pith. sign in

arxiv: 2606.18682 · v1 · pith:SQR2CTDMnew · submitted 2026-06-17 · 💻 cs.CV

Multi-Class Brain Tumor Classification Using Advanced Deep Learning Models: A Comparative Study

Asad Channa , Asghar Ali Chandio , Akhtar Hussain Jalbani , Mehwish Leghari , Shahzad Memon This is my paper

Pith reviewed 2026-06-26 21:36 UTC · model grok-4.3

classification 💻 cs.CV
keywords brain tumor classificationMRI imagesCNN architecturesEfficientNetB0meningioma detectioncomparative studydeep learningmedical image analysis
0
0 comments X

The pith

EfficientNetB0 reaches 95 percent accuracy on multi-class brain tumor classification from MRI and lifts meningioma recall from 20 to 89 percent.

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

The paper compares five CNN architectures on a clinically sourced set of roughly 10,000 MRI images for classifying brain tumors into multiple types. It reports that EfficientNetB0 delivers the highest overall accuracy at 95 percent, ahead of VGG16 at 94.37 percent, VGG19 at 92.29 percent, DenseNet121 at 90.91 percent, and a custom CNN at 78 percent. The largest practical gain appears in meningioma detection, where the advanced model achieves 89 percent recall against roughly 20 percent for the simple CNN. The work also records that the shallower VGG16 outperforms the deeper VGG19, indicating that architectural efficiency can matter more than added depth for this medical imaging task.

Core claim

EfficientNetB0 supplies the best overall performance for multi-class brain tumor classification on MRI, recording 95 percent accuracy together with an 89 percent recall rate on meningiomas, while the same experimental conditions show that model efficiency can outweigh greater depth.

What carries the argument

Side-by-side training and evaluation of VGG16, VGG19, DenseNet121, EfficientNetB0, and a custom CNN on identical MRI data using accuracy plus per-tumor recall as the performance measures.

If this is right

  • EfficientNetB0 supplies the strongest practical option among the tested models for reducing missed meningioma cases.
  • Architectural efficiency can be more decisive than network depth when classifying medical images.
  • Pre-trained models deliver substantially higher per-class recall than a simple custom CNN on this task.
  • The reported ranking of the five architectures rests on both aggregate accuracy and clinically relevant tumor-specific recall.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Hospitals could adopt EfficientNetB0 as a first-line screening tool to lower the rate of undetected meningiomas.
  • The depth-versus-efficiency observation may generalize to other subtle-feature medical imaging problems.
  • Repeating the comparison on multi-center datasets would test whether the 95 percent figure and meningioma recall hold outside the original collection.

Load-bearing premise

The same training setup applied to all models on a single 10,000-image dataset produces performance differences that will hold for other clinical MRI collections.

What would settle it

Retraining the five models on an independent collection of MRI scans and observing that any architecture other than EfficientNetB0 exceeds 95 percent accuracy or that its meningioma recall falls below 80 percent.

Figures

Figures reproduced from arXiv: 2606.18682 by Akhtar Hussain Jalbani, Asad Channa, Asghar Ali Chandio, Mehwish Leghari, Shahzad Memon.

