OSCR

ETDACVO: Structural-Fidelity-Aware Evolutionary Co-Optimization for Robust and Explainable Brain Tumor MRI Classification.

Code ↔ Paper

18 matches between paragraphs of the paper and lines of its authors' code, computed by the harvester (lexical-v1). Click a colored paragraph or line to see its counterpart.

The 18 matches · 13 of them tie a paragraph to a whole file, not to given lines: weak matches, whose lines are not tinted
  1. [1] § 2. Materials and Methods › 2.6. Convergence-Aware Evolutionary Attention Grad-CAM (CA-EA-GradCAM) ↔ explainability/ca_ea_gradcam.py, the whole file · a weak match · score 0.93 · CA EA GradCAM, evolutionary convergence confidence, transformer attention, gated, explanations, localization
  2. [2] § 2. Materials and Methods ↔ optimizer/etdacvo.py, the whole file · a weak match · score 0.85 · Anti Conservative Variable, ETDACVO evolutionary optimization, Tasmanian Devil, EWMA smoothing, augmentation parameters, TDO
  3. [3] § 3. Experimental Evaluation › 3.12. Explainability and Diagnostic Reliability Evaluation ↔ explainability/ca_ea_gradcam.py, the whole file · a weak match · score 0.84 · CA EA GradCAM, Attention Grad CAM, CNN Transformer model, Convergence Aware Evolutionary
  4. [4] § 2. Materials and Methods › 2.3. Hybrid CNN–Transformer Backbone ↔ models/hybrid_model.py, the whole file · a weak match · score 0.78 · transformer encoder, hybrid CNN, Transformer features, backbone, fused, global
  5. [5] § 2. Materials and Methods › 2.3. Hybrid CNN–Transformer Backbone ↔ models/hybrid_model.py, the whole file · a weak match · score 0.76 · transformer encoder, hybrid CNN, flattened, backbone, linear, layer
  6. [6] § 2. Materials and Methods › 2.4. ETDACVO: Evolutionary Co-Optimization Framework › 2.4.2. Anti-Conservative Variable Optimization (ACVO) ↔ optimizer/acvo.py, the whole file · a weak match · score 0.76 · correlation matrix, covariance structure, perturbation, ACVO, population, Anti
  7. [7] § 2. Materials and Methods › 2.7. Theoretical Insight into ETDACVO Convergence ↔ optimizer/etdacvo.py, the whole file · a weak match · score 0.74 · Anti Conservative Variable, Tasmanian Devil, EWMA smoothing, population, exploration, diversity
  8. [8] § 2. Materials and Methods › 2.4. ETDACVO: Evolutionary Co-Optimization Framework › 2.4.1. Tasmanian Devil Optimization (TDO) ↔ optimizer/tdo.py, the whole file · a weak match · score 0.74 · vy flight, stochastic exploratory, exploration mechanism, TDO, Devil, Tasmanian
  9. [9] § 2. Materials and Methods › 2.7. Theoretical Insight into ETDACVO Convergence ↔ optimizer/tdo.py, the whole file · a weak match · score 0.70 · Tasmanian Devil Optimization, vy flight, stochastic, algorithm, component, Variable
  10. [10] § 2. Materials and Methods › 2.3. Hybrid CNN–Transformer Backbone ↔ models/resnet_model.py, lines 6–38 · score 0.68 · ResNet, medical image classification, backbone, convolutional, linear, MRI
  11. [11] § 2. Materials and Methods › 2.2. Preprocessing and Input Standardization ↔ preprocessing/bias_field.py, lines 5–29 · score 0.65 · N4 bias field, preprocessing
  12. [12] § 2. Materials and Methods › 2.6. Convergence-Aware Evolutionary Attention Grad-CAM (CA-EA-GradCAM) › 2.6.3. Evolutionary Convergence Confidence (ECC) ↔ explainability/convergence_confidence.py, the whole file · a weak match · score 0.64 · ETDACVO parameter trajectory, optimization stability, Confidence, ECC, weight, Evolutionary
  13. [13] § 3. Experimental Evaluation › 3.7. Fusion Strategy Evaluation ↔ models/fusion.py, lines 47–68 · score 0.58 · Attention Weighted Fusion, Additive Fusion, Concatenation
  14. [14] § 3. Experimental Evaluation › 3.7. Fusion Strategy Evaluation ↔ models/fusion.py, lines 47–68 · score 0.57 · attention weighted fusion, additive fusion, concatenation
  15. [15] § 2. Materials and Methods › 2.6. Convergence-Aware Evolutionary Attention Grad-CAM (CA-EA-GradCAM) › 2.6.3. Evolutionary Convergence Confidence (ECC) ↔ explainability/convergence_confidence.py, the whole file · a weak match · score 0.56 · indicate stronger convergence, Confidence, ECC, Evolutionary
  16. [16] § 3. Experimental Evaluation › 3.7. Fusion Strategy Evaluation ↔ models/__init__.py, the whole file · a weak match · score 0.56 · attention weighted fusion, additive fusion, concatenation
  17. [17] § 2. Materials and Methods › 2.2. Preprocessing and Input Standardization ↔ preprocessing/__init__.py, the whole file · a weak match · score 0.55 · skull stripping, reorientation, resampling, bias, preprocessing
  18. [18] § 3. Experimental Evaluation › 3.6. Training Stability and Convergence Analysis ↔ analysis/plot_utils.py, lines 5–24 · score 0.50 · loss curves, validation loss, epochs, training

