Deep Brain Net: An Optimized Deep Learning Model for Brain tumor Detection in MRI Images Using EfficientNetB0 and ResNet50 with Transfer Learning

Daniel Onah; Ravish Desai

REVIEW 4 major objections 5 minor 12 references

Deep Brain Net: An Optimized Deep Learning Model for Brain tumor Detection in MRI Images Using EfficientNetB0 and ResNet50 with Transfer Learning

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read DeepBrainNet fuses EfficientNetB0 and ResNet50 to classify four brain MRI tumor categories, claiming 88% accuracy, 88.75% weighted F1, and a 98.17% macro AUC.

desk verdict A standard two-backbone transfer-learning application whose headline accuracy is unsupported by an internally contradictory dataset description and inconsistent metric reporting. read the letter →

arxiv 2507.07011 v1 pith:WGX55CXO submitted 2025-07-09 eess.IV cs.CV

Daniel Onah , Ravish Desai This is my paper

classification eess.IVcs.CV

keywords DeepBrainNettumordetectionMRIclassificationEfficientB0ResNet50transferlearningdepthwiseseparableconvolutionsfuzzyC-means

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper is an engineering claim: a hybrid network called DeepBrainNet, built by running EfficientNetB0 and ResNet50 in parallel and fine-tuning both on a public brain MRI dataset, can sort scans into glioma, meningioma, pituitary tumor, and no tumor at 88% accuracy. It reports a weighted F1 of 88.75% and a macro AUC-ROC of 98.17%, which it reads as evidence that the model separates tumor types reliably while staying computationally cheap. Transfer learning from ImageNet supplies the starting weights, depthwise separable convolutions in the EfficientNetB0 branch cut parameters, and residual connections in ResNet50 keep the deeper branch trainable. The stated purpose is a practical decision-support tool for radiologists, not a new biological mechanism.

What carries the argument

The machine doing the work is a parallel hybrid: an EfficientNetB0 branch, whose mobile inverted bottleneck (MBConv) blocks use depthwise separable convolutions to cut parameters and computational cost, joined with a ResNet50 branch that is initialized from ImageNet weights and fine-tuned, so residual connections keep gradients flowing during transfer learning. Around this sit a preprocessing chain (resizing, CLAHE contrast enhancement, histogram equalization, margin cropping), fuzzy C-means feature selection that retains features with high membership across clusters, and training-time augmentation with rotations, flips, zooms, shears, brightness changes, and shifts. Early stopping on validation loss and dynamic learning-rate reduction stabilize training.

What would settle it

Re-run the architecture on the same 7,023-image data with a strictly disjoint test split and report test accuracy and weighted F1; if test accuracy falls below 88% by more than split sampling error, the central claim is not supported. A cheaper control is an ablation training EfficientNetB0 and ResNet50 alone under identical settings: the hybrid must beat its better single branch, otherwise the fusion is not what carries the result.

Watch

Extended reading notes

Core claim

The central claim of the paper is that a single deep network can get the best of both efficiency and depth by fusing EfficientNetB0's mobile inverted bottleneck blocks with a ResNet50 branch that carries over ImageNet features. Trained with on-the-fly augmentation, dropout, early stopping, and fuzzy C-means feature selection, the hybrid reports 88% overall accuracy on a 7,023-image four-class MRI dataset. The class-level report shows the strongest separation for the no-tumor class (precision 0.946) and the weakest for meningioma (F1 0.808). The paper presents these numbers as outperforming prior deep-learning classifiers on the same benchmark and as evidence that an architecture tuned for efficiency on natural images transfers well to medical imaging.

Load-bearing premise

The claim depends on treating the validation set, which was also used for early stopping and model selection, as a measure of generalization; if that set is not a cleanly separated test set, the 88% accuracy figure is not a supported estimate of performance on new scans.

Editorial extensions

If this is right

If the reported accuracy holds on a clean test set, DeepBrainNet could serve as a first-pass four-class triage tool in settings where an MRI radiologist is not immediately available.
A macro AUC near 0.98 means the model ranks classes well even where hard labels are missed, so threshold tuning could trade sensitivity against specificity per tumor type.
The roughly 24 million trainable parameters and the EfficientNetB0 branch's depthwise separable convolutions keep the model light enough for deployment on clinical workstations rather than large GPU clusters.
Per-class results point to meningioma as the most error-prone category, so future data collection or class weighting should target that class first.