Figure 1
Figure 1. Figure 1: Class-wise distribution of MRI images across the combined dataset after stratified splitting into training, validation, and testing sets. 3.2. Sample MRI Images A single example of an image for each of the four classes is illustrated in [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Representative MRI samples from each diagnostic class. From left to right: (a) Glioma, (b) Meningioma, (c) No Tumor (healthy brain), (d) Pituitary Tumor. Each image demonstrates the typical morphological and enhancement features used for clinical diagnosis. 3.3. Data Preprocessing Each of the MRI images was resized to 224 x 224 pixels so that all images could be fed into the CNN models that have already be… view at source ↗
Figure 3
Figure 3. Figure 3: Demonstration of data augmentation techniques. The panel shows a base meningioma image and its augmented variants generated through random rotation, translation, shear, zoom, horizontal flip, brightness and combined augmentation. These transformations expand the effective training dataset, encouraging the model to learn robust, invariant features. 3.5. Deep Learning Models We tested five different Convolut… view at source ↗
Figure 4
Figure 4. Figure 4: Schematic overview of the deep learning architectures used for comparative brain tumor classification from MRI. (a) Custom Baseline CNN: Architecture of the sequential model built from scratch. (b) Hierarchical Feature Learning: Conceptual illustration of progressive feature extraction from MRI scans. (c) Transfer Learning Framework: Comparative setup of four pre-trained models adapted with a shared classi… view at source ↗
Figure 5
Figure 5. Figure 5: Overview of the experimental workflow for brain tumor classification from MRI images. The pipeline includes dataset integration, preprocessing, data augmentation, deep learning model training, and evaluation using standard performance metrics. We normalized and resized all of the images before training our models, but only used data augmentation on the training set to help improve the generality of our mod… view at source ↗
Figure 6
Figure 6. Figure 6: Visual comparison of accuracy and macro F1-score across different deep learning models evaluated on the test dataset [PITH_FULL_IMAGE:figures/full_fig_p014_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Comparison of meningioma recall across different model architectures In addition to demonstrating high levels of accuracy in detecting gliomas, all models had high levels of recall ranging from 87 percent to 94 percent. It is also interesting to note that the Baseline CNN [PITH_FULL_IMAGE:figures/full_fig_p015_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Confusion matrices for (a) Baseline CNN, (b) VGG16, (c) VGG19, (d) DenseNet121, and (e) EfficientNetB0 evaluated on the independent test set [PITH_FULL_IMAGE:figures/full_fig_p016_8.png] view at source ↗
read the original abstract

Despite recent advancements in deep learning, accurately classifying brain tumors from MRI images continues to pose challenges. In this research, we present a comprehensive evaluation of five different convolutional neural networks (CNN) architectures, including a customized baseline model and four pre-trained models - for use in classifying multi-class brain tumors using a clinically-sourced dataset of approximately 10,000 MRI images. We have utilized five different architectures; VGG16, VGG19, DenseNet121, and EfficientNetB0, which were all tested and trained within an identical experimental framework. Performance was measured by both overall accuracy and tumor-wise recall as a means to measure the clinically-relevant performance of each architecture. We found that EfficientNetB0 had the best overall classification accuracy at 95%, when compared to the other architectures tested; specifically VGG16 (94.37%), VGG19 (92.29%), DenseNet121 (90.91%) and the customized CNN (78.00%). An especially important finding of our research was the considerable improvement in detecting meningiomas; specifically, while simple CNNs could detect meningiomas with a recall rate of approximately 20%, EfficientNetB0 was able to detect meningiomas with a recall rate of 89%. Meningiomas are often difficult to detect because they can appear very subtly on MRI images. Additionally, an interesting finding was that the deeper VGG19 performed worse than the shallower VGG16. This indicates that in many cases the architectural efficiency of a CNN model may be more important than its depth when working with medical images. Overall, EfficientNetB0 appears to provide the optimal trade-off between classification accuracy, number of parameters used in the model and clinically meaningful performance.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 2 minor

Summary. The manuscript evaluates five CNN architectures (a custom baseline CNN plus VGG16, VGG19, DenseNet121, and EfficientNetB0) for multi-class brain tumor classification on a clinically sourced dataset of approximately 10,000 MRI images. All models are stated to have been trained and tested under an identical experimental framework. The central empirical claims are that EfficientNetB0 attains the highest overall accuracy (95 %) and markedly higher meningioma recall (89 %) than the custom CNN (~20 %), with secondary observations that VGG16 outperforms the deeper VGG19 and that EfficientNetB0 offers the best accuracy–parameter trade-off.

Significance. If the reported performance differences can be reproduced under documented, identical conditions and on a properly stratified dataset, the results would supply a concrete, clinically relevant ranking of standard architectures for meningioma detection, where subtle appearance makes recall especially important. The finding that architectural efficiency can outweigh depth on medical images is also potentially useful for resource-constrained clinical deployment.

