pith. machine review for the scientific record. sign in

arxiv: 2604.09468 · v1 · submitted 2026-04-10 · 📡 eess.IV · cs.CV

Recognition: unknown

DSVTLA: Deep Swin Vision Transformer-Based Transfer Learning Architecture for Multi-Type Cancer Histopathological Cancer Image Classification

Muazzem Hussain Khan , Tasdid Hasnain , Md. Jamil khan , Ruhul Amin , Md. Shamim Reza , Md. Al Mehedi Hasan , Md Ashad Alam
Authors on Pith no claims yet

Pith reviewed 2026-05-10 16:34 UTC · model grok-4.3

classification 📡 eess.IV cs.CV
keywords histopathological image classificationSwin Vision Transformertransfer learningmulti-cancer classificationdeep learningmedical imagingvision transformercancer diagnosis
0
0 comments X

The pith

A hybrid Swin Transformer and ResNet50 model classifies multiple cancer histopathology images with up to 100% accuracy.

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

The paper proposes a transfer learning architecture that combines a hierarchical Swin Vision Transformer with ResNet50 convolutional features for classifying images across several cancer types. It evaluates the model on datasets for breast cancer, oral cancer, lung and colon cancer, kidney cancer, and acute lymphocytic leukemia, using both original and segmented versions of the images. Competing models including various CNNs and other vision transformers receive the same preprocessing, training, and validation steps for direct comparison. The proposed architecture records the highest results, including 100% accuracy on lung-colon and segmented leukemia cases plus 99.23% on breast cancer. These outcomes could support the creation of automated tools that assist in examining tissue samples for cancer diagnosis.

Core claim

The proposed DSVTLA framework integrates a hierarchical Swin Transformer with ResNet50-based convolution features extraction, enabling the model to capture both long-range contextual dependencies and fine-grained local morphological patterns within histopathological images, and when benchmarked alongside state-of-the-art CNN and transfer models including DenseNet121, DenseNet201, InceptionV3, ResNet50, EfficientNetB3, multiple ViT variants, and Swin Transformer models using a unified pipeline on multi-cancer datasets including Breast Cancer, Oral Cancer, Lung and Colon Cancer, Kidney Cancer, and Acute Lymphocytic Leukemia with both original and segmented images, it achieves 100% test accurac

What carries the argument

The hybrid architecture that pairs a hierarchical Swin Vision Transformer for global dependencies with ResNet50 convolutional features for local morphological details in a transfer learning setup for multi-cancer image classification.

Load-bearing premise

High accuracies measured on the specific multi-cancer datasets will generalize to new clinical images from different hospitals, scanners, and patient populations without overfitting or data-specific biases.

What would settle it

A clear drop in accuracy when the model is applied to a fresh collection of histopathological images gathered from independent clinical sites that use different scanners or staining methods.

Figures

Figures reproduced from arXiv: 2604.09468 by Md. Al Mehedi Hasan, Md Ashad Alam, Md. Jamil khan, Md. Shamim Reza, Muazzem Hussain Khan, Ruhul Amin, Tasdid Hasnain.

Figure 7
Figure 7. Figure 7: SHAP Global Feature Importance for all proposed datasets (left to right: Breast, Colon, Kidney, Leukemia-original, Leukemia-segmented, Lung, Oral). This performance gain highlights the effectiveness of integrating convolutional feature extraction with hierarchical self-attention mechanisms. ResNet50 effectively captures fine-grained local texture patterns, while the Swin Transformer models broader spatial … view at source ↗
Figure 8
Figure 8. Figure 8: SHAP Summary Plots for all proposed datasets (left to right: Breast, Colon, Kidney, Leukemia￾original, Leukemia-segmented, Lung, Oral) [PITH_FULL_IMAGE:figures/full_fig_p019_8.png] view at source ↗
read the original abstract