Reading between the lines

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

Because no separate held-out test set is described, the 88% figure should be read as a validation-based estimate; a clean test split may land lower.
The paper never compares the hybrid against its own two backbones run alone, so the claimed fusion advantage is not yet isolated; an ablation with identical training would settle it.
Tumor and no-tumor images come from different source repositories, so some of the discriminative signal may be source-specific rather than tumor-specific; testing on an external dataset with matched acquisition is the cleanest check.
The comparison table's baselines [14] and [15] have no corresponding entries in the reference list, so the claim that DeepBrainNet beats those systems cannot be independently verified from this manuscript alone.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

Short version: this is a standard transfer-learning application with an unsupported headline. The architecture is a parallel concatenation of EfficientNetB0 and ResNet50; the paper's own related-work table lists similar ensembles (AlexNet+VGG19, ResNet50+DenseNet201+InceptionV3), so novelty is incremental at best. To its credit, the manuscript is readable and documents preprocessing (resizing, normalization, augmentation, cropping), defines the evaluation metrics, and presents per-class results rather than only a single accuracy number. The FCM feature-selection algorithm is given, although the results section never says whether FCM was actually used in the final pipeline.

The soft spots are not minor. The dataset description in §III-A is internally contradictory: it says the final dataset is 'images from figshare' after removing SARTAJ, but the no-tumor class was taken from Br35H. A figshare-only dataset cannot produce the no-tumor row in Table III or the confusion matrix in Figure 11. If Br35H images were retained, the true dataset is not figshare-only and the per-class counts are unstated. Either way, the input to the experiment is not reproducible.

The numbers are also inconsistent: 88% in the abstract, 89% in Table IV, 93.2% validation accuracy in Section V, 94.7% 'best validation accuracy' elsewhere. There is no defined test split. Early stopping is based on validation loss, and the reported metrics appear to be validation-set metrics, so they are optimistically biased. No code, weights, or seed are provided. The claim that DeepBrainNet 'consistently outperforms' SOTA is contradicted by the paper's own Table I, which lists accuracies of 98.7–99.39%. Finally, Section IV-A describes EfficientNetB0 as if it were a combined EfficientNet/VNet segmentation model, which is a different architecture from what was trained; the fusion of the two backbones is not described.

Who gets value from this? Maybe a reader who wants a template for a simple two-backbone fine-tuning script. The paper does not advance the state of the art, and its central claim cannot be mapped to a defined dataset and split. I would desk reject this in its current form. If the authors resubmit with a consistent dataset description, a clear held-out test set, and code, it could be a minor application note, but the current manuscript is not reproducible enough for peer review.

Referee Report

4 major / 5 minor

Summary. The paper proposes DeepBrainNet, a hybrid deep learning model combining EfficientNetB0 and ResNet50 with transfer learning for four-class brain tumor classification from MRI images. The authors report an overall accuracy of 88%, a weighted F1-score of 88.75%, and a macro AUC-ROC of 98.17% on a publicly available dataset, and claim that this consistently outperforms existing state-of-the-art methods. The manuscript describes dataset collection, preprocessing, fuzzy C-means feature selection, model architecture, training details, and a comparative analysis.

Significance. If the reported results were reproducible and properly validated, the model would be a modest but potentially useful engineering contribution to automated brain tumor classification. The paper's strengths are that it uses publicly available data, provides training details, includes a confusion matrix and ROC curves, and compares with two prior systems. However, the contribution is incremental—a standard late-fusion ensemble of two pretrained CNNs—and the experimental validation as presented is not rigorous enough to support the central claims of consistent superiority and clinical potential.

major comments (4)

