Pith. sign in

REVIEW 5 major objections 6 minor 45 references

Hybrid Interpretable Deep Learning Framework for Skin Cancer Diagnosis: Integrating Radial Basis Function Networks with Explainable AI

T0 review · 5 major / 6 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 a hybrid CNN-RBF network can classify skin lesions with competitive accuracy while making each prediction traceable to a specific image prototype, reporting 83.02% on ISIC 2016 and 76.15% on ISIC 2017 using ResNet-50…

desk verdict The paper's central accuracy claim is undefined because Algorithm 1 never defines how segment-level predictions become image-level predictions, and the abstract's 72.15% conflicts with the body's 76.15% for ISIC 2017. read the letter →

arxiv 2501.14885 v1 pith:WHLD25KB submitted 2025-01-24 cs.CV

Mirza Ahsan Ullah , Tehseen Zia This is my paper · ORCID
classification cs.CV
keywords skincancerdiagnosisradialbasisfunctionnetworksexplainableAIprototype-basedclassificationhybriddeeplearningK-medoidsclusteringISICdatasetinterpretablemachine
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

The paper proposes a hybrid model for skin-cancer diagnosis that splits each image into segments, embeds each segment with a pretrained CNN, and classifies the segment embeddings with an RBF network whose neurons are learned prototypes. The central claim is that this design reaches competitive accuracy—83.02% on ISIC 2016 and 76.15% on ISIC 2017 with ResNet-50 embeddings—while making each prediction explainable by pointing to the prototype that fired most strongly. The authors argue this prototype-based explanation is faithful because it uses the model's actual decision mechanism rather than a post-hoc approximation. A sympathetic reader would care because it suggests diagnostic AI can be both accurate and transparent enough for clinical use.

What carries the argument

The central mechanism is the RBF activation layer: each segment embedding $z$ produces a vector $\Phi(z)$ whose entries are Gaussian similarities $\phi_i(z) = \exp(-\|z-c_i\|^2 / 2\sigma^2)$ to $k$ cluster prototypes $c_i$ chosen by K-medoids clustering over user-selected salient segments. A softmax over a dense layer fed by $\Phi(z)$ yields class probabilities, and the argmax over $\phi_i(z)$ identifies the prototype that explains the decision. The prototypes themselves are shown as the corresponding image segments, grounding each prediction in a visible patch of skin.

What would settle it

Run the pipeline and aggregate the four per-segment predictions by majority vote, by averaged softmax, and by maximum softmax; if none of these reproduces the reported 83.02% (ISIC 2016) and 76.15% (ISIC 2017) figures, the claimed accuracies are not defined by the algorithm in the paper.

Watch

Extended reading notes

Core claim

The paper's core claim is that a hybrid classifier—pretrained CNN embeddings followed by an RBF network over K-medoid prototypes—can classify skin lesions with accuracy competitive with standard deep models while remaining transparent. Concretely, the paper reports 83.02% accuracy on ISIC 2016 and 76.15% on ISIC 2017 using ResNet-50 embeddings, with VGG-16 embeddings giving 79.54% and 70.09%. The new ingredient is that the classifier's decision for each image segment is a softmax over Gaussian similarities to class prototypes, so the explanation for a prediction is literally the prototype segment that activated most strongly. The paper argues that because the explanation comes from the same RBF computation that produces the prediction, it is faithful by construction, unlike post-hoc methods such as Grad-CAM.

Load-bearing premise

The reported accuracies depend on an unstated rule for converting the model's per-segment predictions into a single image-level prediction, because the algorithm as written never combines the $n$ segments of an image.

Editorial extensions

If this is right

  • On ISIC 2016 and ISIC 2017, the RBF-NN with ResNet-50 embeddings reaches 83.02% and 76.15% accuracy, beating its VGG-16 variant while remaining less accurate than the non-interpretable ResNet-50 baseline reported at 92.90% and 83.00%.
  • Each prediction is explained by the single prototype segment whose RBF activation is strongest, so the explanation is an actual image region rather than a heatmap approximation.
  • Using deeper embeddings (ResNet-50 over VGG-16) improves both clustering coherence, with silhouette scores rising from 0.63 to 0.72 on ISIC 2016, and final classification accuracy.
  • The authors claim the same architecture extends to other medical imaging tasks, such as histopathology or radiology, where localized, transparent evidence is needed.