major comments (3)
  1. [Abstract] Abstract and Methods: the abstract reports accuracy and recall figures (EfficientNetB0 95 %, meningioma recall 89 %) but supplies no information on train/validation/test splits, cross-validation procedure, or class-imbalance handling. Without these details the central performance claims cannot be verified or compared across architectures.
  2. [Abstract] Abstract: the claim that all five architectures were evaluated 'within an identical experimental framework' is load-bearing for attributing the observed differences (e.g., 95 % vs. 78 %) to model choice rather than to differing preprocessing, data augmentation, learning-rate schedules, or early-stopping criteria; no supporting protocol is provided.
  3. [Abstract] Abstract: meningioma recall is highlighted as clinically important, yet no per-class precision, F1, or confusion-matrix data are supplied, nor is any statistical significance test reported for the 89 % vs. ~20 % difference, leaving the strength of the improvement unquantified.
minor comments (2)
  1. [Abstract] The exact dataset size, class distribution, and source (hospital, scanner types) are given only approximately; these details are needed for reproducibility and to assess generalizability.
  2. [Abstract] No mention is made of whether the pre-trained models were fine-tuned from ImageNet weights or trained from scratch, nor of the optimizer, batch size, or number of epochs used.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive comments highlighting the need for greater methodological transparency in the abstract. We agree that these details strengthen the manuscript and will revise the abstract and, where appropriate, the methods section to incorporate the requested information on splits, protocols, and metrics.

read point-by-point responses

  1. Referee: [Abstract] Abstract and Methods: the abstract reports accuracy and recall figures (EfficientNetB0 95 %, meningioma recall 89 %) but supplies no information on train/validation/test splits, cross-validation procedure, or class-imbalance handling. Without these details the central performance claims cannot be verified or compared across architectures.

    Authors: We agree these details belong in the abstract. The dataset was partitioned via stratified sampling into 70 % training, 15 % validation and 15 % test sets to preserve class proportions; no cross-validation was performed owing to computational cost. We will add a concise statement of the split and stratification procedure to the abstract. revision: yes

  2. Referee: [Abstract] Abstract: the claim that all five architectures were evaluated 'within an identical experimental framework' is load-bearing for attributing the observed differences (e.g., 95 % vs. 78 %) to model choice rather than to differing preprocessing, data augmentation, learning-rate schedules, or early-stopping criteria; no supporting protocol is provided.

    Authors: All models shared identical preprocessing (resize to 224×224, ImageNet mean/std normalization), the same augmentation pipeline (random rotation ±15°, horizontal flip, zoom 0.8–1.2), Adam optimizer (lr = 0.001, batch size 32), and early-stopping patience of 10 epochs on validation loss. We will insert a brief protocol summary into the abstract. revision: yes

  3. Referee: [Abstract] Abstract: meningioma recall is highlighted as clinically important, yet no per-class precision, F1, or confusion-matrix data are supplied, nor is any statistical significance test reported for the 89 % vs. ~20 % difference, leaving the strength of the improvement unquantified.

    Authors: We will add per-class precision and F1 scores for all models to the abstract. The full manuscript already contains the confusion matrix; we will ensure it is clearly referenced. No statistical significance test was performed originally; we can report the raw difference magnitude or add a post-hoc test in revision if the editor deems it essential. revision: partial

Circularity Check

0 steps flagged

No significant circularity

full rationale

The paper reports empirical classification accuracies and recall rates from training five CNN architectures (VGG16, VGG19, DenseNet121, EfficientNetB0, and a custom CNN) on a fixed ~10k-image MRI dataset under an asserted identical experimental framework. No equations, derivations, fitted parameters, or predictions are presented that could reduce to inputs by construction. All load-bearing claims rest on measured performance differences, with no self-citation chains, ansatzes, or uniqueness theorems invoked. This is a standard empirical comparison study with no circularity.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The paper is an empirical ML comparison and therefore rests on standard supervised-learning assumptions plus the unstated premise that the single reported run per architecture is stable. No free parameters are explicitly fitted to produce the headline numbers beyond ordinary training; no new entities are postulated.

axioms (2)
  • domain assumption The MRI images are independent and identically distributed within each tumor class.
    Implicit in any supervised classification claim on a fixed dataset; location not stated because abstract only.
  • domain assumption The training procedure (optimizer, learning rate schedule, data augmentation) is identical and fair across all five models.
    Stated in the abstract as 'identical experimental framework' but no further detail given.

pith-pipeline@v0.9.1-grok · 5868 in / 1404 out tokens · 19170 ms · 2026-06-26T21:36:46.673014+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

56 extracted references · 23 canonical work pages

  1. [1]

