Enhancing Brain Tumor Classification Accuracy through Multi-Modal MRI Analysis and Advanced Ensemble Deep Learning

Ayyoub Berra; Oussama Bouguerra; Bilal Attallah; Youcef Brik; Mohamed Djerioui; Ibraheem Shayea; Saleh Ibrahim Al-Zahrani; Youssouf Bouzidi; Abdelwahhab Boudjelal

doi:10.1109/ICATEEE68170.2025.11406230

Enhancing Brain Tumor Classification Accuracy through Multi-Modal MRI Analysis and Advanced Ensemble Deep Learning

Ayyoub Berra^*
, Oussama Bouguerra
, Bilal Attallah
, Youcef Brik
, Mohamed Djerioui
, Ibraheem Shayea
, Saleh Ibrahim Al-Zahrani
, Youssouf Bouzidi
, Abdelwahhab Boudjelal

^*Corresponding author for this work

Faculty of Electrical and Electronics

University of M'sila
Imam Abdulrahman Bin Faisal University
École d’ingénieurs CESI

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

This paper focused on the multi-class brain tumor classification problem, using deep learning methodologies based on magnetic resonance imaging (MRI) images. Two state-of-the-art deep learning models, EfficientNet-B3 and ConvNeXt-Tiny, were fine-tuned through transfer learning with custom classifier heads, and used a publicly available dataset from Kaggle, leveraging various data augmentation techniques to help reduce class imbalance. These two models were trained to the same hyperparameters and had the same metrics applied to the evaluation. Both models were assessed using Accuracy, F1-Score, Recall, and Precision. EfficientNet-B3 achieved an accuracy of 96.85% and ConvNeXt-Tiny achieved 95.88%. After observing the strengths of both models, a weighted ensemble (60% EfficientNet-B3 and 40% ConvNeXt-Tiny) was created, achieving a final accuracy of 97.07%. Based on the results of the study, it demonstrates how deep learning and ensemble techniques can improve diagnostic accuracy in a complex multi-class brain tumor classification task.

Original language	English
Title of host publication	2nd International Conference on Advanced Technology in Electronic and Electrical Engineering, ICATEEE 2025
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9798331563349
DOIs	https://doi.org/10.1109/ICATEEE68170.2025.11406230
Publication status	Published - 2025
Event	2nd International Conference on Advanced Technology in Electronic and Electrical Engineering, ICATEEE 2025 - M�sila, Algeria Duration: 10 Dec 2025 → 11 Dec 2025

Publication series

Name	2nd International Conference on Advanced Technology in Electronic and Electrical Engineering, ICATEEE 2025

Conference

Conference	2nd International Conference on Advanced Technology in Electronic and Electrical Engineering, ICATEEE 2025
Country/Territory	Algeria
City	M�sila
Period	10/12/25 → 11/12/25

Bibliographical note

Publisher Copyright:
© 2025 IEEE.

Keywords

Brain Tumor Classification
ConvNeXt-Tiny
Deep Learning
EfficientNet-B3
Ensemble Learning
Image Classification
MRI
Weight Averaging

Access to Document

10.1109/ICATEEE68170.2025.11406230

Cite this

Berra, A., Bouguerra, O., Attallah, B., Brik, Y., Djerioui, M., Shayea, I., Al-Zahrani, S. I., Bouzidi, Y., & Boudjelal, A. (2025). Enhancing Brain Tumor Classification Accuracy through Multi-Modal MRI Analysis and Advanced Ensemble Deep Learning. In 2nd International Conference on Advanced Technology in Electronic and Electrical Engineering, ICATEEE 2025 (2nd International Conference on Advanced Technology in Electronic and Electrical Engineering, ICATEEE 2025). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICATEEE68170.2025.11406230

Berra, Ayyoub ; Bouguerra, Oussama ; Attallah, Bilal et al. / Enhancing Brain Tumor Classification Accuracy through Multi-Modal MRI Analysis and Advanced Ensemble Deep Learning. 2nd International Conference on Advanced Technology in Electronic and Electrical Engineering, ICATEEE 2025. Institute of Electrical and Electronics Engineers Inc., 2025. (2nd International Conference on Advanced Technology in Electronic and Electrical Engineering, ICATEEE 2025).

