Exploiting Test-Time Augmentation in Federated Learning for Brain Tumor MRI Classification

TL;DR

This paper demonstrates that applying test-time augmentation significantly improves federated learning-based MRI brain tumor classification accuracy, recommending TTA as a default inference strategy in medical imaging.

Contribution

It shows that combining test-time augmentation with federated learning enhances MRI classification performance, with minimal preprocessing needed.

Findings

TTA improves federated MRI classification accuracy significantly.

Preprocessing alone yields negligible gains.

TTA should be the default inference strategy in FL medical imaging.

Abstract

Efficient brain tumor diagnosis is crucial for early treatment; however, it is challenging because of lesion variability and image complexity. We evaluated convolutional neural networks (CNNs) in a federated learning (FL) setting, comparing models trained on original versus preprocessed MRI images (resizing, grayscale conversion, normalization, filtering, and histogram equalization). Preprocessing alone yielded negligible gains; combined with test-time augmentation (TTA), it delivered consistent, statistically significant improvements in federated MRI classification (p<0.001). In practice, TTA should be the default inference strategy in FL-based medical imaging; when the computational budget permits, pairing TTA with light preprocessing provides additional reliable gains.