Pith. sign in

REVIEW 4 major objections 5 minor 1 cited by

An Attention-Guided Deep Learning Approach for Classifying 39 Skin Lesion Types

T0 review · 4 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read The paper claims that adding CBAM to a Vision Transformer yields 93.46% accuracy in classifying 39 skin lesion types.

desk verdict Useful 39-class skin lesion benchmark, but the headline accuracy is unverified and likely optimistic until de-duplication and split methodology are fixed. read the letter →

arxiv 2501.05991 v1 pith:5J72IY6T submitted 2025-01-10 eess.IV cs.CVcs.LG

Sauda Adiv Hanum , Ashim Dey , Muhammad Ashad Kabir This is my paper · ORCID
classification eess.IVcs.CVcs.LG
keywords skinlesionclassificationVisionTransformerCBAMattentionmechanismmulti-classdatasetcurationtransferlearningdermatologyimaging
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 claims that a Vision Transformer augmented with a convolutional block attention module can classify 39 types of skin lesions from photographs, reaching 93.46% accuracy on a held-out test set. The authors build a balanced dataset by merging five public skin-image collections and capping every class at 130 images, then compare five deep models with and without two attention modules. Their central result is that ViT+CBAM beats the plain ViT, all CNN baselines, and every ECA-augmented variant on accuracy, precision, recall, F1-score, and specificity. If the result holds, attention-guided transformers become a credible tool for broad multi-class dermatology triage rather than the usual small set of lesion types.

What carries the argument

The load-bearing mechanism is CBAM, the Convolutional Block Attention Module, applied after the Vision Transformer encoder. CBAM first computes a channel attention map from average- and max-pooled features through a shared MLP, then a spatial attention map from channel-pooled features through a 7×7 convolution, applying the two maps sequentially to refine the transformer's feature representation. It is the component that separates the proposed model from the plain ViT baseline and produces the claimed gain from 91.79% to 93.46%. The dataset curation — merging five public sources, capping each class at 130 images, and using a 70/15/15 split — is the supporting machinery that makes the 39-class comparison possible.

What would settle it

Re-run the ViT+CBAM evaluation after removing near-duplicate images and splitting by patient or lesion identity rather than by image; if accuracy drops substantially below 93.46%, the reported test result is inflated by data leakage. A simpler check is to search the training and test sets for identical or near-identical images across the five source collections.

Watch

Extended reading notes

Core claim

The paper's central discovery is an empirical ranking of model configurations on a curated 39-class dataset. Under a 70/15/15 split with 130 images per class, the Vision Transformer with CBAM achieves 93.46% accuracy, 94% precision, 93% recall, 93% F1-score, and 93.67% specificity, outperforming the plain ViT (91.79%) and every other baseline and attention variant tested. The authors attribute this improvement to CBAM's sequential channel and spatial attention, which refines the global features produced by the transformer encoder so the model concentrates on diagnostically relevant regions. The overall ROC AUC is reported as 0.99. The same claim appears in the abstract and the conclusion, so the paper is trying to establish that this particular attention-guided configuration is the strongest among those compared.

Load-bearing premise

The whole result rests on the assumption that the 39 merged classes are correctly labelled and that the random split of individual images, rather than patients or unique lesions, does not put the same or near-identical pictures into both training and testing.

Editorial extensions

If this is right

  • On the curated 39-class dataset, ViT+CBAM outperforms MobileNetV2, Xception, InceptionV3, EfficientNetB1, and the plain Vision Transformer under the same training protocol.
  • Adding either ECA or CBAM improves every baseline model relative to the same architecture without attention, so the paper supports attention modules as a general enhancement for skin-lesion classifiers.
  • Class-wise results show perfect accuracy on 18 of 39 classes and an overall ROC AUC of 0.99, indicating the model separates most classes cleanly.
  • The released dataset and code let other groups reproduce the 39-class comparison and test new architectures against the same benchmark.
  • The known weak spot, Leprosy Lepromatous at 50% accuracy, is explicitly attributed to visual similarity with Molluscum Contagiosum, so the paper identifies where image-only classification reaches its limit.