Reading between the lines

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

  • If the missing aggregation rule is specified as a simple majority vote over the four segments, the reported accuracy could be checked directly; until then the headline numbers are not reproducible from the algorithm as written.
  • The prototype explanation inherits the quality of the clustering step: if K-medoids picks a prototype that is visually similar but clinically irrelevant, the explanation would look plausible while pointing at the wrong feature.
  • A natural next experiment would compare explanation faithfulness across RBF prototypes, ProtoPNet-style prototypes, and Grad-CAM on the same backbone, measuring whether clinicians rank the prototype explanations as more useful.
  • One could test the framework on a class-imbalanced real-world dataset without SMOTE or focal loss, to isolate how much of the reported accuracy comes from the RBF architecture versus the augmentation and loss tricks.
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

5 major / 6 minor

Summary. The paper proposes a hybrid interpretable framework for skin cancer diagnosis that combines pre-trained CNN embeddings (VGG-16, ResNet-50) with a Radial Basis Function network. Each image is segmented with SLIC, segment embeddings are clustered to form prototypes, and an RBF activation vector is fed to a dense softmax layer for classification; the explanation is defined as the most activated prototype per segment. The method is evaluated on ISIC 2016 and ISIC 2017, with the main reported results being 83.02% accuracy on ISIC 2016 and 76.15% on ISIC 2017 using ResNet-50. The paper claims that this architecture provides both competitive accuracy and faithful, prototype-based interpretability.

Significance. If the central claims were substantiated, the paper would address a real need: an interpretable, prototype-based diagnostic model for skin cancer with localized explanations that clinicians can trace. The motivation for combining segmentation, prototype selection, and RBF networks is clear, and the authors explicitly aim for ante-hoc interpretability rather than post-hoc approximation. The paper also attempts to compare against ProtoPNet and ProtoPShare, which is appropriate. However, the current manuscript does not provide a verifiable central claim: the image-level accuracy is not reproducible because the segment-to-image aggregation step is undefined, the reported ISIC 2017 accuracy is inconsistent between the abstract and the results section, and the baseline comparisons contain circular citations. The interpretability claim is also not guaranteed by the architecture. These issues are load-bearing for both the accuracy and the interpretability contributions.

major comments (5)
  1. [Abstract vs. Section 4 and Table 5] The abstract reports 72.15% accuracy on ISIC 2017 with ResNet-50, while Section 4 and Table 5 report 76.15% for the same configuration, and the conclusion also says 76.15%. The manuscript never acknowledges or reconciles this discrepancy, so the central empirical claim is ambiguous and the reader cannot tell which number is the reported result.
  2. [Table 1, Algorithm Steps 1-9] The algorithm operates per segment: Step 1 segments the image into n regions, Steps 2-6 compute embeddings, RBF activations, and softmax probabilities for each segment z, and Step 9 gives y_hat = argmax_c p(y=c|z). No step combines the n segment predictions, segment embeddings, or segment probabilities into an image-level prediction. The introduction's statement that the approach 'averages decisions across localized patches' is never formalized, so the reported image-level accuracies of 83.02% and 76.15% are not well-defined and cannot be reproduced.
  3. [Table 5] The comparison baselines ProtoPNet and ProtoPShare are cited to reference [44], which is the authors' own CA-SoftNet paper, rather than to the original ProtoPNet [40] and ProtoPShare [41] sources. This misattributes the baselines and creates a circular comparison in which the proposed model is evaluated against numbers that are not attributed to the actual methods. Additionally, the row 'ResNet-50 [1]' cites reference [1], which is a systematic review, not the ResNet-50 architecture, and the ProtoPNet/ProtoPShare entries omit the '%' sign, making the table internally inconsistent.
  4. [Section 3.1.6 vs. Table 1 and Table 3] Section 3.1.6 states that prototype selection is based on K-Means clustering, while Algorithm 1 (Table 1), Table 3, and the discussion in Section 4 all refer to K-Medoids. Since prototype selection is the mechanism claimed to provide interpretability, the manuscript must specify which clustering algorithm is actually used; the current inconsistency prevents replication.
  5. [Table 1, Steps 6 and 10] Classification is performed by softmax(W·Phi(z)+b) over an unconstrained dense layer, but the explanation is defined as argmax_i phi_i(z), the most activated prototype. Since the learned weight matrix W can reweight prototype activations arbitrarily, the predicted class is not necessarily the class of the most activated prototype. Therefore the explanation is not guaranteed to be faithful to the model's decision, contradicting the paper's claim that explanations are generated through the exact mechanism within the model.
minor comments (6)
  1. [Title page] The affiliation contains a typo: 'Paksitan' should be 'Pakistan'.
  2. [Section 3.1.1, Table 2] The table lists 'PyTorch (version 3.12)'; 3.12 is a Python version, not a PyTorch version, and the actual PyTorch version is not given.
  3. [Section 3.1.3] The text says 'compactness=100 and sigma=1.*3'; this appears to be a formatting error for sigma=1.3 and should be clarified.
  4. [Table 5] The ProtoPNet and ProtoPShare rows omit the '%' sign, and the table mixes numbers with and without percentage symbols, making the comparison difficult to read.
  5. [Equation (5)] The focal loss equation has a typo in the label ('Lf ocal') and uses notation inconsistently with the loss expression in Table 1; the subscripts and arguments of the loss terms should be aligned.
  6. [References] Several in-text citations do not match the reference list, including [1] used for ResNet-50 in Table 5, [2] used for GDPR in the introduction, and the placement of [44] for ProtoPNet/ProtoPShare; the reference numbering and attribution should be corrected throughout.

