DCGAN-Based Data Augmentation and GridSearch-Optimized CNN for Imbalanced Brain Tumor MRI Classification

DOI

10.2991/978-94-6239-713-2_12How to use a DOI?

Keywords

CNN; Hyperparameter optimization; GridSearchCV; Medical image classification; Brain tumor

Abstract

This paper focuses on more accurately and efficiently detecting the presence of brain tumors from their Magnetic Resonance Imaging (MRI) scans, leading to improved clinical decision-making and treatment outcomes. Medical image classification frequently uses deep learning models that must deal with complications like class imbalance and inappropriate hyperparameter settings, which can harm their performance and accuracy. To overcome these problems, this paper proposes a unified framework of DCGAN–based data augmentation, along with GridSearch-based hyperparameter optimisation for multiclass brain tumor classification. The DCGAN model generates synthetic MRI samples to balance the dataset and increase feature variance, whereas GridSearchCV optimises hyperparameters for the CNN classifier. The experiment is conducted on deep learning models, Custom CNN, VGG19, EfficientNet and BPNN, with different optimisers. Out of these models, the Custom CNN with Adam optimiser attained the best performance, registering an accuracy of 88.79%, recall of 87.96, and an F1-score of 88.13. The results show that using generative data augmentation effectively enhances classification reliability, while hyperparameter optimisation enables automated brain Tumor diagnosis with consistently better performance.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Pradeep Shree Adhikari
AU  - Amit Sharma
PY  - 2026
DA  - 2026/06/25
TI  - DCGAN-Based Data Augmentation and GridSearch-Optimized CNN for Imbalanced Brain Tumor MRI Classification
BT  - Proceedings of the International Conference on Advances in Computing Technology and Artificial Intelligence (COMPUTATIA 2026)
PB  - Atlantis Press
SP  - 159
EP  - 175
SN  - 2589-4919
UR  - https://doi.org/10.2991/978-94-6239-713-2_12
DO  - 10.2991/978-94-6239-713-2_12
ID  - Adhikari2026
ER  -