    Dorfner, F.J., Patel, J.B., Kalpathy -Cramer, J. et al. A review of deep learning for brain tumor analysis in MRI. npj Precis. Onc. 9, 2 (2025). https://doi.org/10.1038/s41698-024-00789-2

    work page doi:10.1038/s41698-024-00789-2 2025

  2. [2]

    CBTRUS Statistical Report: Primary Brain and Other Central Nervous System Tumors Diagnosed in the Uni ted States in 2017 -2021

    Price M, Ballard C, Benedetti J, et al. CBTRUS Statistical Report: Primary Brain and Other Central Nervous System Tumors Diagnosed in the Uni ted States in 2017 -2021. Neuro Oncol. 2024;26(Supplement_6):vi1-vi85. doi:10.1093/neuonc/noae145

    work page doi:10.1093/neuonc/noae145 2017

  3. [3]

    Assessing inter-rater reliability of MRI features in glioma: a multi -radiologist agre ement study

    Mohebbi A, Mohammadzadeh S, Zare AH, Moradi Z, Ardakani AA, Mohammadi A. Assessing inter-rater reliability of MRI features in glioma: a multi -radiologist agre ement study. BMC Medical Imaging. 2025 Dec;25(1):480

    2025

  4. [4]

    Evaluating Brain Tumor Detection with Deep Learning Convolutional Neural Networks Across Multiple MRI Modalities

    Stathopoulos I, Serio L, Karavasilis E, et al. Evaluating Brain Tumor Detection with Deep Learning Convolutional Neural Networks Across Multiple MRI Modalities. J Imaging . 2024;10(12):296. Published 2024 Nov 21. doi:10.3390/jimaging10120296

    work page doi:10.3390/jimaging10120296 2024

  5. [5]

    Tareco Bucho, James A

    Iris van der Loo, Teresa M. Tareco Bucho, James A. Hanley, Regina G.H. Beets-Tan, Alex L.T. Imholz, Stefano Trebeschi, Measurement variability of radiologists when measuring brain tumors, European Journal of Radiology , Volume 183, 2025, 111874, ISSN 0720-048X, https://doi.org/10.1016/j.ejrad.2024.111874

    work page doi:10.1016/j.ejrad.2024.111874 2025

  6. [6]

    Repeatability of radiomic features in magnetic resonance imaging of glioblastoma: Test -retest and image registration analyses

    Shiri I, Hajianfar G, Sohrabi A, et al. Repeatability of radiomic features in magnetic resonance imaging of glioblastoma: Test -retest and image registration analyses. Med Phys . 2020;47(9):4265-4280. doi:10.1002/mp.14368

    work page doi:10.1002/mp.14368 2020

  7. [7]

    Interobserver variabil ity in assessing preoperative imaging biomarkers for cerebellar mutism syndrome: a multiobserver pilot study

    Laustsen AF, Dineen R, Ilginiene J, et al. Interobserver variabil ity in assessing preoperative imaging biomarkers for cerebellar mutism syndrome: a multiobserver pilot study. Pediatric Radiology 2025;55:1915–1926. https://doi.org/10.1007/s00247-025-06326-y

    work page doi:10.1007/s00247-025-06326-y 2025

  8. [8]

    Salama A, Daradkeh YI, Javeed D, Ur Rehman A, Hamam H

    Raza A, Ayub H, Khan JA, Ahmad I, S. Salama A, Daradkeh YI, Javeed D, Ur Rehman A, Hamam H. A hybrid deep learning -based approach for brain tumor classification. Electronics. 2022 Apr 5;11(7):1146

    2022

  9. [9]

    Enhanced brain tumor classification using graph convolutional neural network architecture

    Ravinder M, Saluja G, Allabun S, Alqahtani MS, Abbas M, Othman M, Soufiene BO. Enhanced brain tumor classification using graph convolutional neural network architecture. Scientific Reports. 2023 Sep 11;13(1):14938

    2023

  10. [10]

    Brain tumor classification from MRI using image enhancement and convolutional neural network techniques

    Rasheed Z, Ma YK, Ullah I, Ghadi YY, Khan MZ, Khan MA, Abdusalomov A, Alqahtani F, Shehata AM. Brain tumor classification from MRI using image enhancement and convolutional neural network techniques. Brain Sciences. 2023 Sep 14;13(9):1320