Reading between the lines

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

  • Because the split is image-level and no de-duplication or patient-level separation is reported, the 93.46% figure is likely optimistic; re-evaluating on a deduplicated or patient-level split would reveal how much of the score comes from near-duplicate images.
  • The balanced 130-image-per-class design removes natural prevalence, so the accuracy number does not directly transfer to real-world screening where some lesion types are far rarer than others.
  • The Leprosy Lepromatous confusion with Molluscum Contagiosum suggests the same architecture could gain from clinical metadata or histopathology, as the authors propose in their future-work section, rather than from more image augmentation alone.
  • The same 'transformer plus convolutional attention' recipe could be tested on other fine-grained medical image tasks where global context and local boundary detail both matter.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The paper proposes an attention-guided deep learning framework for classifying 39 types of skin lesions. It curates a dataset by merging five public sources (ISIC 2019, Atlas Dermatology, HAM10000, MSLD 2.0, Dermnet-based collections), caps each class at 130 images, and splits the data 70/15/15 into train/validation/test sets of 90/20/20 images per class. Five architectures (MobileNetV2, Xception, InceptionV3, EfficientNetB1, Vision Transformer) are evaluated in baseline form and with two attention modules (ECA, CBAM). The authors report that ViT+CBAM achieves the best results: 93.46% accuracy, 94% precision, 93% recall, 93% F1-score, and 93.67% specificity. The paper includes class-wise accuracy, confusion matrix, ROC curves, and a GitHub link for code and data.

Significance. If the results are valid, the paper would provide a useful 39-class benchmark for skin lesion classification and evidence that attention mechanisms (especially CBAM) improve a Vision Transformer baseline on such a multi-class task. The dataset integration effort is potentially valuable, and the inclusion of code and dataset links supports reproducibility. However, the headline accuracy and the comparative conclusion are only meaningful if the training/test split is leakage-free and the best model is not selected on the same test set used for evaluation; these issues are the central validity concerns of the paper.

major comments (4)
  1. [3.1] Section 3.1 describes merging five public datasets and capping each class at 130 images, but it does not describe any de-duplication or patient/source-level split. Because HAM10000 images are also part of ISIC 2019, and Dermnet images appear in multiple Kaggle collections, the same lesion photograph (or near-duplicates from the same lesion) can be present in both training and test splits. This would inflate the reported 93.46% accuracy and could change the relative ranking of models. Please specify a de-duplication procedure (e.g., exact-duplicate removal via perceptual hashing, and near-duplicate grouping by lesion or patient identifiers) and provide the exact split (file lists and random seed) so the train/test independence can be audited.
  2. [4.3, Table 4] The test set of 20 images per class is used both to compare all 15 model configurations and to select the 'best' model (ViT+CBAM) in Table 4. Selecting the best configuration on the same data used for final reporting introduces optimistic bias, so the 93.46% accuracy is not an unbiased estimate of the selected model's generalization performance. The authors should use a validation set for model selection and a separately held-out test set for the final evaluation, or report repeated runs with different seeds and confidence intervals for the comparisons.
  3. [4.3, Table 5] Each class-wise accuracy in Table 5 is computed from only 20 test images. For example, the 100% accuracies for 18 classes correspond to 20/20 correct, and the 50% for Leprosy Lepromatous corresponds to 10/20; these values have very wide binomial confidence intervals, so statements such as 'notable successes with 100% accuracy' are statistically fragile. The overall accuracy of 93.46% (729/780) also lacks a confidence interval. Please report exact counts and confidence intervals, and temper the class-wise claims accordingly.
  4. [3.3, 4.1] The Vision Transformer hyperparameters are not reported: there is no learning rate, batch size, number of epochs, optimizer, patch size, or embedding dimension. The other four models are given such details, so this omission prevents reproducibility of the central ViT+CBAM result. In addition, the GitHub repository is cited without a commit hash or exact train/test file lists, and the split seed is not provided. Please supply complete hyperparameters and a versioned, auditable data split.
minor comments (5)
  1. [3.4.2, Eqs. (4)-(6)] The spatial attention equations use F as their input and denote the output as M_c(F), but the spatial attention module is described as operating on the channel-refined feature F' and should produce M_s(F'). The notation should be corrected to avoid confusion.
  2. [Tables 3 and 5] The class numbering and naming are inconsistent: in Table 3, '34. Psoriasis' appears before '33. Scabies Lyme Disease', and Table 5 lists 'Lerva Migrans' for what is elsewhere 'Larva Migrans'. These should be unified and corrected.
  3. [Figure 1] Figure 1 lists SGD, Adam, and RectifiedAdam as optimizers, but the text never specifies which optimizer is used for each model or configuration; please clarify.
  4. [5.1] The statement 'The overall area under the curve (AUC) score of 0.99' does not specify whether this is macro-average, micro-average, or per-class; Figure 9 would benefit from a clear definition in the caption or text.
  5. [References [4] and [17]] References [4] and [17] appear to be the same paper (Choudhary et al., Chemometrics and Intelligent Laboratory Systems 2022) cited twice in different contexts; please consolidate.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: all reported results are direct empirical measurements on an independent curated dataset; no derivation reduces to its own inputs.