[§III-A (Dataset)] The dataset description is internally contradictory. The section first states that the study uses 7023 images from figshare, SARTAJ, and Br35H, with no-tumor images taken from Br35H. It then states that SARTAJ images were removed and 'The final dataset used for this study consists of images from figshare.' If the figshare collection contains only the three tumor classes (as is standard for this dataset), then a figshare-only dataset cannot yield the no-tumor results reported in Table III and the four-class confusion matrix in Figure 11. If Br35H no-tumor images were retained, then the final dataset is not figshare-only and the exact per-class counts and total are not specified. Either way, the input dataset for the experiments is not uniquely defined, which makes the reported 88% accuracy unreproducible.
[§V (Results) and Abstract/Table IV] The central accuracy claim is reported inconsistently and without a defined evaluation split. The Abstract states 88% accuracy, Table IV lists 89% for DeepBrainNet, and Section V reports a validation accuracy of 93.2% (and later 94.7%) for the same model. The manuscript never defines a held-out test set; the models are selected by early stopping on validation loss, so the reported metrics appear to be validation-set metrics that are optimistically biased. Without a clearly separated test partition, the claim that the model 'generalizes well to unseen data' is unsupported.
[§III-D (Feature Selection)] Fuzzy C-means (FCM) feature selection is described in detail and Algorithm 1 is provided, but FCM never appears in the results section, no ablation study is performed, and there is no explanation of how the selected features are integrated into the classification pipeline. As written, the feature-selection component has no observable effect on the empirical results, so its role in the claimed performance is unclear.
[Table IV (Comparative Analysis)] The comparison in Table IV is not adequately grounded. The table claims 'same dataset used in this study,' but given the ambiguity in §III-A, it is unclear which dataset and split were used for the cited baselines [14] and [15]. No details are provided on the preprocessing, augmentation, or evaluation protocol for the baselines, and no error bars or multiple-run statistics are given. The claim that DeepBrainNet 'consistently outperforms' prior work is therefore not supported by the reported evidence.

minor comments (5)

[§IV-A] The description of EfficientNetB0 incorrectly states that it combines EfficientNet and VNet for segmentation; this paper addresses classification, and the quoted text appears to have been reused from a segmentation study.
[Equation (1)] The variables in the resizing equation are inconsistently named: 'Iresized = f(I original, Dnew)' but the definition list mentions 'Original image' rather than 'I original'.
[Algorithm 1] The membership update equation in Algorithm 1 has a typesetting error: 'uij = db(xi, vj)Pc k=1 db(xi, vk)' is missing the division symbol, making the formula unreadable.
[Reference [4]] Reference [4] cites 'arXiv:2304.100394', which appears to contain an extra digit; the correct ID appears to be arXiv:2304.10039.
[Figure 3] The bar chart in Figure 3 is not annotated with axis labels or class counts, making it difficult to verify the class distribution described in the text.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper reports an empirical training-and-evaluation outcome with no derivation chain that reduces to its own inputs.

full rationale

This paper makes no formal derivation claim. Its central result is an empirical measurement: a hybrid EfficientNetB0/ResNet50 model trained with transfer learning and evaluated on brain MRI data, yielding 88% accuracy, an 88.75% weighted F1-score, and a 98.17% macro AUC-ROC. The preprocessing equations (resizing, normalization, CLAHE, histogram equalization) and the fuzzy C-means algorithm are standard operations, not derived results, and none of them is defined in terms of the final accuracy or F1-score. The evaluation metrics such as precision, recall, and F1 are textbook definitions applied to the model's predictions, not quantities that were fitted to produce the reported values. There are no load-bearing self-citations: the cited works are external and, while some are the sources of dataset descriptions and algorithm conventions, the reported performance numbers are not obtained from those citations. No uniqueness theorem from the authors is invoked, and no ansatz is smuggled in via self-citation. The paper does contain a serious internal inconsistency in the dataset description: Section III-A says the final dataset consists of images from figshare, while also stating that no-tumor images were taken from Br35H, which would be needed for the four-class results in Table III and Figure 11. That is a reproducibility and correctness risk, not a circularity of the kind this axis targets. Similarly, the absence of a clearly defined held-out test set and the possibility of validation-based early stopping inflating reported metrics are empirical validity concerns, not circular reasoning. Because the central claim is an experimental outcome with no derivation chain that reduces to its inputs, the appropriate circularity score is 0.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

The central performance claim rests on fitted network weights and hand-chosen hyperparameters rather than on a derivation; no invented entities are proposed.

free parameters (2)

Trained ResNet50 and EfficientNetB0 weights including fusion head = ~24M trainable parameters; final weights not released
The 88% accuracy and AUC values are properties of these fitted weights; no checkpoint or seed is provided.
Training hyperparameters: epochs 40, batch size 32, input 224x224, augmentation settings = As stated in Section V; learning rate and dropout rates not specified
These hand-chosen settings determine the reported scores, and no sensitivity analysis or ablation is given.