Circularity Check

1 steps flagged · score 4.0 of 10

Baseline comparison rests on self-citation; the reported accuracies are not themselves circular but are unreproducible because Algorithm 1 omits the image-level aggregation rule.

  1. self citation load bearing [Table 5, Section 4 (Results and discussion); reference [44]]
    "ProtoPNet [44] 84.50 82.90 ProtoPShare [44] 92.25 90.10"

    The table labels these rows as ProtoPNet and ProtoPShare but cites both to reference [44], which is the authors' own CA-SoftNet paper (Ullah et al., PLoS ONE 2024), not to the original ProtoPNet [40] or ProtoPShare [41]. The paper's comparative claim that the proposed model is 'comparable or superior' and that 'ProtoPShare slightly outperformed the proposed model' is therefore supported by numbers from the authors' own prior work rather than by the independent baselines it purports to compare against. The external benchmark is replaced by a self-citation, so the comparison is not an independent check on the proposed method.

full rationale

The proposed model's headline accuracies (83.02% on ISIC 2016 and 76.15% on ISIC 2017 with ResNet-50) are experimental measurements on public datasets, not quantities derived from fitted parameters renamed as predictions, so the core empirical claim is not circular by construction. The main circular element is comparative: Table 5 labels baseline rows as ProtoPNet and ProtoPShare but cites both to reference [44], which is the authors' own CA-SoftNet paper, rather than to ProtoPNet [40] or ProtoPShare [41]. The discussion uses those numbers to argue the proposed model is 'comparable or superior' and that ProtoPShare 'slightly outperformed' it, making the self-citation load-bearing for the comparison claim. Two additional issues are correctness or reproducibility concerns rather than circularity: Algorithm 1 defines predictions per segment z but never states how the n segment predictions are combined into the reported image-level accuracy, and Step 10's 'most activated prototype' explanation is not guaranteed to match the class chosen by the dense softmax in Step 6. Those undermine verifiability and faithfulness but are not reductions of outputs to inputs. Overall score 4: one self-citation is load-bearing for the benchmark comparison, while the central accuracy result retains independent empirical content.

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

No new physical or conceptual entities are introduced; the prototypes are cluster centers of existing embedding features. The central claim rests on several domain assumptions about transfer learning, prototype meaningfulness, explanation faithfulness, and dataset representativeness, none of which are validated with external evidence.