In this study, we proposed a deep Swin-Vision Transformer-based transfer learning architecture for robust multi-cancer histopathological image classification. The proposed framework integrates a hierarchical Swin Transformer with ResNet50-based convolution features extraction, enabling the model to capture both long-range contextual dependencies and fine-grained local morphological patterns within histopathological images. To validate the efficiency of the proposed architecture, an extensive experiment was executed on a comprehensive multi-cancer dataset including Breast Cancer, Oral Cancer, Lung and Colon Cancer, Kidney Cancer, and Acute Lymphocytic Leukemia (ALL), including both original and segmented images were analyzed to assess model robustness across heterogeneous clinical imaging conditions. Our approach is benchmarked alongside several state-of-the-art CNN and transfer models, including DenseNet121, DenseNet201, InceptionV3, ResNet50, EfficientNetB3, multiple ViT variants, and Swin Transformer models. However, all models were trained and validated using a unified pipeline, incorporating balanced data preprocessing, transfer learning, and fine-tuning strategies. The experimental results demonstrated that our proposed architecture consistently gained superior performance, reaching 100% test accuracy for lung-colon cancer, segmented leukemia datasets, and up to 99.23% accuracy for breast cancer classification. The model also achieved near-perfect precision, f1 score, and recall, indicating highly stable scores across divers cancer types. Overall, the proposed model establishes a highly accurate, interpretable, and also robust multi-cancer classification system, demonstrating strong benchmark for future research and provides a unified comparative assessment useful for designing reliable AI-assisted histopathological diagnosis and clinical decision-making.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 3 minor

Summary. The paper proposes DSVTLA, a hybrid architecture integrating a hierarchical Swin Vision Transformer with ResNet50-based convolutional feature extraction for transfer learning-based classification of multi-type cancer histopathological images (breast, oral, lung-colon, kidney, and acute lymphocytic leukemia, including both original and segmented variants). It benchmarks the model against multiple CNNs (DenseNet121/201, InceptionV3, ResNet50, EfficientNetB3) and ViT/Swin variants under a single unified pipeline of preprocessing, transfer learning, and fine-tuning, claiming consistent superiority with 100% test accuracy on lung-colon and segmented leukemia datasets and 99.23% on breast cancer, plus near-perfect precision/recall/F1 scores.

Significance. If the performance claims are supported by rigorous validation, the work would offer a useful empirical benchmark for hybrid transformer-CNN transfer learning in multi-cancer histopathology, highlighting the value of combining long-range context modeling with local morphological features. The unified comparative evaluation across heterogeneous datasets and model families is a constructive element for the field.

major comments (3)
  1. [§4, §3] §4 (Experimental Results) and §3 (Methodology): The reported 100% test accuracy on lung-colon cancer and segmented leukemia datasets (and 99.23% on breast) is presented without any description of the train/test split protocol. In histopathology, random per-image splits routinely induce leakage because patches from the same slide or patient share staining, scanner, and cellular patterns; the absence of patient-level or slide-level partitioning details directly undermines the central claim that the model achieves robust, generalizable superiority rather than an artifact of data correlation.
  2. [§4] §4 (Results tables/figures): No ablation studies isolate the contribution of the Swin Transformer hierarchy versus the ResNet50 backbone, nor are repeated random splits, cross-validation folds, or statistical significance tests (e.g., paired t-tests or McNemar’s test against baselines) reported. Without these, the assertion of consistent outperformance over DenseNet, EfficientNet, and ViT variants cannot be evaluated as load-bearing evidence.
  3. [§3] §3 (Dataset description): Dataset sizes, class balances, and the exact number of images per cancer type (original vs. segmented) are not quantified, nor is any multi-center or external hold-out validation mentioned. This omission makes it impossible to assess whether the near-perfect metrics reflect genuine robustness across the claimed “heterogeneous clinical imaging conditions.”
minor comments (3)
  1. [Abstract] Abstract: “gained superior performance” should read “achieved superior performance”; “divers cancer types” should be “diverse cancer types”; the phrase “also robust” is redundant.
  2. [Figures/Tables] Figure and table captions throughout: Ensure all axes, color legends, and metric definitions (e.g., whether accuracy is macro- or micro-averaged) are explicitly labeled for reproducibility.
  3. [§2] §2 (Related Work): A few recent transformer-based histopathology papers (e.g., post-2022 Swin or hybrid ViT works) appear to be missing; adding them would strengthen the positioning of DSVTLA.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive and detailed feedback, which highlights important aspects of rigorous validation in histopathological image classification. We have addressed each major comment by clarifying our experimental setup where possible and committing to revisions that strengthen the manuscript's transparency and evidence.