assumptions (3)

domain assumption The merged figshare, SARTAJ, and Br35H dataset has accurate labels after removal of misclassified SARTAJ glioma images.
Section III-A states that SARTAJ images were removed due to incorrect categorization, but no independent verification of the remaining labels is described.
domain assumption The validation split is representative and does not leak between training and validation.
Section V reports validation metrics but no split ratio, stratification, or patient-level separation is given.
domain assumption ImageNet-pretrained features transfer to MRI brain images.
Transfer learning from ImageNet is the core mechanism in Sections IV-B and V; this is a plausible but unverified domain assumption for radiology images.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Deep Brain Net: An Optimized Deep Learning Model for Brain tumor Detection in MRI Images Using EfficientNetB0 and ResNet50 with Transfer Learning." pith.science (2026). https://pith.science/paper/WGX55CXO

@misc{pith2026250707011,
  author       = {Pith},
  title        = {Pith review of: Deep Brain Net: An Optimized Deep Learning Model for Brain tumor Detection in MRI Images Using EfficientNetB0 and ResNet50 with Transfer Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WGX55CXO}},
  note         = {Machine review of arXiv:2507.07011}
}

read the original abstract

In recent years, deep learning has shown great promise in the automated detection and classification of brain tumors from MRI images. However, achieving high accuracy and computational efficiency remains a challenge. In this research, we propose Deep Brain Net, a novel deep learning system designed to optimize performance in the detection of brain tumors. The model integrates the strengths of two advanced neural network architectures which are EfficientNetB0 and ResNet50, combined with transfer learning to improve generalization and reduce training time. The EfficientNetB0 architecture enhances model efficiency by utilizing mobile inverted bottleneck blocks, which incorporate depth wise separable convolutions. This design significantly reduces the number of parameters and computational cost while preserving the ability of models to learn complex feature representations. The ResNet50 architecture, pre trained on large scale datasets like ImageNet, is fine tuned for brain tumor classification. Its use of residual connections allows for training deeper networks by mitigating the vanishing gradient problem and avoiding performance degradation. The integration of these components ensures that the proposed system is both computationally efficient and highly accurate. Extensive experiments performed on publicly available MRI datasets demonstrate that Deep Brain Net consistently outperforms existing state of the art methods in terms of classification accuracy, precision, recall, and computational efficiency. The result is an accuracy of 88 percent, a weighted F1 score of 88.75 percent, and a macro AUC ROC score of 98.17 percent which demonstrates the robustness and clinical potential of Deep Brain Net in assisting radiologists with brain tumor diagnosis.

Figures

Figures reproduced from arXiv: 2507.07011 by the authors.

**Figure 1.** Classes of Brain tumor from MRI images [2] The objectives of DeepBrainNet are: • To develop an optimized deep learning model (DeepBrainNet) for brain tumor detection in MRI images, integrating EfficientNetB0 with depthwise separable convolutions and ResNet50 with transfer learning to improve accuracy and computational efficiency. • To benchmark the performance of DeepBrainNet on publicly available MRI datasets, co… view at source ↗

**Figure 2.** Architecture Diagram of DeepBrainNet III. MATERIALS AND METHODS This section presents the materials and methods utilized in conducting the study. A. Dataset The dataset used in this research is a combination of three different datasets: figshare, SARTAJ, and Br35H. It contains a total of 7023 human brain MRI images that are classified into four distinct categories: glioma, meningioma, no tumor, and pituitary tumor. … view at source ↗

**Figure 3.** bar chart of distribution of tumor images in dataset B. Data Pre-processing 1) Image Resizing: The MRI images in the dataset may vary in size. It is essential to resize them to a consistent dimension as input into the deep learning models. This resizing ensures that all images conform to the required input size of the model; in addition, this helps in improving both the computational efficiency and the model perform… view at source ↗

Figures from the paper (9 more)

**Figure 4.** Figure 4: Fuzzy C means Feature Selection Architecture [10] precisely segment the tumor region from the MRI scans. The combination of EfficientNet’s efficient feature extraction with VNet’s robust segmentation capabilities makes EfficientNetB0 an ideal architecture for both cl…

**Figure 5.** Figure 5: Efficient V-net Architecture [11] [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]

**Figure 6.** Figure 6: Transfer Learning on Pretrained Resnet Model [12] C. Evaluation Metrics This subsection provides a comprehensive evaluation of the model’s performance using various metrics, including classification report, confusion matrix, F1 score, and AUC-ROC score. These metrics …

**Figure 10.** Figure 10: Classification Report Metrics Visualization [PITH_FULL_IMAGE:figures/full_fig_p007_10.png]

**Figure 9.** Figure 9: Training and validation accuracy and loss curves and a validation accuracy of 93.2%, which indicates strong performance. The training loss of 0.112 and the validation loss of 0.145 further suggest that the model is fitting the data effectively without overfitting. The …