Paper

Loaded from Europe PMC by your browser, not stored by OSCR: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

Python · 117 lines · 2.8 KB · other · 2 matches

  1. import torch
  2. import torch.nn.functional as F
  5. class CA_EA_GradCAM:
  6. """
  7. Convergence-Aware Evolutionary Attention GradCAM
  9. Combines:
  10. - CNN GradCAM localization
  11. - Transformer attention context
  12. - Evolutionary convergence confidence
  13. """
  15. def __init__(self, model, target_layer):
  16. """
  17. model: trained CNN-Transformer model
  18. target_layer: CNN feature layer used for GradCAM
  19. """
  21. self.model = model
  22. self.target_layer = target_layer
  24. self.feature_maps = None
  25. self.gradients = None
  27. # hooks
  28. self._register_hooks()
  30. def _register_hooks(self):
  32. def forward_hook(module, input, output):
  33. self.feature_maps = output.detach()
  35. def backward_hook(module, grad_in, grad_out):
  36. self.gradients = grad_out[0].detach()
  38. self.target_layer.register_forward_hook(forward_hook)
  39. self.target_layer.register_backward_hook(backward_hook)
  41. def compute_gradcam(self, input_tensor, class_idx=None):
  43. """
  44. Compute CNN GradCAM heatmap
  45. """
  47. self.model.zero_grad()
  49. output = self.model(input_tensor)
  51. if class_idx is None:
  52. class_idx = torch.argmax(output)
  54. score = output[:, class_idx]
  56. score.backward(retain_graph=True)
  58. gradients = self.gradients
  59. feature_maps = self.feature_maps
  61. weights = torch.mean(gradients, dim=(2, 3), keepdim=True)
  63. cam = torch.sum(weights * feature_maps, dim=1)
  65. cam = F.relu(cam)
  67. cam = cam / (torch.max(cam) + 1e-8)
  69. return cam
  71. def compute_entropy(self, heatmap):
  72. """
  73. Spatial entropy of GradCAM map
  74. """
  76. p = heatmap / (torch.sum(heatmap) + 1e-8)
  78. entropy = -torch.sum(p * torch.log(p + 1e-8))
  80. return entropy
  82. def fuse_maps(self, cnn_map, transformer_map, psi):
  83. """
  84. Fuse CNN and transformer explanations
  85. """
  87. entropy = self.compute_entropy(cnn_map)
  89. gate = torch.sigmoid(entropy)
  91. fused = psi * (
  92. gate * cnn_map +
  93. (1 - gate) * transformer_map
  94. )
  96. fused = fused / (torch.max(fused) + 1e-8)
  98. return fused
  100. def generate(self, input_tensor, transformer_attention, psi):
  101. """
  102. Full CA-EA-GradCAM generation pipeline
  103. """
  105. cnn_map = self.compute_gradcam(input_tensor)
  107. # resize transformer map to CNN resolution
  108. transformer_map = F.interpolate(
  109. transformer_attention.unsqueeze(1),
  110. size=cnn_map.shape[-2:],
  111. mode="bilinear",
  112. align_corners=False,
  113. ).squeeze(1)
  115. heatmap = self.fuse_maps(cnn_map, transformer_map, psi)
  117. return heatmap