full rationale

This paper is an empirical benchmarking study rather than a mathematical derivation. The central claim — that ViT+CBAM reaches 93.46% accuracy on the curated 39-class dataset — is a measured test-set result reported in Table 4, not a quantity derived from a fitted parameter or from a self-referential definition. The attention modules (ECA, CBAM) are taken from cited external sources ([45], [46]) and inserted into standard pretrained architectures; there is no step in which a model's success is defined in terms of the outcome it is claimed to predict. The dataset is assembled from five public sources with per-class capping, and the 70/15/15 split is described in Section 3.1; while the absence of de-duplication and patient-level splitting is a legitimate validity or leakage risk that the paper does not address, that is an experimental-design concern, not circular reasoning. No load-bearing self-citation appears: references [10] and [41] are prior merging works by other authors, and the paper does not invoke a uniqueness theorem or a prior claim by the same authors to rule out alternatives. The selection of ViT+CBAM as 'best' is based on the same test set used for comparison, which raises a mild selection-bias concern, but the accuracy numbers themselves are independent measurements and do not reduce by construction to the inputs of the study. Consequently, there is no specific step that qualifies as circular under the stated criteria, and the honest finding is no significant circularity (score 0).

Assumptions & free parameters 6 free parameters · 4 assumptions · 0 invented entities

The central result is an empirical benchmark, so the ledger lists the main hand-chosen design choices and unstated assumptions that the reported accuracy depends on. There are no invented entities. The most consequential items are the 130-image cap, the 70/15/15 split, and the absence of de-duplication, which together determine how much signal the test set contains and how likely the accuracy is to generalize.

free parameters (6)
  • Per-class image cap = 130
    All classes are downsampled to 130 images to balance the dataset before splitting; this choice affects the amount of data per class and the reported accuracy.
  • Train/validation/test split = 70/15/15
    The random split assigns 90 training, 20 validation, and 20 test images per class; with only 20 test images per class, the reported accuracy has wide uncertainty.
  • Learning rate = 0.001
    All five models use a fixed learning rate of 0.001; this hand-chosen value influences convergence and final accuracy.
  • Batch size = 8
    All models use batch size 8; chosen for the Kaggle environment and affects training dynamics.
  • Training epochs per model = MobileNetV2: 40, Xception: 8, InceptionV3: 8, EfficientNetB1: 70, ViT: not reported
    Epoch counts differ across models and were apparently chosen to optimize each baseline; the ViT epoch count is not stated.
  • Augmentation parameters = Not specified
    Height/width shift, rotation, zoom, shear, and flips are applied but their ranges are not quantified, making exact reproduction impossible.
assumptions (4)
  • domain assumption The five public source datasets provide correct, mutually exclusive labels for all 39 classes.
    Section 3.1 merges Atlas Dermatology, ISIC 2019, HAM10000, MSLD 2.0, Dermnet, and other Kaggle sources without a label-verification or de-duplication step; if labels are wrong or overlapping, the accuracy numbers lose meaning.
  • domain assumption The random 70/15/15 split creates independent train and test sets.
    No patient-level or source-level split is used; near-duplicates between Dermnet and Atlas images at different sources could appear in both training and test sets, inflating accuracy.
  • domain assumption The reported metrics (accuracy, precision, recall, F1, specificity) are computed correctly.
    The equations in Section 4.2 are standard, but no confusion matrix or per-class counts for the aggregate metrics are shown, and rounding or macro/micro averaging are not specified.
  • ad hoc to paper Attention modules were implemented correctly in each architecture.
    The integration points of ECA and CBAM into each of the five architectures are not described in enough detail; the ViT+CBAM figure suggests post-transformer insertion, but the code is not in the paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of An Attention-Guided Deep Learning Approach for Classifying 39 Skin Lesion Types." pith.science (2026). https://pith.science/paper/5J72IY6T