**Figure 11.** Figure 11: Confusion Matrix The weighted F1-score across all classes, weighted by the number of true instances per class, is 0.88. This value indicates a well-balanced model that performs strongly in both precision and recall, making it wellsuited for multi-class classification…

**Figure 12.** Figure 12: ROC Curve for Multi-Class Classification (One-vs-Rest) model performs well in separating the types of tumor. These evaluation metrics provide a thorough analysis of the model performance. The model demonstrates strong performance across all tumor classes, with high pr…

**Figure 13.** Figure 13: Correctly Predicted Example for Each Class Each image in the plot corresponds to a different class, with the title indicating both the actual label and the predicted label. In cases where the model has correctly predicted the label, we visualise a set of random predic…

**Figure 14.** Figure 14: Random Predictions VI. CONCLUSION In conclusion, the developed model demonstrates strong performance in classifying brain tumors from MRI images, achieving high accuracy and reliable generalization to unseen data. The combination of ResNet50 and EfficientNetB0 archite…

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

12 extracted references · 12 canonical work pages

[1]

MRI brain tumour detection us- ing deep learning and machine learning ap- proaches

S. Anantharajan, S. Gunasekaran, and T. Sub- ramanian. “MRI brain tumour detection us- ing deep learning and machine learning ap- proaches”. In:Measurement: Sensors31.1 (2024), p. 101026.DOI: 10.1016/j.measen.2024. 101026

work page doi:10.1016/j.measen.2024 2024
[2]

Employing deep learn- ing and transfer learning for accurate brain tumour detection

S. K. Mathivanan et al. “Employing deep learn- ing and transfer learning for accurate brain tumour detection”. In:Scientific Reports14.2 (2024), p. 7232.DOI: 10 . 1038 / s41598 - 024 - 57970-7

work page 2024
[3]

Brain tumour detection from images and compar- ison with transfer learning methods and 3- layer CNN

M. Z. Khaliki and M. S. Bas ¸arslan. “Brain tumour detection from images and compar- ison with transfer learning methods and 3- layer CNN”. In:Scientific Reports14.3 (2024), p. 2664.DOI: 10.1038/s41598-024-52823-9

work page doi:10.1038/s41598-024-52823-9 2024
[4]

Brain tumor multi classification and segmentation in MRI images using deep learning

B. Amin et al. “Brain tumour multi clas- sification and segmentation in MRI images using deep learning”. In:arXiv preprint arXiv:2304.100394 (2023).DOI: 10 . 48550 / arXiv.2304.10039

work page Pith review arXiv doi:10.48550/arxiv.2304.10039 2023
[5]

Enhancing brain tumour classification in MRI scans with a multi-layer customized convolutional neural network ap- proach

E. Albalawi et al. “Enhancing brain tumour classification in MRI scans with a multi-layer customized convolutional neural network ap- proach”. In:Frontiers in Computational Neuro- science18.5 (2024), p. 1418546.DOI: 10.3389/ fncom.2024.1418546

work page arXiv 2024
[6]

Automated brain tumour iden- tification in biomedical radiology images: A multi-model ensemble deep learning approach

S. Natha et al. “Automated brain tumour iden- tification in biomedical radiology images: A multi-model ensemble deep learning approach”. In:Applied Sciences14.6 (2024), p. 2210.DOI: 10.3390/app14052210

work page doi:10.3390/app14052210 2024
[7]

Realization of the contrast limited adaptive histogram equalization (CLAHE) for real-time image enhancement

Ali M. Reza. “Realization of the contrast limited adaptive histogram equalization (CLAHE) for real-time image enhancement”. In:Journal of VLSI signal processing systems for signal, im- age and video technology38.7 (2004), pp. 35– 44.DOI: 10.1023/B:VLSI.0000028532.53893. 82

work page arXiv 2004
[8]

A review of histogram equalization techniques in image enhancement application

W. A. Mustafa and M. M. M. Abdul Kader. “A review of histogram equalization techniques in image enhancement application”. In:Journal of Physics: Conference Series. V ol. 1019. 8. IOP Publishing. 2018, p. 012026

work page 2018

Show all 12 references

[9]

Comparison of Fuzzy C-Means, Fuzzy Kernel C-Means, and Fuzzy Kernel Robust C-Means to Classify Thalassemia Data