ca_ea_gradcam.py at commit e756b1c, under other · at the source

Overview

Authors: Indrakumar Krishnamurthy1, Ravikumar Manjunath1, Mohammed A S Al-Mohamadi1, Lubna A Gabralla2, Sami F Karali3, Mohammed I Thanoon3, Abed Saif Ahmed Alghawli4, Abdulbasit A Darem5
ORCID iDs: Indrakumar Krishnamurthy, Mohammed A S Al-Mohamadi, Sami F Karali, Mohammed I Thanoon, Abed Saif Ahmed Alghawli, Abdulbasit A Darem
  1. Department of MCA and Computer Science, Jnanasahyadri, Kuvempu University, Shivamogga 577 451, India; (I.K.); (R.M.); (M.A.S.A.-M.)
  2. Department of Computer Sciences, Applied College, Princess Nourah bint Abdulrahman University, Riyadh 11671, Saudi Arabia
  3. Computing Department, College of Engineering and Computing at Al-Lith, Umm Al-Qura University, Mecca 24382, Saudi Arabia; (S.F.K.); (M.I.T.)
  4. Computer Science Department, College of Sciences and Humanities, Prince Sattam Bin Abdulaziz University, Aflaj 16278, Saudi Arabia
  5. Center for Scientific Research and Entrepreneurship, Northern Border University, Arar 73213, Saudi Arabia
Institutions: Kuvempu University (India); Princess Nourah bint Abdulrahman University (Saudi Arabia); Umm al-Qura University (Saudi Arabia); Prince Sattam Bin Abdulaziz University (Saudi Arabia); Northern Border University (Saudi Arabia)
Journal: Biomedicines, volume 14, issue 7, article 1475
Dates: received 24 May 2026; accepted 26 June 2026; published online 29 June 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.3390/biomedicines14071475 · PMID 42511950 · PMCID PMC13405514 · OpenAlex W7166538492
Open access: gold, a free copy (OpenAlex)
Status: code verified
Categories: structural MRI / diffusion (modality), other condition (population), methods / tools (subfield)
Methods: Connectivity, Preprocessing, Machine learning, Smoothing, state filtering, decompositions
Keywords: adaptive optimization, evolutionary algorithms, medical image learning, brain tumor MRI, domain generalization, explainable AI, hybrid CNN–transformer models, ETDACVO
Topic: Brain Tumor Detection and Classification (Neurology, Neuroscience), according to OpenAlex
Funding: Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia and Prince Sattam bin Ab-dulaziz University (PSAU/2026/R/1447); Deanship of Scientific Research at Northern Border University (NBU-FFR-2026-2903-02); Princess Nourah bint Abdulrahman University Researchers Supporting Project (PNURSP2026R178)
Citations: not cited yet (Europe PMC); 22 references in the paper

Abstract

Background/Objectives: Heterogeneous imaging protocols, a lack of labeled data, and domain shifts continue to make training deep learning models to analyze medical images a challenge. This study presents ETDACVO (Enhanced Tasmanian Devil Anti-Conservative Variable Optimization), a hybrid evolutionary optimization system designed to improve convergence stability and cross-domain robustness in brain tumor MRI classification. Methods: ETDACVO combines Tasmanian Devil Optimization (TDO), Anti-Conservative Variable Optimization (ACVO), and Exponentially Weighted Moving Average (EWMA) smoothing to stabilize evolutionary parameter updates. Unlike existing approaches that optimize augmentation policies or optimizer dynamics separately, ETDACVO simultaneously evolves both components within a single evolutionary loop. The framework was evaluated on four MRI datasets (Nickparvar, Mendeley, BRISC, and Figshare), comprising 28,151 images. In addition, a convergence-aware explainability mechanism, CA-EA-GradCAM, was developed by integrating gradient saliency, transformer attention, and evolutionary convergence confidence to generate confidence-sensitive tumor localization maps. Results: Experimental results demonstrated that ETDACVO achieved a 2.3–2.5% improvement in classification accuracy and converged 19–22 epochs faster than baseline optimizers. The statistical significance of these improvements was confirmed using paired statistical tests (p < 1 × 10−5). Cross-dataset transfer experiments further showed strong domain-shift resilience, with performance retention reaching 92.8%. The proposed CA-EA-GradCAM mechanism provided interpretable and confidence-aware tumor localization maps. Conclusions: ETDACVO provides a robust and computationally efficient optimization framework for deep-learning-based medical image analysis. By jointly optimizing augmentation strategies and optimizer dynamics, the framework enhances convergence stability, cross-domain robustness, and interpretability, making it a promising approach for reliable brain tumor MRI classification under heterogeneous imaging conditions.

Reproduced under the paper's license (CC BY), from the paper cited above.

Repository

Its files are read in the Code ↔ Paper reader above, with 18 matches between paragraphs and lines of code.