read point-by-point responses

  1. Referee: [§4, §3] §4 (Experimental Results) and §3 (Methodology): The reported 100% test accuracy on lung-colon cancer and segmented leukemia datasets (and 99.23% on breast) is presented without any description of the train/test split protocol. In histopathology, random per-image splits routinely induce leakage because patches from the same slide or patient share staining, scanner, and cellular patterns; the absence of patient-level or slide-level partitioning details directly undermines the central claim that the model achieves robust, generalizable superiority rather than an artifact of data correlation.

    Authors: We agree that explicit details on the train/test split are essential and that patient- or slide-level partitioning is the gold standard to mitigate leakage in histopathology. Our experiments followed a standard 80/20 random per-image split for each public dataset, consistent with many prior transfer-learning studies on these collections. However, we recognize this does not fully eliminate the risk of correlation. In the revised manuscript we will (i) explicitly state the split ratios and random seed, (ii) discuss the leakage concern as a limitation, and (iii) where patient/slide metadata exists in the source datasets, re-run and report results under patient-level partitioning. These additions will appear in §3 and §4. revision: yes

  2. Referee: [§4] §4 (Results tables/figures): No ablation studies isolate the contribution of the Swin Transformer hierarchy versus the ResNet50 backbone, nor are repeated random splits, cross-validation folds, or statistical significance tests (e.g., paired t-tests or McNemar’s test against baselines) reported. Without these, the assertion of consistent outperformance over DenseNet, EfficientNet, and ViT variants cannot be evaluated as load-bearing evidence.

    Authors: We concur that ablation studies and statistical rigor are necessary to substantiate the hybrid architecture’s superiority. We will add (i) ablation experiments that systematically remove or replace the Swin Transformer hierarchy and the ResNet50 backbone, (ii) results from five independent random splits with mean ± standard deviation, and (iii) McNemar’s tests (and paired t-tests where appropriate) comparing DSVTLA against each baseline. These will be presented in revised §4 tables and text. revision: yes

  3. Referee: [§3] §3 (Dataset description): Dataset sizes, class balances, and the exact number of images per cancer type (original vs. segmented) are not quantified, nor is any multi-center or external hold-out validation mentioned. This omission makes it impossible to assess whether the near-perfect metrics reflect genuine robustness across the claimed “heterogeneous clinical imaging conditions.”

    Authors: We will expand §3 to report the precise number of images, class distributions, and original-versus-segmented counts for every cancer type. The datasets are drawn from well-known public repositories; we will cite their sources and note any documented multi-center provenance. We acknowledge the absence of an external hold-out set as a limitation and will add a dedicated paragraph discussing this point together with suggestions for future multi-center validation. revision: yes

Circularity Check

0 steps flagged

No circularity: empirical ML benchmark with independent test metrics

full rationale

The paper is an empirical study proposing a hybrid Swin Transformer + ResNet50 transfer-learning architecture and reporting its classification accuracies on multi-cancer histopathology datasets (original and segmented versions of breast, oral, lung-colon, kidney, and leukemia images). No mathematical derivation chain, predictive equations, or first-principles results are claimed; performance numbers (100% on lung-colon and segmented leukemia, 99.23% on breast) are direct outputs of a single unified training pipeline evaluated on held-out test splits. No self-definitional loops, fitted-input predictions, or load-bearing self-citations appear in the provided text. The central claims rest on observable benchmark comparisons against DenseNet, ResNet, ViT, and Swin baselines under identical preprocessing, which are falsifiable against external data and do not reduce to the model's own inputs by construction.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on standard computer-vision assumptions that combining global context from transformers with local features from CNNs improves classification on histopathology images; no new entities are postulated and no free parameters are explicitly named in the abstract.

free parameters (1)
  • Hyperparameters for transfer learning and fine-tuning
    Typical in such models (learning rate, batch size, epochs) but not enumerated; their values are fitted during training.
axioms (1)
  • domain assumption Hierarchical Swin Transformer captures long-range dependencies while ResNet50 extracts fine-grained local patterns
    Invoked in the description of the proposed framework integration.