@inproceedings{355e25c7dd0c40aea602b530d645900e,

title = "Enhancing Brain Tumor Classification Accuracy through Multi-Modal MRI Analysis and Advanced Ensemble Deep Learning",

abstract = "This paper focused on the multi-class brain tumor classification problem, using deep learning methodologies based on magnetic resonance imaging (MRI) images. Two state-of-the-art deep learning models, EfficientNet-B3 and ConvNeXt-Tiny, were fine-tuned through transfer learning with custom classifier heads, and used a publicly available dataset from Kaggle, leveraging various data augmentation techniques to help reduce class imbalance. These two models were trained to the same hyperparameters and had the same metrics applied to the evaluation. Both models were assessed using Accuracy, F1-Score, Recall, and Precision. EfficientNet-B3 achieved an accuracy of 96.85\% and ConvNeXt-Tiny achieved 95.88\%. After observing the strengths of both models, a weighted ensemble (60\% EfficientNet-B3 and 40\% ConvNeXt-Tiny) was created, achieving a final accuracy of 97.07\%. Based on the results of the study, it demonstrates how deep learning and ensemble techniques can improve diagnostic accuracy in a complex multi-class brain tumor classification task.",

keywords = "Brain Tumor Classification, ConvNeXt-Tiny, Deep Learning, EfficientNet-B3, Ensemble Learning, Image Classification, MRI, Weight Averaging",

author = "Ayyoub Berra and Oussama Bouguerra and Bilal Attallah and Youcef Brik and Mohamed Djerioui and Ibraheem Shayea and Al-Zahrani, \{Saleh Ibrahim\} and Youssouf Bouzidi and Abdelwahhab Boudjelal",

note = "Publisher Copyright: {\textcopyright} 2025 IEEE.; 2nd International Conference on Advanced Technology in Electronic and Electrical Engineering, ICATEEE 2025 ; Conference date: 10-12-2025 Through 11-12-2025",

year = "2025",

doi = "10.1109/ICATEEE68170.2025.11406230",

language = "English",

series = "2nd International Conference on Advanced Technology in Electronic and Electrical Engineering, ICATEEE 2025",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2nd International Conference on Advanced Technology in Electronic and Electrical Engineering, ICATEEE 2025",

address = "United States",

}

Berra, A, Bouguerra, O, Attallah, B, Brik, Y, Djerioui, M, Shayea, I, Al-Zahrani, SI, Bouzidi, Y & Boudjelal, A 2025, Enhancing Brain Tumor Classification Accuracy through Multi-Modal MRI Analysis and Advanced Ensemble Deep Learning. in 2nd International Conference on Advanced Technology in Electronic and Electrical Engineering, ICATEEE 2025. 2nd International Conference on Advanced Technology in Electronic and Electrical Engineering, ICATEEE 2025, Institute of Electrical and Electronics Engineers Inc., 2nd International Conference on Advanced Technology in Electronic and Electrical Engineering, ICATEEE 2025, M�sila, Algeria, 10/12/25. https://doi.org/10.1109/ICATEEE68170.2025.11406230

Enhancing Brain Tumor Classification Accuracy through Multi-Modal MRI Analysis and Advanced Ensemble Deep Learning. / Berra, Ayyoub; Bouguerra, Oussama; Attallah, Bilal et al.
2nd International Conference on Advanced Technology in Electronic and Electrical Engineering, ICATEEE 2025. Institute of Electrical and Electronics Engineers Inc., 2025. (2nd International Conference on Advanced Technology in Electronic and Electrical Engineering, ICATEEE 2025).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Enhancing Brain Tumor Classification Accuracy through Multi-Modal MRI Analysis and Advanced Ensemble Deep Learning