free parameters (4)
  • RBF spread sigma = not reported
    Controls the width of the Gaussian similarity in the RBF layer; no value or selection procedure is given in Section 3.1.6 or Table 4.
  • Number of prototypes per class = 15
    Set to 15 per class with no sensitivity analysis or ablation (Section 3.1.6).
  • Number of superpixel segments per image = 4
    SLIC configured with compactness=100 and sigma=1.3 generates four segments per image; this choice is not ablated (Section 3.1.3).
  • SLIC compactness and sigma = compactness=100, sigma=1.3
    Preprocessing hyperparameters chosen by hand for segmentation (Section 3.1.3).
assumptions (4)
  • domain assumption Pre-trained ImageNet CNN embeddings transfer to dermoscopic image segments without fine-tuning
    Section 3.1.5 uses frozen ResNet-50 and VGG-16 to embed skin lesion segments; transferability is assumed and never validated against a fine-tuned baseline.
  • domain assumption K-medoids cluster centers represent clinically meaningful prototypes
    Section 3.1.6 assumes cluster centroids correspond to salient lesion features; only a silhouette score is reported, with no clinical evaluation.
  • domain assumption The most activated prototype explains the prediction faithfully
    Algorithm Table 1 step 10 equates explanation with argmax RBF activation; no faithfulness metric or user study supports this.
  • domain assumption ISIC 2016 and ISIC 2017 are sufficient to generalize to clinical skin cancer diagnosis
    Sections 3.1.2 and 4 generalize from two small public datasets to clinical workflows without external validation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Hybrid Interpretable Deep Learning Framework for Skin Cancer Diagnosis: Integrating Radial Basis Function Networks with Explainable AI." pith.science (2026). https://pith.science/paper/WHLD25KB

@misc{pith2026250114885,
  author       = {Pith},
  title        = {Pith review of: Hybrid Interpretable Deep Learning Framework for Skin Cancer Diagnosis: Integrating Radial Basis Function Networks with Explainable AI},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WHLD25KB}},
  note         = {Machine review of arXiv:2501.14885}
}
read the original abstract

Skin cancer is one of the most prevalent and potentially life-threatening diseases worldwide, necessitating early and accurate diagnosis to improve patient outcomes. Conventional diagnostic methods, reliant on clinical expertise and histopathological analysis, are often time-intensive, subjective, and prone to variability. To address these limitations, we propose a novel hybrid deep learning framework that integrates convolutional neural networks (CNNs) with Radial Basis Function (RBF) Networks to achieve high classification accuracy and enhanced interpretability. The motivation for incorporating RBF Networks lies in their intrinsic interpretability and localized response to input features, which make them well-suited for tasks requiring transparency and fine-grained decision-making. Unlike traditional deep learning models that rely on global feature representations, RBF Networks allow for mapping segments of images to chosen prototypes, exploiting salient features within a single image. This enables clinicians to trace predictions to specific, interpretable patterns. The framework incorporates segmentation-based feature extraction, active learning for prototype selection, and K-Medoids clustering to focus on these salient features. Evaluations on the ISIC 2016 and ISIC 2017 datasets demonstrate the model's effectiveness, achieving classification accuracies of 83.02\% and 72.15\% using ResNet50, respectively, and outperforming VGG16-based configurations. By generating interpretable explanations for predictions, the framework aligns with clinical workflows, bridging the gap between predictive performance and trustworthiness. This study highlights the potential of hybrid models to deliver actionable insights, advancing the development of reliable AI-assisted diagnostic tools for high-stakes medical applications.

Figures

Figures reproduced from arXiv: 2501.14885 by the authors.

Figure 1
Figure 1. Architecture of proposed Radial basis neural network [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Images from ISIC 2016 of both classes Both datasets, despite their relatively small size, were chosen to demonstrate that our proposed model, designed with intrinsically interpretable architecture, achieves superior performance compared to existing approaches. 3.1.3 Data pre-processing All images are resized to a uniform size of 224×224 pixels (as required by the most pre-trained deep learning models). Pixel values … view at source ↗
Figure 3
Figure 3. Images from ISIC 2017 of all classes 3.1.6 Custom Clustering Layer Our proposed RBF neural network uses a custom clustering layering, which selects the embeddings of salient features from each class. Initially, we used 15 prototype embeddings from each class. The selection is based on K-Means clustering, which learns the embeddings from the concepts/segments selected by the users in the previous step. Corresponding … view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Segments from ISIC 2016 chosen as prototypes by the model from each class [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: Segments from ISIC 2017 chosen as prototypes by the model from each class [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: Visualization of results and explanations of the proposed model [PITH_FULL_IMAGE:figures/full_fig_p011_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

45 extracted references · 32 canonical work pages

  1. [44]

    An inherently interpretable deep learning model for local explanations using visual concepts