IndrakumarK/Evolutionary-Co-Optimization-of-Data-Augmentation-and-Learning-Dynamics-for-BT-MRI-Classification

License: other
State: the link answers, verified on 27 September 2026
Evidence: files inventoried
Commit: e756b1c23a41b505e6af6cb3caf330f65575a254, 16 March 2026
Languages: Python (45)
Size: 53 files, 45 scripts
Software Heritage: not archived
Found in: “Data Availability Statement”
Holds: README, license file, environment (requirements.txt, setup.py)
Not found: CITATION.cff, tests, continuous integration, documentation
Tools: PyTorch (27 files), NumPy (16 files), SimpleITK (5 files), Matplotlib (3 files), scikit-learn (3 files), SciPy (2 files), NiBabel (1 file), pandas (1 file), scikit-image (1 file)
Availability: 1 check, the latest on 27 September 2026: the link answers
  • 27 September 2026: the link answers
47 files
At the source: github.com/IndrakumarK/Evolutionary-Co-Optimization-of-Data-Augmentation-and-Learning-Dynamics-for-BT-MRI-Classification.git

The paper's code and data availability statement is in the Data section.

Tracing map

Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.

What the map holds:

  • 1 repository of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
  • 45 scripts, each with its path and the digest of its content;
  • 18 matches between paragraphs of the paper and lines of the code (method lexical-v1);
  • neither the text of the paper nor the code itself.

Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.

Data

No dataset and no data link were found in the paper.

Data Availability Statement

The source code for ETDACVO is publicly available at: https://github.com/IndrakumarK/Evolutionary-Co-Optimization-of-Data-Augmentation-and-Learning-Dynamics-for-BT-MRI-Classification.git (accessed on 25 June 2026). The datasets used in this study are publicly available from their respective official repositories. Direct links to all datasets are provided in the manuscript and are also documented in the GitHub repository for reproducibility.

Reproduced under the paper's license (CC BY), from the paper cited above.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 1, 27 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 8 authors, 8 keywords, 3 funders, 17 references.

Cite

This paper

Krishnamurthy, I., Manjunath, R., Al-Mohamadi, M. A. S., Gabralla, L. A., Karali, S. F., Thanoon, M. I., Alghawli, A. S. A., & Darem, A. A. (2026). ETDACVO: Structural-Fidelity-Aware Evolutionary Co-Optimization for Robust and Explainable Brain Tumor MRI Classification. Biomedicines, 14(7), 1475. https://doi.org/10.3390/biomedicines14071475

BibTeX

@article{krishnamurthy2026etdacvo,
author = {Krishnamurthy, Indrakumar and Manjunath, Ravikumar and Al-Mohamadi, Mohammed A S and Gabralla, Lubna A and Karali, Sami F and Thanoon, Mohammed I and Alghawli, Abed Saif Ahmed and Darem, Abdulbasit A},
title = {{ETDACVO: Structural-Fidelity-Aware Evolutionary Co-Optimization for Robust and Explainable Brain Tumor MRI Classification}},
journal = {Biomedicines},
year = {2026},
month = jun,
volume = {14},
number = {7},
pages = {1475},
publisher = {Multidisciplinary Digital Publishing Institute (MDPI)},
issn = {2227-9059},
doi = {10.3390/biomedicines14071475},
url = {https://doi.org/10.3390/biomedicines14071475},
pmid = {42511950},
pmcid = {PMC13405514}
}

RIS

TY - JOUR
AU - Krishnamurthy, Indrakumar
AU - Manjunath, Ravikumar
AU - Al-Mohamadi, Mohammed A S
AU - Gabralla, Lubna A
AU - Karali, Sami F
AU - Thanoon, Mohammed I
AU - Alghawli, Abed Saif Ahmed
AU - Darem, Abdulbasit A
TI - ETDACVO: Structural-Fidelity-Aware Evolutionary Co-Optimization for Robust and Explainable Brain Tumor MRI Classification
T2 - Biomedicines
J2 - Biomedicines
PY - 2026
DA - 2026/06/29
VL - 14
IS - 7
SP - 1475
SN - 2227-9059
PB - Multidisciplinary Digital Publishing Institute (MDPI)
DO - 10.3390/biomedicines14071475
UR - https://doi.org/10.3390/biomedicines14071475
LA - en
ER -