pith-pipeline@v0.9.0 · 5626 in / 1303 out tokens · 64349 ms · 2026-05-10T16:34:42.855235+00:00 · methodology

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

43 extracted references · 28 canonical work pages

  1. [1]

    and Laversanne, Mathieu and Soerjomataram, Isabelle and Jemal, Ahmedin and Bray, Freddie , title =

    H. Sung et al., “Global Cancer Statistics 2020 : GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries,” vol. 71, no. 3, pp. 209 –249, 2021, doi: 10.3322/caac.21660

    work page doi:10.3322/caac.21660 2020

  2. [2]

    Cancer Statistics , 2020,

    R. L. Siegel and K. D. Miller, “Cancer Statistics , 2020,” vol. 0, no. 0, pp. 1 –24, 2020, doi: 10.3322/caac.21590

    work page doi:10.3322/caac.21590 2020

  3. [3]

    Artificial intelligence in digital pathology — new tools for diagnosis and precision oncology,

    K. A. Schalper and D. L. Rimm, “Artificial intelligence in digital pathology — new tools for diagnosis and precision oncology,” Nat. Rev. Clin. Oncol., doi: 10.1038/s41571-019-0252-y

    work page doi:10.1038/s41571-019-0252-y

  4. [4]

    Multi-omics data integration and drug screening of AML cancer using Generative Adversarial Network,

    S. Afroz, N. Isalam, M. A. Hahib, M. S. Reza, M. A. Alam, “Multi-omics data integration and drug screening of AML cancer using Generative Adversarial Network,” Methods, 226:138–50, 2024

    2024

  5. [5]

    Influence Function of Multiple Kernel Canonical Analysis to Identify Outliers in Imaging Genetics Data

    M. A. Alam, V. Calhoun, and Y. Wang, “Influence Function of Multiple Kernel Canonical Analysis to Identify Outliers in Imaging Genetics Data”

  6. [6]

    Higher-order Regularized Kernel CCA,

    M. A. Alam, “Higher-order Regularized Kernel CCA,” no. 2, pp. 1–4, 2013, doi: 10.1109/ICMLA.2013.76

    work page doi:10.1109/icmla.2013.76 2013

  7. [7]

    Deep residual learning for image recognition

    K. He and J. Sun, “Deep Residual Learning for Image Recognition,” 2016, doi: 10.1109/CVPR.2016.90

    work page doi:10.1109/cvpr.2016.90 2016

  8. [8]

    Densely Connected Convolutional Networks

    G. Huang, Z. Liu, L. Van Der Maaten, and K. Q. Weinberger, “Densely Connected Convolutional Networks,” 2017, doi: 10.1109/CVPR.2017.243

    work page doi:10.1109/cvpr.2017.243 2017

  9. [9]

    Kernel Choice for Unsupervised Kernel Methods ,

    M. A. Alam, “Kernel Choice for Unsupervised Kernel Methods ,” The Graduate University for Advanced Studies, Hayama, Kanagawa, Japan, 2014

    2014

  10. [10]

    Associations between genetically predicted plasma protein levels and Alzheimer ’ s disease risk : a study using genetic prediction models,

    J. Zhu et al., “Associations between genetically predicted plasma protein levels and Alzheimer ’ s disease risk : a study using genetic prediction models,” Alzheimers. Res. Ther., pp. 1–16, 2024, doi: 10.1186/s13195- 023-01378-4

    work page doi:10.1186/s13195- 2024

  11. [11]

    Swin Transformer : Hierarchical Vision Transformer using Shifted Windows,

    Y. Lin et al., “Swin Transformer : Hierarchical Vision Transformer using Shifted Windows,” pp. 1 –11

  12. [12]

    A Dataset for Breast Cancer Histopathological Image Classification,

    F. A. Spanhol, L. S. Oliveira, C. Petitjean, and L. Heutte, “A Dataset for Breast Cancer Histopathological Image Classification,” vol. 9294, no. c, pp. 1–8, 2015, doi: 10.1109/TBME.2015.2496264

    work page doi:10.1109/tbme.2015.2496264 2015

  13. [13]