    2023

  11. [11]

    Convolutional Neural Network in Medical Image Analysis: A Review

    Kshatri, S.S., Singh, D. Convolutional Neural Network in Medical Image Analysis: A Review. Arch Computat Methods Eng 30, 2793 –2810 (2023). https://doi.org/10.1007/s11831-023-09898-w

    work page doi:10.1007/s11831-023-09898-w 2023

  12. [12]

    Development of hybrid models based on deep learning and optimized machine learning algorithms for brain tumor Multi -Classification

    Celik M, Inik O. Development of hybrid models based on deep learning and optimized machine learning algorithms for brain tumor Multi -Classification. Expert Systems with Applications. 2024 Mar 15;238:122159

    2024

  13. [13]

    T1 -weighted MRI-based brain tumor classification using hybrid deep learning models

    Ilani MA, Shi D, Banad YM. T1 -weighted MRI-based brain tumor classification using hybrid deep learning models. Scientific Reports. 2025 Feb 27;15(1):7010

    2025

  14. [14]

    Niakan Kalhori S

    Saeedi S, Rezayi S, Keshavarz H, R. Niakan Kalhori S. MRI-based brain tumor detection using convolutional deep learning methods and chosen machine learning techniques. BMC Medical Informatics and Decision Making. 2023 Jan 23;23(1):16. https://doi.org/10.1186/s12911-023-02114-6

    work page doi:10.1186/s12911-023-02114-6 2023

  15. [15]

    Imagenet: A large-scale hierarchical image database

    Deng J, Dong W, Socher R, Li LJ, Li K, Fei -Fei L. Imagenet: A large-scale hierarchical image database. In2009 IEEE conference on computer vision and pattern recognition 2009 Jun 20 (pp. 248-255). Ieee

    2009

  16. [16]

    Brain tumor classification for MR images using transfer learning and fine -tuning

    Swati ZN, Zhao Q, Kabir M, Ali F, Ali Z, Ahmed S, Lu J. Brain tumor classification for MR images using transfer learning and fine -tuning. Computerized Medical Imaging and Graphics. 2019 Jul 1;75:34-46

    2019

  17. [17]

    Advanced brain tumor classi fication in MR images using transfer learning and pre -trained deep CNN models

    Disci R, Gurcan F, Soylu A. Advanced brain tumor classi fication in MR images using transfer learning and pre -trained deep CNN models. Cancers. 2025 Jan 2;17(1):121. https://doi.org/10.3390/cancers17010121

    work page doi:10.3390/cancers17010121 2025

  18. [18]

    Transfer learning for medical image classification: a literature review

    Kim HE, Cosa-Linan A, Santhanam N, Jannesari M, Maros ME, Ganslandt T. Transfer learning for medical image classification: a literature review. BMC medical imaging. 2022 Apr 13;22(1):69. https://doi.org/10.1186/s12880-022-00793-7

    work page doi:10.1186/s12880-022-00793-7 2022

  19. [19]

    Brain tumor classification using VGG 16, ResNet50, and inception V3 transfer learning models

    Pillai R, Sharma A, Sharma N, Gupta R. Brain tumor classification using VGG 16, ResNet50, and inception V3 transfer learning models. In2023 2nd International Conference for Innovation in Technology (INOCON) 2023 Mar 3 (pp. 1-5). IEEE

    2023

  20. [20]

    Abnormal brain tumors classification using resnet50 and its comprehensive evaluation

    Younis A, Li Q, Afzal Z, Adamu MJ, Kawuwa HB, Hussain F, Hussain H. Abnormal brain tumors classification using resnet50 and its comprehensive evaluation. IEEE Access. 2024 May 21;12:78843-53

    2024

  21. [21]

    Brainnet: precision brain tumor classification with optimized efficientnet architecture

    Islam MM, Talukder MA, Uddin MA, Akhter A, Khalid M. Brainnet: precision brain tumor classification with optimized efficientnet architecture. International Journal of Intelligent Systems. 2024;2024(1):3583612

    2024

  22. [22]

    Comparative study on architecture of deep neural networks for segmentation of brain tumor using magnetic resonance images

