Models & Optimisation and Mathematical Analysis Journal
Volume 12, Numéro 1, Pages 16-21
2025-12-31
Improving Brain Tumor Classification In Mri Through Hybrid Data Preprocessing And Random Forests
Authors : Djafri Laouni . Gafour Yacine . Hoceini Mohammed . Horri Moura . Silarbi Ibrahim .
Abstract
Brain tumor classification using Magnetic Resonance Imaging (MRI) plays a crucial role in medical diagnosis and treatment planning. However, the automatic analysis of MRI data remains challenging due to noise, variability, and class imbalance. In this work, two approaches for binary brain tumor classification are investigated. The first approach is based on classical machine learning using a Random Forest classifier combined with a hybrid data preprocessing pipeline that integrates oversampling and data cleaning techniques. The second approach relies on a deep learning model based on a Convolutional Neural Network (CNN). Both approaches are evaluated under identical experimental conditions. The experimental results demonstrate that the machine learning approach combined with effective data preprocessing achieves superior performance compared to the CNN model, particularly in terms of recall and F1-score, which are critical metrics for medical diagnosis.
Keywords
Brain tumor classification, Random forest, Smote, Data analyses
Les articles similaires
Bouzid-daho Abdellatif
.
pages 1-7.
Azrou Nacima
.
pages 79-99.
Gherabi Abdesselem
.
pages 572-590.
Boucif Beddad
.
Kaddour Hachemi
.
Jack-gerard Postaire
.
pages 30-35.
Lamiche Chaabane
.
Moussaoui Abdelouahab
.
pages 37-41.