@misc{pith2026250105991,
  author       = {Pith},
  title        = {Pith review of: An Attention-Guided Deep Learning Approach for Classifying 39 Skin Lesion Types},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5J72IY6T}},
  note         = {Machine review of arXiv:2501.05991}
}
read the original abstract

The skin, as the largest organ of the human body, is vulnerable to a diverse array of conditions collectively known as skin lesions, which encompass various dermatoses. Diagnosing these lesions presents significant challenges for medical practitioners due to the subtle visual differences that are often imperceptible to the naked eye. While not all skin lesions are life-threatening, certain types can act as early indicators of severe diseases, including skin cancers, underscoring the critical need for timely and accurate diagnostic methods. Deep learning algorithms have demonstrated remarkable potential in facilitating the early detection and prognosis of skin lesions. This study advances the field by curating a comprehensive and diverse dataset comprising 39 categories of skin lesions, synthesized from five publicly available datasets. Using this dataset, the performance of five state-of-the-art deep learning models -- MobileNetV2, Xception, InceptionV3, EfficientNetB1, and Vision Transformer - is rigorously evaluated. To enhance the accuracy and robustness of these models, attention mechanisms such as the Efficient Channel Attention (ECA) and the Convolutional Block Attention Module (CBAM) are incorporated into their architectures. Comprehensive evaluation across multiple performance metrics reveals that the Vision Transformer model integrated with CBAM outperforms others, achieving an accuracy of 93.46%, precision of 94%, recall of 93%, F1-score of 93%, and specificity of 93.67%. These results underscore the significant potential of the proposed system in supporting medical professionals with accurate and efficient prognostic tools for diagnosing a broad spectrum of skin lesions. The dataset and code used in this study can be found at https://github.com/akabircs/Skin-Lesions-Classification.

Figures

Figures reproduced from arXiv: 2501.05991 by the authors.

Figure 1
Figure 1. Methodological Flow of the Proposed Skin Disease Classification Approach [PITH_FULL_IMAGE:figures/full_fig_p007_1.png] view at source ↗
Figure 2
Figure 2. Efficient channel attention module [45] 3.4.2. Convolutional Block Attention Module (CBAM) CBAM is a lightweight yet powerful attention mechanism designed to enhance feature extraction in deep learn￾ing models [46]. CBAM combines channel attention and spatial attention modules in a sequential manner to refine feature representations adaptively, as illustrated in [PITH_FULL_IMAGE:figures/full_fig_p013_2.png] view at source ↗
Figure 3
Figure 3. Convolutional block attention module domain. Both average-pooling and max-pooling operations are used to generate global descriptors, which are then passed through a shared network consisting of a multi-layer perceptron (MLP) with one hidden layer. The outputs of these two pooling operations are merged to produce the final channel attention map: Mc(F) = σ [PITH_FULL_IMAGE:figures/full_fig_p014_3.png] view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: Channel attention module, redrawn based on [PITH_FULL_IMAGE:figures/full_fig_p014_4.png]
Figure 5
Figure 5. Figure 5: Spatial Attention Module, redrawn based on [ [PITH_FULL_IMAGE:figures/full_fig_p015_5.png]
Figure 6
Figure 6. Figure 6: Vision transformer with CBAM 4. Experimental Evaluation 4.1. Experiment Setup The experiments conducted in this study were implemented using the Python programming language and exe￾cuted on Kaggle’s notebook environment. Kaggle provides access to powerful computational…
Figure 7
Figure 7. Figure 7: Confusion Matrix of proposed model balanced representation across most classes, it is likely that the diversity of training samples for Leprosy Lepromatous is insufficient to capture its unique features. This lack of diversity may hinder the model’s ability to generali…
Figure 8
Figure 8. Figure 8: Visual similarity between (a) Molluscum Contagiosum and (b) Leprosy Lepromatous. [PITH_FULL_IMAGE:figures/full_fig_p021_8.png]
Figure 9
Figure 9. Figure 9: ROC curves for 39 classes (GANs), have been shown to effectively augment datasets and introduce greater variability in lesion appearance [50]. While this study explored several baseline and attention-guided deep learning models, future research could focus on enhancing…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Toward Accessible Dermatology: Skin Lesion Classification Using Deep Learning Models on Mobile-Acquired Images