    Preetha R, Priyadarsini MJ, Nisha JS. Comparative study on architecture of deep neural networks for segmentation of brain tumor using magnetic resonance images. IEEE Access. 2023 Dec 7;11:138549-67

    2023

  23. [23]

    Multiple Brain Tumor Classification with Dense CNN Architecture Using Brain MRI Images

    Özkaraca, O.; Bağrıaçık, O.İ.; Gürüler, H.; Khan, .; Hussain, J.; Khan, J.; Laila, U.e. Multiple Brain Tumor Classification with Dense CNN Architecture Using Brain MRI Images. Life 2023, 13, 349. https://doi.org/10.3390/life13020349

    work page doi:10.3390/life13020349 2023

  24. [24]

    Multi -classification of brain tumor MRI images using deep convolutional neural network with fully optimized framework

    Irmak E. Multi -classification of brain tumor MRI images using deep convolutional neural network with fully optimized framework. Iranian Journal of Science and Technology, Transactions of Electrical Engineering. 2021 Sep;45(3):1015 -36. https://doi.org/10.1007/s40998-021-00426-9

    work page doi:10.1007/s40998-021-00426-9 2021

  25. [25]

    Brain tumour detection using machine and deep learning: a systematic review

    Rasool N, Bhat JI. Brain tumour detection using machine and deep learning: a systematic review. Multimedia tools and applications. 2025 Apr;84(13):11551-604

    2025

  26. [26]

    Convolutional neural networks in medical image understanding: a survey

    Sarvamangala DR, Kulkarni RV. Convolutional neural networks in medical image understanding: a survey. Evolutionary intelligence. 2022 Mar;15(1):1 -22. https://doi.org/10.1007/s12065-020-00540-3

    work page doi:10.1007/s12065-020-00540-3 2022

  27. [27]

    Susanto, A.; Sari, C.A.; Rahmalan, H.; Doheir, M. a. S. Support Vector Machine Based Discrete Wavelet Transform for Magnetic Resonance Imaging Brain Tumor Classification. TELKOMNIKA (Telecom munication Computing Electronics and Control) 2023, 21, 592, doi:10.12928/telkomnika.v21i3.24928

    work page doi:10.12928/telkomnika.v21i3.24928 2023

  28. [28]

    Early Brain Tumor Detection Using Random Forest Classification

    Upadhyay, A.; Palival, U.; Jaiswal, S. Early Brain Tumor Detection Using Random Forest Classification. In Advances in intelligent systems and computing; 2020; pp. 258–264

    2020

  29. [29]

    -N.; Guillevin, C.; Lallouette, A.; Thomas, C.; Rittano, G.; Wager, M.; Guillevin, R.; Vallée, A

    Vallée, R.; Vallée, J. -N.; Guillevin, C.; Lallouette, A.; Thomas, C.; Rittano, G.; Wager, M.; Guillevin, R.; Vallée, A. Machine Learning Decision Tree Models for Multiclass Classification of Common Malignant Brain Tumors Using Perfusion and Spectroscopy MRI Data. Frontiers in Oncology 2023, 13, doi:10.3389/fonc.2023.1089998

    work page doi:10.3389/fonc.2023.1089998 2023

  30. [30]

    Brain tumor detection from MRI images using Naive classifier

    Divyamary D, Gopika S, Pradeeba S, Bhuvaneswari M. Brain tumor detection from MRI images using Naive classifier. In2020 6th international conference on advanced computing and communication systems (ICACCS) 2020 Mar 6 (pp. 620-622). IEEE

    2020

  31. [31]

    Magnetic Resonance Imaging (MRI) Brain Tumor Image Classification Based on Five Machine Learning Algorithms

    Jiang, S.; Gu, Y.; Kumar, E. Magnetic Resonance Imaging (MRI) Brain Tumor Image Classification Based on Five Machine Learning Algorithms. Cloud Computing and Data Science 2023, 122–133, doi:10.37256/ccds.4220232740

    work page doi:10.37256/ccds.4220232740 2023

  32. [32]

    and Mengko, T.R., 2018, May

    Abiwinanda, N., Hanif, M., Hesaputra, S.T., Handayani, A. and Mengko, T.R., 2018, May. Brain tumor classification using convolutional neural network. In World Congress on Medical Physics and Biomedical Engineering 2018: June 3 –8, 2018, Prague, Czech Republic (Vol. 1) (pp. 183-189). Singapore: Springer Nature Singapore