Zuherman Rustam et al. “Comparison of Fuzzy C-Means, Fuzzy Kernel C-Means, and Fuzzy Kernel Robust C-Means to Classify Thalassemia Data”. In:International Journal on Advanced Science, Engineering and Information Technol- ogy9.9 (2019), p. 1205.DOI: 10.18517/ijaseit. 9.4.9580

2019 doi
[10]

Un- supervised Multiview Fuzzy C-Means Cluster- ing Algorithm

I. Hussain, K. P. Sinaga, and M.-S. Yang. “Un- supervised Multiview Fuzzy C-Means Cluster- ing Algorithm”. In:Electronics12.10 (2023), p. 4467.DOI: 10.3390/electronics12214467

2023 doi
[11]

Efficientnet: Rethinking model scaling for convolutional neu- ral networks

Mingxing Tan and Quoc V . Le. “Efficientnet: Rethinking model scaling for convolutional neu- ral networks”. In:International conference on machine learning. 11. PMLR. 2019, pp. 6105– 6114

2019
[12]

Potato Leaf Disease De- tection and Classification Using VGG16

V . L. Adluri et al. “Potato Leaf Disease De- tection and Classification Using VGG16”. In: Proceedings of International Joint Conference on Advances in Computational Intelligence. 12. Springer Nature Singapore. 2024, pp. 147–158. DOI: 10.1007/978-981-97-0180-3 13

2024 doi

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [1]

MRI brain tumour detection us- ing deep learning and machine learning ap- proaches

S. Anantharajan, S. Gunasekaran, and T. Sub- ramanian. “MRI brain tumour detection us- ing deep learning and machine learning ap- proaches”. In:Measurement: Sensors31.1 (2024), p. 101026.DOI: 10.1016/j.measen.2024. 101026

work page doi:10.1016/j.measen.2024 2024

[2] [2]

Employing deep learn- ing and transfer learning for accurate brain tumour detection

S. K. Mathivanan et al. “Employing deep learn- ing and transfer learning for accurate brain tumour detection”. In:Scientific Reports14.2 (2024), p. 7232.DOI: 10 . 1038 / s41598 - 024 - 57970-7

work page 2024

[3] [3]

Brain tumour detection from images and compar- ison with transfer learning methods and 3- layer CNN

M. Z. Khaliki and M. S. Bas ¸arslan. “Brain tumour detection from images and compar- ison with transfer learning methods and 3- layer CNN”. In:Scientific Reports14.3 (2024), p. 2664.DOI: 10.1038/s41598-024-52823-9

work page doi:10.1038/s41598-024-52823-9 2024

[4] [4]

Brain tumor multi classification and segmentation in MRI images using deep learning

B. Amin et al. “Brain tumour multi clas- sification and segmentation in MRI images using deep learning”. In:arXiv preprint arXiv:2304.100394 (2023).DOI: 10 . 48550 / arXiv.2304.10039

work page Pith review arXiv doi:10.48550/arxiv.2304.10039 2023

[5] [5]

Enhancing brain tumour classification in MRI scans with a multi-layer customized convolutional neural network ap- proach

E. Albalawi et al. “Enhancing brain tumour classification in MRI scans with a multi-layer customized convolutional neural network ap- proach”. In:Frontiers in Computational Neuro- science18.5 (2024), p. 1418546.DOI: 10.3389/ fncom.2024.1418546

work page arXiv 2024

[6] [6]

Automated brain tumour iden- tification in biomedical radiology images: A multi-model ensemble deep learning approach

S. Natha et al. “Automated brain tumour iden- tification in biomedical radiology images: A multi-model ensemble deep learning approach”. In:Applied Sciences14.6 (2024), p. 2210.DOI: 10.3390/app14052210

work page doi:10.3390/app14052210 2024

[7] [7]

Realization of the contrast limited adaptive histogram equalization (CLAHE) for real-time image enhancement

Ali M. Reza. “Realization of the contrast limited adaptive histogram equalization (CLAHE) for real-time image enhancement”. In:Journal of VLSI signal processing systems for signal, im- age and video technology38.7 (2004), pp. 35– 44.DOI: 10.1023/B:VLSI.0000028532.53893. 82

work page arXiv 2004

[8] [8]

A review of histogram equalization techniques in image enhancement application

W. A. Mustafa and M. M. M. Abdul Kader. “A review of histogram equalization techniques in image enhancement application”. In:Journal of Physics: Conference Series. V ol. 1019. 8. IOP Publishing. 2018, p. 012026

work page 2018