ACADEMIC WORK

Published paper · Deep Learning · Medical Imaging

MRI-based Brain Tumor Classification using Transfer Learning: A Comparative Analysis

Anil Poudyal, Niruta Devkota, Gayatri Budha

London Metropolitan University
International Journal of Computer Applications, Vol. 186, No. 40, pp. 7–11 · September 2024

Summary

Fast, accurate brain tumour diagnosis shapes treatment and outcomes, but reading MRI scans by eye varies between clinicians. We compared five ImageNet pre-trained CNNs (VGG16, ResNet50, DenseNet121, InceptionV3 and InceptionResNetV2) on 7,023 brain MRI images across four classes: glioma, meningioma, pituitary tumour and no tumour. Under identical training settings, DenseNet121 performed best with 94% validation accuracy and precision, recall and F1-score of 0.94.

Keywords: transfer learning, CNN, brain tumour classification, MRI, DenseNet121, medical image analysis

7,023MRI images, 4 classes
5Pre-trained models compared
94%Best validation accuracy (DenseNet121)
0.27Lowest validation loss

Approach

  • Data: the Kaggle Brain Tumor MRI Dataset (5,712 training, 1,311 testing images), roughly balanced at 1,300 to 1,500 images per class.
  • Preprocessing: images resized to 224 × 224 and augmented with rotation, shearing, zoom and horizontal flips.
  • Model setup: early layers of each pre-trained network frozen; batch normalisation added after the base model, then a 128-neuron dense layer with L1/L2 regularisation and a softmax output.
  • Training: Adam optimiser, categorical cross-entropy, up to 50 epochs with early stopping, on a single GTX 1660 Ti GPU.

Results

0 0.25 0.50 0.75 1.00 DenseNet121 0.94 InceptionV3 0.91 InceptionResNetV2 0.90 VGG16 0.88 ResNet50 0.51
Figure 1. Validation accuracy of each pre-trained model under the same training configuration.

Table 1. Validation results for all five models.

ModelAccuracyLossPrecisionRecallF1-score
DenseNet1210.940.270.940.940.94
InceptionV30.910.330.910.910.91
InceptionResNetV20.900.340.900.900.90
VGG160.880.410.880.880.88
ResNet500.511.120.510.510.49

Key findings

  • DenseNet121, the deepest yet one of the smallest models (8.1M parameters), gave the best accuracy and lowest loss.
  • The Inception models and VGG16 performed close behind, while ResNet50 did not adapt well in this configuration.
  • Increasing the dense layer from 64 to 128 neurons noticeably reduced validation loss; going beyond 128 gave no further gain.

Future directions

Larger and more diverse MRI datasets, multi-modal inputs such as CT and PET, semi-supervised learning to reduce reliance on labels, lightweight models for low-resource clinics, and Grad-CAM explanations to make predictions easier for clinicians to trust.

Cite this paper

A. Poudyal, N. Devkota and G. Budha, “MRI-based Brain Tumor Classification using Transfer Learning: A Comparative Analysis,” International Journal of Computer Applications, vol. 186, no. 40, pp. 7–11, Sept. 2024. doi: 10.5120/ijca2024923988

Summary of the published paper. All results are from the original article.