    Mirza Ahsan Ullah, Tehseen Zia, Jungeun Kim, and Seifedine Kadry. An inherently interpretable deep learning model for local explanations using visual concepts. Plos one, 19(10):e0311879, 2024

    work page 2024

  2. [40]

    This looks like that: deep learning for interpretable image recognition

    Chaofan Chen, Oscar Li, Daniel Tao, Alina Barnett, Cynthia Rudin, and Jonathan K Su. This looks like that: deep learning for interpretable image recognition. Advances in neural information processing systems, 32, 2019

    2019

  3. [41]

    Protopshare: Prototypical parts sharing for similarity discovery in interpretable image classification

    Dawid Rymarczyk, Łukasz Struski, Jacek Tabor, and Bartosz Zieli´nski. Protopshare: Prototypical parts sharing for similarity discovery in interpretable image classification. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining, pages 1420–1430, 2021

    work page 2021

  4. [1]

    Xiaoxuan Liu and et al. A comparison of deep learning performance against health-care professionals in detecting diseases from medical imaging: a systematic review and meta-analysis.The Lancet Digital Health, 1(6):e271–e297, 2019

    work page 2019

  5. [2]

    A benchmark for interpretability methods in deep neural networks

    Sara Hooker and et al. A benchmark for interpretability methods in deep neural networks. In Advances in Neural Information Processing Systems, volume 32, 2019

    work page 2019

  6. [3]

    A benchmark for neural network robustness in skin cancer classification

    RC Maron, JG Schlager, S Haggenmüller, and et al. A benchmark for neural network robustness in skin cancer classification. European Journal of Cancer, 155:191–199, 2021

    work page 2021

  7. [4]

    Robustness of convolutional neural networks in recognition of pigmented skin lesions

    RC Maron, S Haggenmüller, C von Kalle, and et al. Robustness of convolutional neural networks in recognition of pigmented skin lesions. European Journal of Cancer, 145:81–91, 2021

    work page 2021

  8. [5]

    White paper on artificial intelligence - a european approach to excellence and trust, 2020

    European Commission. White paper on artificial intelligence - a european approach to excellence and trust, 2020. Accessed: 28 December 2021

    work page 2020

Show all 45 references
  1. [6]

    Making the black box more transparent: Understanding the physical implications of machine learning

    Amy McGovern and et al. Making the black box more transparent: Understanding the physical implications of machine learning. Bulletin of the American Meteorological Society, 100(11):2175–2199, 2019

    2019

  2. [7]

    Toms, Elizabeth A

    Benjamin A. Toms, Elizabeth A. Barnes, and Imme Ebert-Uphoff. Physically interpretable neural networks for the geosciences: Applications to earth system variability. Journal of Advances in Modeling Earth Systems, 12(9):e2019MS002002, 2020

    2020

  3. [8]

    Attention-based interpretable neural network for building cooling load prediction

    Ao Li and et al. Attention-based interpretable neural network for building cooling load prediction. Applied Energy, 299:117238, 2021

    2021

  4. [9]

    Interpretable neural network via algorithm unrolling for mechanical fault diagnosis

    Botao An and et al. Interpretable neural network via algorithm unrolling for mechanical fault diagnosis. IEEE Transactions on Instrumentation and Measurement, 71:1–11, 2022

    2022

  5. [10]

    General pitfalls of model-agnostic interpretation methods for machine learning models

    Christoph Molnar and et al. General pitfalls of model-agnostic interpretation methods for machine learning models. In International Workshop on Extending Explainable AI Beyond Deep Models and Classifiers. Springer International Publishing, 2020

    2020

  6. [11]

    Methods for interpreting and understanding deep neural networks

    Grégoire Montavon, Wojciech Samek, and Klaus-Robert Müller. Methods for interpreting and understanding deep neural networks. Digital Signal Processing, 73:1–15, 2018

    2018

  7. [12]

    David Gutman, Noel CF Codella, Emre Celebi, Brian Helba, Michael Marchetti, Nabin Mishra, and Allan Halpern. Skin lesion analysis toward melanoma detection: A challenge at the international symposium on biomedical imaging (isbi) 2016, hosted by the international skin imaging c...

