural network for early prediction of cognitive deficits in very preterm infants using brain functional connectome data,‖ Pediatr. Radiol., vol. 52, n o. 11, pp. 2227 –2240, 2022

  124. [132]

    R., & Raza, B

    Saleem, H., Shahid, A. R., & Raza, B. (2021). Visual interpretability in 3D brain tumor segmentation network. Computers in Biology and Medicine, 133(104410), 104410. https://doi.org/10.1016/j.compbiomed.2021.104410

  125. [133]

    Wu et al., ―Intelligent segmentation medical assistance system for MRI images of osteosarcoma in developing countries,‖ Comput

    J. Wu et al., ―Intelligent segmentation medical assistance system for MRI images of osteosarcoma in developing countries,‖ Comput. Math. Methods Med., vol. 2022, p. 7703583, 2022

  126. [134]

    W. Chen, M. Yao, Z. Zhu, Y. Sun, and X. Han, ―The application research of AI image recognition and processing technology in the early diagnosis of the COVID-19,‖ BMC Med. Imaging, vol. 22, no. 1, p. 29, 2022

  127. [135]

    Niyaz, A

    U. Niyaz, A. S. Sambyal, and Devanand, ―Advances in deep learning techniques for medical image analysis,‖ in 2018 Fifth International Conference on Parallel, Distributed and Grid Computing (PDGC), 2018, pp. 271–277

  128. [136]

    A., Palma, D

    Mattonen, S. A., Palma, D. A., Haasbeek, C. J. A., Senan, S., & Ward, A. D. (2014). Early prediction of tumor r ecurrence based on CT texture changes after stereotactic ablative radiotherapy (SABR) for lung cancer: Predicting recurrence after SABR using second -order texture s...

  129. [137]

    R., Yates, E

    Eldesoky, A. R., Yates, E. S., Nyeng, T. B., Thomsen, M. S., Nielsen, H. M., Poortmans, P., Kirkove, C., Krause, M., Kamby, C., Mjaaland, I., Blix, E. S., Jensen, I., Berg, M., Lorenzen, E. L., Taheri -Kadkhoda, Z., & Offersen, B. V. (2016). Internal and external validation of...

  130. [138]

    de Oliveira, J. P. S., Conci, A., Perez, M. G., & Andaluz, V. H. (2015). Segmentation of infrared ima ges: A new technology for early detection of breast diseases. 2015 IEEE International Conference on Industrial Technology (ICIT), 1765–1771

  131. [139]

    G. Danala et al., ―Improving efficacy of metastatic tumor segmentation to facilitate early prediction of ovarian c ancer patients‘ response to chemotherapy,‖ in Biophotonics and Immune Responses XII, 2017, vol. 10065, pp. 47–52

  132. [140]

    K., & Wardlaw, J

    Rekik, I., Allassonnière, S., Carpenter, T. K., & Wardlaw, J. M. (2012). Medical image analysis methods in MR/CT -imaged acute -subacute ischemic stroke lesion: Segmentation, prediction and insights into dynamic evolution simulation models. A critical appraisal. NeuroImage. Cl...

  133. [141]

    G., de Haan, R

    Zandbergen, E. G., de Haan, R. J., Stoutenbeek, C. P., Koelman, J. H., & Hijdra, A. (1998). Systematic review of early prediction of poor outcome in anoxic -ischaemic coma. Lancet, 352(9143), 1808 –1812. https://doi.org/10.1016/S0140-6736(98)04076-8

  134. [142]

    U., Johannes, R

    Wu, B. U., Johannes, R. S., Sun, X., Tabak, Y., Conwell, D. L., & Banks, P. A. (2008). The early prediction of mortality in acute pancreatitis: a large population -based study. Gut, 57(12), 1698 –1703. https://doi.org/10.1136/gut.2008.152702

  135. [143]

    Hoeben, B. A. W., Troost, E. G. C., Span, P. N., van Herpen, C. M. L., Bussink, J., Oyen, W. J. G., & Kaanders, J. H. A. M. (2013). 18F -FLT PET during radiotherapy or chemoradiotherapy in head and neck squamous cell carcinoma is an early predictor of outcome. Journal of Nucle...

  136. [144]

    Fang, Y.-J., Mukundan, A., Tsao, Y.-M., Huang, C.-W., & Wang, H.-C. (2022). Ide ntification of early esophageal cancer by semantic segmentation. Journal of Personalized Medicine, 12(8), 1204. https://doi.org/10.3390/jpm12081204

  137. [145]

    Liu, S., Liu, K., Zhu, W., Shen, Y., & Fernandez -Granda, C. (2022). Adaptive early -learning correction for segmentation from noisy annotations. 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2606–2616

  138. [146]

    [Online]

    VinBigData, ―10 open data repositories for the medical image processing communi ty,‖ VinBigdata - Blog , 28 -Jun-2021. [Online]. Available: https://blog.vinbigdata.org/10 -kho-du-lieu-mo-danh-cho- cong-dong-xu-ly-anh-y-te/. [Accessed: 01-Oct-2022]

  139. [147]

    Joshi, ―4 Ways to Tackle the Lack of Machine Learning Datasets,‖ Bbntimes.com

    N. Joshi, ―4 Ways to Tackle the Lack of Machine Learning Datasets,‖ Bbntimes.com. [Onlin e]. Available: https://www.bbntimes.com/technology/4-ways-to-tackle-the-lack-of- machine-learning-datasets. [Accessed: 01-Oct-2022]

  140. [148]

    Zhang, J

    Z. Zhang, J. Li, Z. Zhong, Z. Jiao, and X. Gao, ―A sparse annotation strategy based on attention-guided active learning for 3D medical image segmentation,‖ arXiv [cs.CV], 2019

  141. [149]

    U. Vovk, F. Pernus, and B. Likar, ―A review of methods for correction of intensity inhomogeneity in MRI,‖ IEEE Trans. Med. Imaging, vol. 26, no. 3, pp. 405–421, 2007

  142. [150]

    [Online]

    ―What is Overfitting in Deep Lear ning [+10 Ways to Avoid It],‖ V7labs.com. [Online]. Available: https://www.v7labs.com/blog/overfitting. [Accessed: 01-Oct-2022]

  143. [151]

    Ashkani Chenarlogh et al., ―Clinical target segmentation using a novel deep neural network: double attention Res -U-Net,‖ Sci

    V. Ashkani Chenarlogh et al., ―Clinical target segmentation using a novel deep neural network: double attention Res -U-Net,‖ Sci. Rep., vol. 12, no. 1, p. 6717, 2022

  144. [152]

    Y. S. Jeon, H. Yang, and M. Feng, ―FCSN: Global context aware segmentation by learning the Fourier coefficients of objects in medical images,‖ arXiv [eess.IV], 2022

  145. [153]

    J. Sun, F. Darbehani, M. Zaidi, and B. Wang, ―SAU-Net: Shape Attentive U-Net for Interpretable Medical Image Segmentation,‖ arXiv [eess.IV], 2020

  146. [154]

    Seibold, J

    C. Seibold, J. Künzel, A. Hilsmann, and P. Eisert, ―From explanations to segmentation: Using Explainable AI for image segmentation,‖ arXiv [cs.CV], 2022

  147. [155]

    MIASNet: A medical image segmentation method predicting future based on past and current cases,

    S. Ma, X. Li, J. Tang and F. Guo, "MIASNet: A medical image segmentation method predicting future based on past and current cases," 2021 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 2021, pp. 1014 -1019, d oi: 10.1109/BIBM52615.2021.966963

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.