    cs.CV 2025-09 conditional novelty 6.0 of 10

    On a new 51-class dataset of everyday skin photos, the Swin Transformer outperforms eight CNN and transformer baselines, reaching about 81 percent accuracy.

Reference graph

Works this paper leans on

58 extracted references · 45 canonical work pages · cited by 1 Pith paper

  1. [1]

    Accessed on 18 May 2024

    Skin disorder, https://www.britannica.com/science/human-skin-disease , 2024. Accessed on 18 May 2024

    work page 2024

  2. [2]

    Accessed on 18 July 2023

    Abnormal behaviour of skin, https://my.clevelandclinic.org/health/diseases/15818-skin-cancer , 2024. Accessed on 18 July 2023

    work page 2024

  3. [3]

    Accessed on 18 May 2023

    Statistics for skin disease, https://jamanetwork.com/journals/jamadermatology/fullarticle/2790344, 2024. Accessed on 18 May 2023

    work page arXiv 2024

  4. [4]

    Choudhary, J

    P. Choudhary, J. Singhai, J. Yadav, Skin lesion detection based on deep neural networks, Chemometrics and Intelligent Laboratory Systems 230 (2022) 104659

    work page 2022

  5. [5]

    M. A. Al-Masni, D.-H. Kim, T.-S. Kim, Multiple skin lesions diagnostics via integrated deep convolutional networks for segmentation and classification, Computer methods and programs in biomedicine 190 (2020) 105351

    work page 2020

  6. [6]

    Accessed on 18 May 2023

    Cause of skin disease, https://www.healthline.com/health/skin-disorders#causes, 2024. Accessed on 18 May 2023

    work page 2024

  7. [7]

    T. G. Debelee, Skin lesion classification and detection using machine learning techniques: A systematic review, Diagnostics 13 (2023) 3147

    work page 2023

  8. [8]

    Arora, A

    G. Arora, A. K. Dubey, Z. A. Ja ffery, A. Rocha, A comparative study of fourteen deep learning networks for multi skin lesion classification (mslc) on unbalanced data, Neural Computing and Applications 35 (2023) 7989–8015

    work page 2023

Show all 58 references
  1. [9]

    M. A. Kassem, K. M. Hosny, R. Dama ˇseviˇcius, M. M. Eltoukhy, Machine learning and deep learning methods for skin lesion classification and diagnosis: a systematic review, Diagnostics 11 (2021) 1390

    2021

  2. [10]

    Rafay, W

    A. Rafay, W. Hussain, E fficientskindis: An e fficientnet-based classification model for a large manually curated dataset of 31 skin diseases, Biomedical Signal Processing and Control 85 (2023) 104869

    2023

  3. [11]

    Accessed on 24 May 2024

    Skin disease2, https://www.kaggle.com/datasets/arsanyfawzy/skin-diseases, 2024. Accessed on 24 May 2024. xxiii

    2024

  4. [12]

    Accessed on 18 May 2024

    Skin lesions classification, https://www.kaggle.com/datasets/ahmedxc4/skin-ds, 2024. Accessed on 18 May 2024

    2024

  5. [13]

    Accessed on 18 May 2024

    Dataset skin, https://www.kaggle.com/datasets/onurinan1/dateset-23-skin , 2024. Accessed on 18 May 2024

    2024

  6. [14]

    Accessed on 24 May 2024

    Skindiseasess, https://www.kaggle.com/datasets/ascanipek/skin-diseases, 2024. Accessed on 24 May 2024

    2024

  7. [15]

    Prasanna Kumar, K

    R. Prasanna Kumar, K. Venkatraman, C. Jawahar, B. Harish, S. Bharathraj, K. Mukesh, Attention-guided residual network for skin lesion classification using deep reinforcement learning, in: 2023 International Conference on Integrated Intelligence and Communication Systems (ICIIC...