    Robust kernel canonical correlation analysis to detect gene-gene co-associations: A case study in genetics,

    M.A. Alam. et al., “ Robust kernel canonical correlation analysis to detect gene-gene co-associations: A case study in genetics,” J Bioinform Comput Biol. 2019;17(4):1950028.2021

    work page arXiv 2019

  14. [14]

    On Breast Cancer Detection : An Application of Machine Learning Algorithms on the Wisconsin Diagnostic Dataset,

    A. F. M. Agarap, “On Breast Cancer Detection : An Application of Machine Learning Algorithms on the Wisconsin Diagnostic Dataset,” no. 1, pp. 7–11, 2018

    2018

  15. [15]

    Vision transformer and explainable transfer learning models for auto detection of kidney cyst , stone and tumor from CT ‑ radiography,

    N. Islam, M. Hasan, K. Hossain, G. R. Alam, Z. Uddin, and A. Soylu, “Vision transformer and explainable transfer learning models for auto detection of kidney cyst , stone and tumor from CT ‑ radiography,” Sci. Rep., pp. 1–14, 2022, doi: 10.1038/s41598-022-15634-4

    work page doi:10.1038/s41598-022-15634-4 2022

  16. [16]

    Lung and Colon Cancer Histopathological Image,

    D. Lc et al., “Lung and Colon Cancer Histopathological Image,” pp. 1–2

  17. [17]

    Data in brief Histopathological imaging database for oral cancer analysis,

    T. Yesmin, L. B. Mahanta, A. K. Das, and J. D. Sarma, “Data in brief Histopathological imaging database for oral cancer analysis,” Data Br., vol. 29, p. 105114, 2020, doi: 10.1016/j.dib.2020.105114

    work page doi:10.1016/j.dib.2020.105114 2020

  18. [18]

    Lung Cancer Detection and Classification from Chest CT Scans Using Machine Learning Techniques,

    M. Kashif and I. Abunadi, “Lung Cancer Detection and Classification from Chest CT Scans Using Machine Learning Techniques,” pp. 6–9, 2021

    2021

  19. [19]

    On Breast Cancer Detection : An Application of Machine Learning Algorithms on the Wisconsin Diagnostic Dataset,

    A. F. M. Agarap, “On Breast Cancer Detection : An Application of Machine Learning Algorithms on the Wisconsin Diagnostic Dataset,” no. 1, pp. 5–9, 2018

    2018

  20. [20]

    Detection of Cervical Cancer with Texture Analysis using Machine Learning Models,

    T. J. Nagalakshmi, N. Nalini, P. Jagadeesh, P. S. Bharathi, V. Amudha, and G. Ramkumar, “Detection of Cervical Cancer with Texture Analysis using Machine Learning Models,” 2022 Int. Conf. Adv. Comput. Commun. Appl. Informatics, pp. 1–6, doi: 10.1109/ACCAI53970.2022.9752550

    work page doi:10.1109/accai53970.2022.9752550 2022

  21. [21]

    Detection of Breast Cancer Using Machine Learning and Deep Learning Methods,

    D. Kale, “Detection of Breast Cancer Using Machine Learning and Deep Learning Methods,” 2022 3rd Int. Conf. Intell. Eng. Manag., pp. 1–6, 2022, doi: 10.1109/ICIEM54221.2022.9853080

    work page doi:10.1109/iciem54221.2022.9853080 2022

  22. [22]

    Study of Machine Learning Models for the Prediction and Detection of Lungs Cancer,

    A. Singh, P. R. Kumar, and R. Rastogi, “Study of Machine Learning Models for the Prediction and Detection of Lungs Cancer,” 2022 11th Int. Conf. Syst. Model. Adv. Res. Trends , pp. 1243 –1248, 2022, doi: 10.1109/SMART55829.2022.10047610

    work page doi:10.1109/smart55829.2022.10047610 2022

  23. [23]