CSL-JSON

{
"id": "10.3390/biomedicines14071475",
"type": "article-journal",
"title": "ETDACVO: Structural-Fidelity-Aware Evolutionary Co-Optimization for Robust and Explainable Brain Tumor MRI Classification",
"container-title": "Biomedicines",
"author": [
{
"family": "Krishnamurthy",
"given": "Indrakumar"
},
{
"family": "Manjunath",
"given": "Ravikumar"
},
{
"family": "Al-Mohamadi",
"given": "Mohammed A S"
},
{
"family": "Gabralla",
"given": "Lubna A"
},
{
"family": "Karali",
"given": "Sami F"
},
{
"family": "Thanoon",
"given": "Mohammed I"
},
{
"family": "Alghawli",
"given": "Abed Saif Ahmed"
},
{
"family": "Darem",
"given": "Abdulbasit A"
}
],
"container-title-short": "Biomedicines",
"volume": "14",
"issue": "7",
"page": "1475",
"DOI": "10.3390/biomedicines14071475",
"PMID": "42511950",
"PMCID": "PMC13405514",
"ISSN": "2227-9059",
"publisher": "Multidisciplinary Digital Publishing Institute (MDPI)",
"URL": "https://doi.org/10.3390/biomedicines14071475",
"language": "en",
"issued": {
"date-parts": [
[
2026,
6,
29
]
]
}
}

The tracing map gets a citation of its own once an author has validated it and it has a DOI.

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.3389/frai.2026.1771088 [code]
Few-shot deployment of pretrained MRI transformers in brain imaging tasks.
Journal: Frontiers in artificial intelligence
In common: SimpleITK, scikit-image, NiBabel, 6 other tools, methods / tools, structural MRI / diffusion, 1 reference
[2] doi:10.21037/qims-2026-0792 [code]
An nnU-Net-based framework with adaptive feature representation for 3D brain tumor segmentation.
Journal: Quantitative imaging in medicine and surgery
In common: SimpleITK, scikit-image, NiBabel, 6 other tools, methods / tools, structural MRI / diffusion, other condition
[3] doi:10.3389/fmed.2026.1875760 [code]
Adaptive multi-stage domain unlearning for white-matter lesion segmentation.
Journal: Frontiers in medicine
In common: SimpleITK, scikit-image, NiBabel, 6 other tools, methods / tools, structural MRI / diffusion
[4] doi:10.3390/jimaging12070276 [code]
Hyperelastic Regularization for Near-Diffeomorphic Transformer-Based Brain MRI Registration.
Journal: Journal of imaging
In common: SimpleITK, scikit-image, NiBabel, 6 other tools, methods / tools, structural MRI / diffusion
[5] doi:10.1002/alz.71649 [code]
Postmortem brain MRI reveals differential associations of subcortical and limbic volumes with cortical thinning and neurodegenerative pathologies.
Journal: Alzheimer's & dementia : the journal of the Alzheimer's Association
In common: SimpleITK, scikit-image, NiBabel, 6 other tools, structural MRI / diffusion, other condition
[6] doi:10.1002/hipo.70124 [code]
Association Between Anterior Hippocampal Gyrification and Episodic Memory Performance in Neurotypical Young Adults.
Journal: Hippocampus
In common: SimpleITK, scikit-image, NiBabel, 6 other tools, structural MRI / diffusion
[7] doi:10.1371/journal.pcbi.1014555 [code]
Body surface potential driven personalisation of electrophysiological digital twins in hypertrophic cardiomyopathy.
Journal: PLoS computational biology
In common: SimpleITK, scikit-image, NiBabel, 6 other tools, structural MRI / diffusion
[8] doi:10.64898/2026.07.15.26357954 [code]
Portable Ultra-Low Field MRI Deep-Learning Algorithms for White Matter Lesion Segmentation Improve Accuracy and Reflect Clinical Disability in Multiple Sclerosis
Journal: medRxiv (preprint)
In common: SimpleITK, scikit-image, NiBabel, 6 other tools, structural MRI / diffusion
[9] doi:10.3389/fnins.2026.1870124 [code]
An end-to-end pipeline for automated fetal brain segmentation and biometry from 3D SSFP MRI.
Journal: Frontiers in neuroscience
In common: SimpleITK, scikit-image, NiBabel, 6 other tools, structural MRI / diffusion
[10] doi:10.1136/jnnp-2025-335884 [code]
Diffusivity anisotropy signature of slowly expanding lesions predicts progression independent of relapse activity in multiple sclerosis.
Journal: Journal of neurology, neurosurgery, and psychiatry
In common: SimpleITK, scikit-image, NiBabel, 6 other tools, structural MRI / diffusion

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion, reproductions, activity

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.