    2023

  8. [16]

    N. C. Codella, D. Gutman, M. E. Celebi, B. Helba, M. A. Marchetti, S. W. Dusza, A. Kalloo, K. Liopyris, N. Mishra, H. Kittler, et al., Skin lesion analysis toward melanoma detection: A challenge at the 2017 international symposium on biomedical imaging (isbi), hosted by the in...

    2017

  9. [17]

    Choudhary, J

    P. Choudhary, J. Singhai, J. Yadav, Skin lesion detection based on deep neural networks, Chemometrics and Intelligent Laboratory Systems 230 (2022) 104659. doi:10.1016/j.chemolab.2022.104659

    2022

  10. [18]

    Codella, V

    N. Codella, V . Rotemberg, P. Tschandl, M. E. Celebi, S. Dusza, D. Gutman, B. Helba, A. Kalloo, K. Liopyris, M. Marchetti, et al., Skin lesion analysis toward melanoma detection 2018: A challenge hosted by the international skin imaging collaboration (isic), arXiv preprint arX...

    2019 arXiv

  11. [19]

    Young, G

    K. Young, G. Booth, B. Simpson, R. Dutton, S. Shrapnel, Deep neural network or dermatologist?, Springer, 2019, pp. 48–55

    2019

  12. [20]

    Tschandl, C

    P. Tschandl, C. Rosendahl, H. Kittler, The ham10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions, Scientific data 5 (2018) 1–9

    2018

  13. [21]

    Chowdhury, A

    T. Chowdhury, A. R. Bajwa, T. Chakraborti, J. Rittscher, U. Pal, Exploring the correlation between deep learned and clinical features in melanoma detection, Springer, 2021, pp. 3–17

    2021

  14. [22]

    P. N. Srinivasu, J. G. SivaSai, M. F. Ijaz, A. K. Bhoi, W. Kim, J. J. Kang, Classification of skin disease using deep learning neural networks with mobilenet v2 and lstm, Sensors 21 (2021) 2852

    2021

  15. [23]

    M. U. Ali, M. Khalid, H. Alshanbari, A. Zafar, S. W. Lee, Enhancing skin lesion detection: A multistage multiclass convolutional neural network-based framework, Bioengineering 10 (2023). doi:10.3390/bioengineering10121430

    2023 doi

  16. [24]

    Ayas, Multiclass skin lesion classification in dermoscopic images using swin transformer model, Neural Computing and Applications 35 (2023) 6713–6722

    S. Ayas, Multiclass skin lesion classification in dermoscopic images using swin transformer model, Neural Computing and Applications 35 (2023) 6713–6722

    2023

  17. [25]

    Combalia, N

    M. Combalia, N. C. Codella, V . Rotemberg, B. Helba, V . Vilaplana, O. Reiter, C. Carrera, A. Barreiro, A. C. Halpern, S. Puig, et al., Bcn20000: Dermoscopic lesions in the wild, arXiv preprint arXiv:1908.02288 (2019)

    2019 arXiv

  18. [26]

    Desale, P

    R. Desale, P. Patil, An efficient multi-class classification of skin cancer using optimized vision transformer, Medical & Biological Engineering & Computing 62 (2024) 773–789

    2024

  19. [27]

    D. K. Saha, A. M. Joy, A. Majumder, Yotransvit: A transformer and cnn method for predicting and classifying skin diseases using segmen- tation techniques, Informatics in Medicine Unlocked 47 (2024) 101495

    2024

  20. [28]

    M. A. Kassem, K. M. Hosny, M. M. Fouad, Skin lesions classification into eight classes for isic 2019 using deep convolutional neural network and transfer learning, IEEE Access 8 (2020) 114822–114832. doi: 10.1109/ACCESS.2020.3003890

    2020

  21. [29]

    Y . S. Alsahafi, M. A. Kassem, K. M. Hosny, Skin-net: a novel deep residual network for skin lesions classification using multilevel feature extraction and cross-channel correlation with detection of outlier, Journal of Big Data 10 (2023) 105

    2023

  22. [30]