    2018

  33. [33]

    A Novel Convolutional Neural Network‐based Approach for Brain Tumor Classification Using Magnetic Resonance Images

    Cinar, N.; Ka a, M.; Ka a, B. A Novel Convolutional Neural Network‐based Approach for Brain Tumor Classification Using Magnetic Resonance Images. International J ournal of Imaging Systems and Technology 2022, 33, 895–908, doi:10.1002/ima.22839

    work page doi:10.1002/ima.22839 2022

  34. [34]

    Multi -grade brain tumor classification using deep CNN with extensive data augmentation

    Sajjad M, Khan S, Muhammad K, Wu W, Ullah A, Baik SW. Multi -grade brain tumor classification using deep CNN with extensive data augmentation. Journal of computational science. 2019 Jan 1;30:174-82

    2019

  35. [35]

    Capsule networks for brain tumor classification based on MRI images and coarse tumor boundaries

    Afshar P, Plataniotis KN, Mohammadi A. Capsule networks for brain tumor classification based on MRI images and coarse tumor boundaries. InICASSP 2019 -2019 IEEE international conference on acoustics, speech and signal processing (ICASSP) 2019 May 12 (pp. 1368-1372). IEEE

    2019

  36. [36]

    Deep learning for brain tumor classification

    Paul JS, Plassard AJ, Landman BA, Fabbri D. Deep learning for brain tumor classification. InMedical Imaging 2017: Biomedical Applications in Molecular, Structural, and Functional Imaging 2017 Mar 13 (Vol. 10137, pp. 253-268). SPIE

    2017

  37. [37]

    Brain Tumor Classification in MRI Image Using Convolutional Neural Network

    Khan, H.A.; Jue, W.; Mushtaq, M.; Mushtaq, M.U. Brain Tumor Classification in MRI Image Using Convolutional Neural Network. Mathematical Biosciences & Engineering 2020, 17, 6203–6216, doi:10.3934/mbe.2020328

    work page doi:10.3934/mbe.2020328 2020

  38. [38]

    Deep transfer learning approaches in performance analysis of brain tumor classification using MRI images

    Srinivas C, KS NP, Zakariah M, Alothaibi YA, Shaukat K, Partibane B, Awal H. Deep transfer learning approaches in performance analysis of brain tumor classification using MRI images. Journal of Healthcare Engineering. 2022;2022(1):3264367

    2022

  39. [39]

    Rasheed, M., Jaffar, M.A., Akram, A. et al. Improved brain tumor classification through DenseNet121 based transfer learning. Discov Onc 16, 1645 (2025). https://doi.org/10.1007/s12672-025-03501-3

    work page doi:10.1007/s12672-025-03501-3 2025

  40. [40]

    Brain Tumor Detection and Prediction in MRI Images Utilizing a Fine-Tuned Transfer Learning Model Integrated Within Deep Learning Frameworks

    Rastogi, D.; Johri, P.; Donelli, M.; Kumar, L.; Bindewari, S.; Raghav, A.; Khatri, S.K. Brain Tumor Detection and Prediction in MRI Images Utilizing a Fine-Tuned Transfer Learning Model Integrated Within Deep Learning Frameworks. Life 2025, 15, 327. https://doi.org/10.3390/life15030327

    work page doi:10.3390/life15030327 2025

  41. [41]

    A hybrid transfer learning framework for brain tumor diagnosis

    Tonni SI, Sheakh MA, Tahosin MS, Hasan MZ, Shuva TF, Bhuiyan T, Almoyad MA, Orka NA, Rahman MT, Khan RT, Kaiser MS. A hybrid transfer learning framework for brain tumor diagnosis. Advanced Intelligent Systems. 2025 Mar;7(3):2400495

    2025

  42. [42]

    A transfer learning-based approach for brain tumor classification

    Bibi N, Wahid F, Ma Y, Ali S, Abbasi IA, Alkhayyat A. A transfer learning-based approach for brain tumor classification. IEEE Access. 2024 Jul 9;12:111218-38

    2024

  43. [43]

    Brain tumor identification using data augmentation and transfer learning approach