    2016 arXiv

  8. [13]

    Noel CF Codella, David Gutman, M Emre Celebi, Brian Helba, Michael A Marchetti, Stephen W Dusza, Aadi Kalloo, Konstantinos Liopyris, Nabin Mishra, Harald Kittler, et al. Skin lesion analysis toward melanoma detection: A challenge at the 2017 international symposium on biomedic...

    2017

  9. [14]

    Deep learning applications for iot in health care: A systematic review

    Hamidreza Bolhasani, Maryam Mohseni, and Amir Masoud Rahmani. Deep learning applications for iot in health care: A systematic review. Informatics in Medicine Unlocked, 23:100550, 2021

    2021

  10. [15]

    A guide to deep learning in healthcare

    Andre Esteva, Alexandre Robicquet, Bharath Ramsundar, V olodymyr Kuleshov, Mark DePristo, Katherine Chou, Claire Cui, Greg Corrado, Sebastian Thrun, and Jeff Dean. A guide to deep learning in healthcare. Nature medicine, 25(1):24–29, 2019

    2019

  11. [16]

    Recent deep learning techniques, challenges and its applications for medical healthcare system: a review

    Saroj Kumar Pandey and Rekh Ram Janghel. Recent deep learning techniques, challenges and its applications for medical healthcare system: a review. Neural Processing Letters, 50(2):1907–1935, 2019

    1907

  12. [17]

    Deep learning in healthcare

    Deeksha Kaul, Harika Raju, and BK Tripathy. Deep learning in healthcare. Deep Learning in Data Analytics: Recent Techniques, Practices and Applications, pages 97–115, 2022

    2022

  13. [18]

    From machine learning to deep learning: Advances of the recent data-driven paradigm shift in medicine and healthcare

    Chiranjib Chakraborty, Manojit Bhattacharya, Soumen Pal, and Sang-Soo Lee. From machine learning to deep learning: Advances of the recent data-driven paradigm shift in medicine and healthcare. Current Research in Biotechnology, 7:100164, 2024

    2024

  14. [19]

    Multiclass skin lesion localization and classification using deep learning based features fusion and selection framework for smart healthcare

    Sarmad Maqsood and Robertas Damaševiˇcius. Multiclass skin lesion localization and classification using deep learning based features fusion and selection framework for smart healthcare. Neural networks, 160:238–258, 2023

    2023

  15. [20]

    Leveraging deep learning for designing healthcare analytics heuristic for diagnostics

    Sarah Shafqat, Maryyam Fayyaz, Hasan Ali Khattak, Muhammad Bilal, Shahid Khan, Osama Ishtiaq, Almas Abbasi, Farzana Shafqat, Waleed S Alnumay, and Pushpita Chatterjee. Leveraging deep learning for designing healthcare analytics heuristic for diagnostics. Neural processing lett...

    2023

  16. [21]

    Learning deep features for discriminative localization

    Bolei Zhou, Aditya Khosla, Agata Lapedriza, Aude Oliva, and Antonio Torralba. Learning deep features for discriminative localization. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 2921–2929, 2016

    2016

  17. [22]

    Grad-cam++: General- ized gradient-based visual explanations for deep convolutional networks

    Aditya Chattopadhay, Anirban Sarkar, Prantik Howlader, and Vineeth N Balasubramanian. Grad-cam++: General- ized gradient-based visual explanations for deep convolutional networks. In 2018 IEEE winter conference on applications of computer vision (WACV), pages 839–847. IEEE, 2018

    2018

  18. [23]

    Opti-cam: Optimizing saliency maps for interpretability

    Hanwei Zhang, Felipe Torres, Ronan Sicre, Yannis Avrithis, and Stephane Ayache. Opti-cam: Optimizing saliency maps for interpretability. Computer Vision and Image Understanding, 248:104101, 2024

    2024

  19. [24]

    Deep residual learning for image recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 770–778, 2016

    2016

  20. [25]

    Melanoma detection using adversarial training and deep transfer learning

    Hasib Zunair and A Ben Hamza. Melanoma detection using adversarial training and deep transfer learning. Physics in Medicine & Biology, 65(13):135005, 2020

    2020

  21. [26]

    Very deep convolutional networks for large-scale image recognition

    Karen Simonyan. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556, 2014

    2014 arXiv

  22. [27]

    From attribution maps to human-understandable explanations through concept relevance propagation