    X. Zeng, Z. Ji, H. Zhang, R. Chen, Q. Liao, J. Wang, T. Lyu, L. Zhao, Dsp-kd: dual-stage progressive knowledge distillation for skin disease classification, Bioengineering 11 (2024) 70

    2024

  23. [31]

    Nakai, Y .-W

    K. Nakai, Y .-W. Chen, X.-H. Han, Enhanced deep bottleneck transformer model for skin lesion classification, Biomedical Signal Processing and Control 78 (2022) 103997

    2022

  24. [32]

    D. de A. Rodrigues, R. F. Ivo, S. C. Satapathy, S. Wang, J. Hemanth, P. P. R. Filho, A new approach for classification skin lesion based on transfer learning, deep learning, and iot system, Pattern Recognition Letters 136 (2020) 8–15. doi:10.1016/j.patrec.2020.05.019

    2020 doi

  25. [33]

    Mendonc ¸a, M

    T. Mendonc ¸a, M. Celebi, T. Mendonca, J. Marques, Ph2: A public database for the analysis of dermoscopic images, Dermoscopy image analysis 2 (2015). xxiv

    2015

  26. [34]

    Rezaee, H

    K. Rezaee, H. G. Zadeh, Self-attention transformer unit-based deep learning framework for skin lesions classification in smart healthcare, Discover Applied Sciences 6 (2024) 3

    2024

  27. [35]

    Ahmad, J

    I. Ahmad, J. Amin, M. I. Lali, F. Abbas, M. I. Sharif, A novel deeplabv3 + and vision-based transformer model for segmentation and classification of skin lesions, Biomedical Signal Processing and Control 92 (2024) 106084

    2024

  28. [36]

    Rotemberg, N

    V . Rotemberg, N. Kurtansky, B. Betz-Stablein, L. Caffery, E. Chousakos, N. Codella, M. Combalia, S. Dusza, P. Guitera, D. Gutman, et al., A patient-centric dataset of images and metadata for identifying melanomas using clinical context, Scientific data 8 (2021) 34

    2021

  29. [37]

    Nagadevi, K

    D. Nagadevi, K. Suman, P. S. Lakshmi, An enhanced skin lesion detection and classification model using hybrid convolution-based ensemble learning model, Research on Biomedical Engineering (2024) 1–26

    2024

  30. [38]

    M. A. Khan, K. Muhammad, M. Sharif, T. Akram, S. Kadry, Intelligent fusion-assisted skin lesion localization and classification for smart healthcare, Neural Computing and Applications 36 (2024) 37–52

    2024

  31. [39]

    Gutman, N

    D. Gutman, N. C. Codella, E. Celebi, B. Helba, M. Marchetti, N. Mishra, A. Halpern, Skin lesion analysis toward melanoma detection: A challenge at the international symposium on biomedical imaging (isbi) 2016, hosted by the international skin imaging collaboration (isic), arXi...

    2016 arXiv

  32. [40]

    Accessed on 24 May 2024

    Atlas dermatology, https://www.atlasdermatologico.com.br/, 2024. Accessed on 24 May 2024

    2024

  33. [41]

    Sadik, A

    R. Sadik, A. Majumder, A. A. Biswas, B. Ahammad, M. M. Rahman, An in-depth analysis of convolutional neural network architectures with transfer learning for skin disease diagnosis, Healthcare Analytics 3 (2023) 100143

    2023

  34. [42]

    Accessed on 24 May 2024

    Dermnet, https://dermnet.com/, 2024. Accessed on 24 May 2024

    2024

  35. [43]

    S. N. Ali, M. T. Ahmed, T. Jahan, J. Paul, S. S. Sani, N. Noor, A. N. Asma, T. Hasan, A web-based mpox skin lesion detection system using state-of-the-art deep learning models considering racial diversity, Biomedical Signal Processing and Control 98 (2024) 106742

    2024

  36. [44]

    Accessed on 18 May 2023

    Dermnet dataset, https://dermnet.com/, 2024. Accessed on 18 May 2023

    2024

  37. [45]

    Q. Wang, B. Wu, P. Zhu, P. Li, W. Zuo, Q. Hu, Eca-net: E fficient channel attention for deep convolutional neural networks, in: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2020, pp. 11534–11542

    2020

  38. [46]