    Kumar KK, Dinesh PM, Rayavel P, Vijayaraja L, Dhanasekar R, Kesavan R, Raju K, Khan AA, Wechtaisong C, Haq MA, Alzamil ZS. Brain tumor identification using data augmentation and transfer learning approach. Computer Systems Science & Engineering. 2023 Aug 1;46(2)

    2023

  44. [44]

    Brain Tumor Classification Using Dense Efficient-Net

    Nayak, D.R.; Padhy, N.; Mall ick, P.K.; Zymbler, M.; Kumar, S. Brain Tumor Classification Using Dense Efficient-Net. Axioms 2022, 11, 34. https://doi.org/10.3390/axioms11010034

    work page doi:10.3390/axioms11010034 2022

  45. [45]

    Comparative analysis for accurate multi -classification of brain tumor based on significant deep learning models

    Elhadidy MS, Elgohr AT, El -Geneedy M, Akram S, Kasem HM. Comparative analysis for accurate multi -classification of brain tumor based on significant deep learning models. Computers in Biology and Medicine. 2025 Apr 1;188:109872

    2025

  46. [46]

    brain tumor dataset

    Cheng, Jun (2017). brain tumor dataset. figshare. Dataset. https://doi.org/10.6084/m9.figshare.1512427.v5

    work page doi:10.6084/m9.figshare.1512427.v5 2017

  47. [47]

    Very deep convolutional networks for large-scale image recognition

    Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556. 2014 Sep 4

    Pith/arXiv arXiv 2014

  48. [48]

    Densely connected convo lutional networks

    Huang G, Liu Z, Van Der Maaten L, Weinberger KQ. Densely connected convo lutional networks. InProceedings of the IEEE conference on computer vision and pattern recognition 2017 (pp. 4700-4708)

    2017

  49. [49]

    Efficientnet: Rethinking model scaling for convolutional neural networks

    Tan M, Le Q. Efficientnet: Rethinking model scaling for convolutional neural networks. InInternational conference on machine learning 2019 May 24 (pp. 6105-6114). PMLR

    2019

  50. [50]

    Adam: A method for stochastic optimization

    Kingma DP. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980. 2014

    Pith/arXiv arXiv 2014

  51. [51]

    Information Processing and Management45(4), 427–437 (2009)

    Sokolova, M.; Lapalme, G. A Systematic Analysis of Performance Measures for Classification Tasks. Information Processing & Management 2009, 45, 427 –437, doi:10.1016/j.ipm.2009.03.002

    work page doi:10.1016/j.ipm.2009.03.002 2009

  52. [52]

    Identification and prediction of brain tumor using VGG-16 empowered with explainable artificial intelligence

    Ahmed F, Asif M, Saleem M, Mushtaq UF, Imran M. Identification and prediction of brain tumor using VGG-16 empowered with explainable artificial intelligence. International Journal of Computational and Innovative Sciences. 2023 Jun 30;2(2):24-33

    2023

  53. [53]

    Brain Tumor Classification Based on Fine-Tuned Models and the Ensemble Method

    Noreen N, Palaniappan S, Qayyum A, Ahmad I, Alassafi MO. Brain Tumor Classification Based on Fine-Tuned Models and the Ensemble Method. Computers, Materials & Continua. 2021 Jun 1;67(3)

    2021

  54. [54]

    Multi -class classification of brain tumor types from MR images using EfficientNets

    Zulfiqar F, Bajwa UI, Mehmood Y. Multi -class classification of brain tumor types from MR images using EfficientNets. Biomedical Signal Processing and Control. 2023 Jul 1;84:104777

    2023

  55. [55]

    Refined automatic brain tumor classification using hybrid convolutional neural networks for MRI scans

    AlTahhan FE, Khouqeer GA, Saadi S, Elgarayhi A, Sallah M. Refined automatic brain tumor classification using hybrid convolutional neural networks for MRI scans. Diagnostics. 2023 Feb 23;13(5):864

    2023

  56. [56]

    Brain tumor detection and classification using an optimized convolutional neural network

    Aamir M, Namoun A, Munir S, Aljohani N, Alanazi MH, Alsahafi Y, Alotibi F. Brain tumor detection and classification using an optimized convolutional neural network. Diagnostics. 2024 Aug 7;14(16):1714

    2024