    Kidney Cancer Detection using Deep Learning Models,

    K. Rajkumar, “Kidney Cancer Detection using Deep Learning Models,” 2023 7th Int. Conf. Trends Electron. Informatics, no. Icoei, pp. 1197–1203, 2023, doi: 10.1109/ICOEI56765.2023.10125589

    work page doi:10.1109/icoei56765.2023.10125589 2023

  24. [24]

    Detection and Classification of Various Types of Leukemia Using Image Processing , Transfer Learning and Ensemble Averaging Techniques,

    S. Rajeswari, C. S. Vasanth, C. Bhavana, and K. S. S. Chowdary, “Detection and Classification of Various Types of Leukemia Using Image Processing , Transfer Learning and Ensemble Averaging Techniques,” 2022 2nd Asian Conf. Innov. Technol., pp. 1–6, 2022, doi: 10.1109/ASIANCON55314.2022.9909377

    work page doi:10.1109/asiancon55314.2022.9909377 2022

  25. [25]

    Timely Diagnosis of Acute Lymphoblastic Leukemia Using Artificial Intelligence -Oriented Deep Learning Methods,

    S. Rezayi, N. Mohammadzadeh, H. Bouraghi, S. Saeedi, and A. Mohammadpour, “Timely Diagnosis of Acute Lymphoblastic Leukemia Using Artificial Intelligence -Oriented Deep Learning Methods,” vol. 2021, 2021, doi: 10.1155/2021/5478157

    work page doi:10.1155/2021/5478157 2021

  26. [26]

    Research Article A Semi -supervised Deep Learning Method for Cervical Cell Classification,

    S. Zhao, Y. He, J. Qin, and Z. Wang, “Research Article A Semi -supervised Deep Learning Method for Cervical Cell Classification,” vol. 2022, 2022, doi: 10.1155/2022/4376178

    work page doi:10.1155/2022/4376178 2022

  27. [27]

    A Machine Learning Approach to Diagnosing Lung and Colon Cancer Using a Deep Learning-Based Classification Framework,

    M. Masud, N. Sikder, A. Nahid, and A. K. Bairagi, “A Machine Learning Approach to Diagnosing Lung and Colon Cancer Using a Deep Learning-Based Classification Framework,” pp. 1–20, 2021

    2021

  28. [28]

    Intelligent Model for Brain Tumor Identification Using Deep Learning,

    A. H. Khan et al., “Intelligent Model for Brain Tumor Identification Using Deep Learning,” vol. 2022, 2022, doi: 10.1155/2022/8104054

    work page doi:10.1155/2022/8104054 2022

  29. [29]

    Boosting Breast Cancer Detection Using Convolutional Neural Network,

    S. A. Alanazi et al., “Boosting Breast Cancer Detection Using Convolutional Neural Network,” vol. 2021, 2021, doi: 10.1155/2021/5528622

    work page doi:10.1155/2021/5528622 2021

  30. [30]

    Automatic Detection and Segmentation of Colorectal Cancer with Deep Residual Convolutional Neural Network,

    A. Akilandeswari et al., “Automatic Detection and Segmentation of Colorectal Cancer with Deep Residual Convolutional Neural Network,” vol. 2022, 2022, doi: 10.1155/2022/3415603

    work page doi:10.1155/2022/3415603 2022

  31. [31]

    Review Article Deep Learning in Cancer Diagnosis and Prognosis Prediction : A Minireview on Challenges , Recent Trends , and Future Directions,

    A. Bin Tufail et al. , “Review Article Deep Learning in Cancer Diagnosis and Prognosis Prediction : A Minireview on Challenges , Recent Trends , and Future Directions,” vol. 2021, 2021, doi: 10.1155/2021/9025470

    work page doi:10.1155/2021/9025470 2021

  32. [32]

    Explainable lung cancer classification with ensemble transfer learning of VGG16 , Resnet50 and InceptionV3 using grad-cam,

    Y. K. S, J. J. Jeya, T. R. Mahesh, S. B. Khan, S. Alzahrani, and M. Alojail, “Explainable lung cancer classification with ensemble transfer learning of VGG16 , Resnet50 and InceptionV3 using grad-cam,” pp. 1– 19, 2024

    2024

  33. [33]

    Neuroscience Informatics The DCT -CNN-ResNet50 architecture to classify brain tumors with super-resolution , convolutional neural network , and the ResNet50,