    S. Woo, J. Park, J.-Y . Lee, I. S. Kweon, Cbam: Convolutional block attention module, in: Proceedings of the European conference on computer vision (ECCV), 2018, pp. 3–19

    2018

  39. [47]

    M. Yin, Z. Chen, C. Zhang, A cnn-transformer network combining cbam for change detection in high-resolution remote sensing images, Remote Sensing 15 (2023). doi:10.3390/rs15092406

    2023 doi

  40. [48]

    van Hooij, E

    A. van Hooij, E. M. Tjon Kon Fat, S. J. van den Eeden, L. Wilson, M. Batista da Silva, C. G. Salgado, J. S. Spencer, P. L. Corstjens, A. Geluk, Field-friendly serological tests for determination of m. leprae-specific antibodies, Scientific reports 7 (2017) 8868

    2017

  41. [49]

    Alshahrani, M

    M. Alshahrani, M. Al-Jabbar, E. M. Senan, I. A. Ahmed, J. A. Mohammed Saif, Analysis of dermoscopy images of multi-class for early detection of skin lesions by hybrid systems based on integrating features of cnn models, Plos one 19 (2024) 1–36. doi: 10.1371/journal. pone.0298305

    2024 doi

  42. [50]

    R. R. Mekala, F. Pahde, S. Baur, S. Chandrashekar, M. Diep, M. Wenzel, E. L. Wisotzky, G. ¨Umit Yolcu, S. Lapuschkin, J. Ma, P. Eis- ert, M. Lindvall, A. Porter, W. Samek, Synthetic generation of dermatoscopic images with gan and closed-form factorization, 2024. arXiv:2410.05114

    2024 arXiv

  43. [51]

    Yunusa, S

    H. Yunusa, S. Qin, A. H. A. Chukkol, A. A. Yusuf, I. Bello, A. Lawan, Exploring the synergies of hybrid cnns and vits architectures for computer vision: A survey, 2024. arXiv:2402.02941

    2024

  44. [52]

    N. Luo, X. Zhong, L. Su, Z. Cheng, W. Ma, P. Hao, Artificial intelligence-assisted dermatology diagnosis: from unimodal to multimodal, Computers in Biology and Medicine (2023) 107413. doi:10.1016/j.compbiomed.2023.107413

    2023

  45. [53]

    S. Yan, Z. Yu, C. Primiero, C. Vico-Alonso, Z. Wang, L. Yang, P. Tschandl, M. Hu, G. Tan, V . Tang, A. B. Ng, D. Powell, P. Bonning- ton, S. See, M. Janda, V . Mar, H. Kittler, H. P. Soyer, Z. Ge, A general-purpose multimodal foundation model for dermatology, 2024. arXiv:2410.15038

    2024 arXiv

  46. [54]

    B. H. Van der Velden, H. J. Kuijf, K. G. Gilhuijs, M. A. Viergever, Explainable artificial intelligence (xai) in deep learning-based medical image analysis, Medical Image Analysis 79 (2022) 102470. doi: 10.1016/j.media.2022.102470. xxv

    2022

  47. [55]

    E. Tjoa, C. Guan, A survey on explainable artificial intelligence (xai): Toward medical xai, IEEE transactions on neural networks and learning systems 32 (2020) 4793–4813. doi:10.1109/TNNLS.2020.3027314

    2020

  48. [56]

    Ghnemat, S

    R. Ghnemat, S. Alodibat, Q. Abu Al-Haija, Explainable artificial intelligence (xai) for deep learning based medical imaging classification, Journal of Imaging 9 (2023) 177. doi:10.3390/jimaging9090177

    2023 doi

  49. [57]

    Tmamna, E

    J. Tmamna, E. B. Ayed, R. Fourati, M. Gogate, T. Arslan, A. Hussain, M. B. Ayed, Pruning deep neural networks for green energy-e fficient models: A survey, Cognitive Computation (2024) 1–22. doi:10.1007/s12559-024-10313-0

    2024 doi

  50. [58]

    Mishra, H

    R. Mishra, H. Gupta, Transforming large-size to lightweight deep neural networks for iot applications, ACM Computing Surveys 55 (2023) 1–35. doi:10.1145/3570955. xxvi

    2023 doi

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.