AU - Berra, Ayyoub

AU - Bouguerra, Oussama

AU - Attallah, Bilal

AU - Brik, Youcef

AU - Djerioui, Mohamed

AU - Shayea, Ibraheem

AU - Al-Zahrani, Saleh Ibrahim

AU - Bouzidi, Youssouf

AU - Boudjelal, Abdelwahhab

PY - 2025

Y1 - 2025

N2 - This paper focused on the multi-class brain tumor classification problem, using deep learning methodologies based on magnetic resonance imaging (MRI) images. Two state-of-the-art deep learning models, EfficientNet-B3 and ConvNeXt-Tiny, were fine-tuned through transfer learning with custom classifier heads, and used a publicly available dataset from Kaggle, leveraging various data augmentation techniques to help reduce class imbalance. These two models were trained to the same hyperparameters and had the same metrics applied to the evaluation. Both models were assessed using Accuracy, F1-Score, Recall, and Precision. EfficientNet-B3 achieved an accuracy of 96.85% and ConvNeXt-Tiny achieved 95.88%. After observing the strengths of both models, a weighted ensemble (60% EfficientNet-B3 and 40% ConvNeXt-Tiny) was created, achieving a final accuracy of 97.07%. Based on the results of the study, it demonstrates how deep learning and ensemble techniques can improve diagnostic accuracy in a complex multi-class brain tumor classification task.

AB - This paper focused on the multi-class brain tumor classification problem, using deep learning methodologies based on magnetic resonance imaging (MRI) images. Two state-of-the-art deep learning models, EfficientNet-B3 and ConvNeXt-Tiny, were fine-tuned through transfer learning with custom classifier heads, and used a publicly available dataset from Kaggle, leveraging various data augmentation techniques to help reduce class imbalance. These two models were trained to the same hyperparameters and had the same metrics applied to the evaluation. Both models were assessed using Accuracy, F1-Score, Recall, and Precision. EfficientNet-B3 achieved an accuracy of 96.85% and ConvNeXt-Tiny achieved 95.88%. After observing the strengths of both models, a weighted ensemble (60% EfficientNet-B3 and 40% ConvNeXt-Tiny) was created, achieving a final accuracy of 97.07%. Based on the results of the study, it demonstrates how deep learning and ensemble techniques can improve diagnostic accuracy in a complex multi-class brain tumor classification task.

KW - Brain Tumor Classification

KW - ConvNeXt-Tiny

KW - Deep Learning

KW - EfficientNet-B3

KW - Ensemble Learning

KW - Image Classification

KW - MRI

KW - Weight Averaging

UR - https://www.scopus.com/pages/publications/105035996661

U2 - 10.1109/ICATEEE68170.2025.11406230

DO - 10.1109/ICATEEE68170.2025.11406230

M3 - Conference contribution

AN - SCOPUS:105035996661

T3 - 2nd International Conference on Advanced Technology in Electronic and Electrical Engineering, ICATEEE 2025

BT - 2nd International Conference on Advanced Technology in Electronic and Electrical Engineering, ICATEEE 2025

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2nd International Conference on Advanced Technology in Electronic and Electrical Engineering, ICATEEE 2025

Y2 - 10 December 2025 through 11 December 2025

ER -

Berra A, Bouguerra O, Attallah B, Brik Y, Djerioui M, Shayea I et al. Enhancing Brain Tumor Classification Accuracy through Multi-Modal MRI Analysis and Advanced Ensemble Deep Learning. In 2nd International Conference on Advanced Technology in Electronic and Electrical Engineering, ICATEEE 2025. Institute of Electrical and Electronics Engineers Inc. 2025. (2nd International Conference on Advanced Technology in Electronic and Electrical Engineering, ICATEEE 2025). doi: 10.1109/ICATEEE68170.2025.11406230

Enhancing Brain Tumor Classification Accuracy through Multi-Modal MRI Analysis and Advanced Ensemble Deep Learning

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this