    A. Deshpande, V. V Estrela, and P. Patavardhan, “Neuroscience Informatics The DCT -CNN-ResNet50 architecture to classify brain tumors with super-resolution , convolutional neural network , and the ResNet50,” Neurosci. Informatics, vol. 1, no. 4, p. 100013, 2021, doi: 10.1016/j.neuri.2021.100013

    work page doi:10.1016/j.neuri.2021.100013 2021

  34. [34]

    Classification of Breast Cancer Histopathological Images Using DenseNet and Transfer Learning,

    M. A. Wakili et al., “Classification of Breast Cancer Histopathological Images Using DenseNet and Transfer Learning,” vol. 2022, 2022, doi: 10.1155/2022/8904768

    work page doi:10.1155/2022/8904768 2022

  35. [35]

    Putra and F

    D. Putra and F. Axel, “ScienceDirect ScienceDirect Improving Warehouse Layout Effectiveness and Process Picking Improving Layout Effectiveness and Process Picking Efficiency Warehouse with the Discrete Event System Simulation Approach Efficiency with the Discrete Event System Simulation Approach Seventh Information Systems International Conference ( ISICO...

    work page doi:10.1016/j.procs.2024.03.182 2023

  36. [36]

    Cancer image classification based on DenseNet model Cancer image classification based on DenseNet model,

    C. Series, “Cancer image classification based on DenseNet model Cancer image classification based on DenseNet model,” 2020, doi: 10.1088/1742-6596/1651/1/012143

    work page doi:10.1088/1742-6596/1651/1/012143 2020

  37. [37]

    Rethinking the Inception Architecture for Computer Vision,

    C. Szegedy, V. Vanhoucke, and J. Shlens, “Rethinking the Inception Architecture for Computer Vision,” 2014

    2014

  38. [38]

    Integration of feature enhancement technique in Google inception network for breast cancer detection and classification,

    W. S. Admass, Y. Y. Munaye, and A. O. Salau, “Integration of feature enhancement technique in Google inception network for breast cancer detection and classification,” pp. 1 –30, 2024

    2024

  39. [39]

    Revolutionizing breast ultrasound diagnostics with EfficientNet ‑ B7 and Explainable,

    M. Latha, P. S. Kumar, R. R. Chandrika, T. R. Mahesh, V. V. Kumar, and S. Guluwadi, “Revolutionizing breast ultrasound diagnostics with EfficientNet ‑ B7 and Explainable,” BMC Med. Imaging , 2024, doi: 10.1186/s12880-024-01404-3

    work page doi:10.1186/s12880-024-01404-3 2024

  40. [40]

    Swin Transformer V2 : Scaling Up Capacity and Resolution,

    Z. Liu et al., “Swin Transformer V2 : Scaling Up Capacity and Resolution,” pp. 12009–12019

  41. [41]

    SwinCup : Cascaded swin transformer for histopathological structures segmentation in colorectal cancer,

    U. Zidan, M. Medhat, and M. M. Abdelsamea, “SwinCup : Cascaded swin transformer for histopathological structures segmentation in colorectal cancer,” vol. 216, no. December 2022, 2023

    2022

  42. [42]

    Learning Schizophrenia Imaging Genetics Data Via Multiple Kernel Canonical Correlation Analysis. ,

    O. Richfield, M. A. Alam, V. D . Calhoun, Y.P. Wang, “Learning Schizophrenia Imaging Genetics Data Via Multiple Kernel Canonical Correlation Analysis. ,” BIBM 2016, Proceeding of the IEEE International Conference on Bioinformatics and Biomedicine; Shenzhen, China: BIBM; 2016

    2016

  43. [43]

    A Systematic Approach for MRI Brain Tumor Localization and Segmentation Using Deep Learning and Active Contouring,

    S. R. Gunasekara, H. N. T. K. Kaldera, and M. B. Dissanayake, “A Systematic Approach for MRI Brain Tumor Localization and Segmentation Using Deep Learning and Active Contouring,” vol. 2021, 2021, doi: 10.1155/2021/6695108

    work page doi:10.1155/2021/6695108 2021