    Reduan Achtibat, Maximilian Dreyer, Ilona Eisenbraun, Sebastian Bosse, Thomas Wiegand, Wojciech Samek, and Sebastian Lapuschkin. From attribution maps to human-understandable explanations through concept relevance propagation. Nature Machine Intelligence, 5(9):1006–1019, 2023

    2023

  23. [28]

    The disagreement problem in explainable machine learning: A practitioner’s perspective

    Satyapriya Krishna, Tessa Han, Alex Gu, Javin Pombra, Shahin Jabbari, Steven Wu, and Himabindu Lakkaraju. The disagreement problem in explainable machine learning: A practitioner’s perspective. arXiv preprint arXiv:2202.01602, 2022

    2022 arXiv

  24. [29]

    Evaluating the explainers: Black- box explainable machine learning for student success prediction in moocs

    Vinitra Swamy, Bahar Radmehr, Natasa Krco, Mirko Marras, and Tanja Käser. Evaluating the explainers: Black- box explainable machine learning for student success prediction in moocs. In International Conference on Educational Data Mining, 2022

    2022

  25. [30]

    Interpretcc: Intrinsic user-centric interpretability through global mixture of experts, 2024

    Vinitra Swamy and et al. Interpretcc: Intrinsic user-centric interpretability through global mixture of experts, 2024

    2024

  26. [31]

    Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead

    Cynthia Rudin. Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead. Nature Machine Intelligence, 1:206–215, 2019

    2019

  27. [32]

    Zachary C. Lipton. The mythos of model interpretability. Queue, 16(3):31–57, 2018

    2018

  28. [33]

    Ross Quinlan

    J. Ross Quinlan. Induction of decision trees. Machine learning, 1:81–106, 1986

    1986

  29. [34]

    Decision trees

    Lior Rokach and Oded Maimon. Decision trees. Data mining and knowledge discovery handbook, pages 165–192, 2005

    2005

  30. [35]

    A novel hybrid classification model of artificial neural networks and multiple linear regression models

    Mehdi Khashei, Ali Zeinal Hamadani, and Mehdi Bijari. A novel hybrid classification model of artificial neural networks and multiple linear regression models. Expert Systems with Applications, 39(3):2606–2620, 2012. 13 A PREPRINT - JANUARY 28, 2025

    2012

  31. [36]

    Fifty years of computer analysis in chest imaging: rule-based, machine learning, deep learning

    Bram van Ginneken. Fifty years of computer analysis in chest imaging: rule-based, machine learning, deep learning. Radiological physics and technology, 10:23–32, 2017

    2017

  32. [37]

    Using rule-based machine learning for candidate disease gene prioritization and sample classification of cancer gene expression data

    Enrico Glaab, Jaume Bacardit, Jonathan M Garibaldi, and Natalio Krasnogor. Using rule-based machine learning for candidate disease gene prioritization and sample classification of cancer gene expression data. PloS one, 7(7):e39932, 2012

    2012

  33. [38]

    Radial basis functions

    Martin Dietrich Buhmann. Radial basis functions. Acta numerica, 9:1–38, 2000

    2000

  34. [39]

    Radial basis function approximations: comparison and applications

    Zuzana Majdisova and Vaclav Skala. Radial basis function approximations: comparison and applications. Applied Mathematical Modelling, 51:728–743, 2017

    2017

  35. [42]

    This looks like those: Illuminating prototypical concepts using multiple visualizations

    Chiyu Ma, Brandon Zhao, Chaofan Chen, and Cynthia Rudin. This looks like those: Illuminating prototypical concepts using multiple visualizations. Advances in Neural Information Processing Systems, 36:39212–39235, 2023

    2023

  36. [43]

    Softnet: A concept-controlled deep learning architecture for interpretable image classification

    Tehseen Zia, Nauman Bashir, Mirza Ahsan Ullah, and Shakeeb Murtaza. Softnet: A concept-controlled deep learning architecture for interpretable image classification. Knowledge-Based Systems, 240:108066, 2022

    2022

  37. [45]

    Patchnet: interpretable neural networks for image classification

    Adityanarayanan Radhakrishnan, Charles Durham, Ali Soylemezoglu, and Caroline Uhler. Patchnet: interpretable neural networks for image classification. arXiv preprint arXiv:1705.08078, 2017. 14

    2017 arXiv

Pith tools

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