Explainability of Deep Neural Networks for Brain Tumor Detection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Park, S., Kim, J.
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912067272835072
author Park, S.
Kim, J.
author_facet Park, S.
Kim, J.
contents Medical image classification is crucial for supporting healthcare professionals in decision-making and training. While Convolutional Neural Networks (CNNs) have traditionally dominated this field, Transformer-based models are gaining attention. In this study, we apply explainable AI (XAI) techniques to assess the performance of various models on real-world medical data and identify areas for improvement. We compare CNN models such as VGG-16, ResNet-50, and EfficientNetV2L with a Transformer model: ViT-Base-16. Our results show that data augmentation has little impact, but hyperparameter tuning and advanced modeling improve performance. CNNs, particularly VGG-16 and ResNet-50, outperform ViT-Base-16 and EfficientNetV2L, likely due to underfitting from limited data. XAI methods like LIME and SHAP further reveal that better-performing models visualize tumors more effectively. These findings suggest that CNNs with shallower architectures are more effective for small datasets and can support medical decision-making.
format Preprint
id arxiv_https___arxiv_org_abs_2410_07613
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Explainability of Deep Neural Networks for Brain Tumor Detection
Park, S.
Kim, J.
Computer Vision and Pattern Recognition
Medical image classification is crucial for supporting healthcare professionals in decision-making and training. While Convolutional Neural Networks (CNNs) have traditionally dominated this field, Transformer-based models are gaining attention. In this study, we apply explainable AI (XAI) techniques to assess the performance of various models on real-world medical data and identify areas for improvement. We compare CNN models such as VGG-16, ResNet-50, and EfficientNetV2L with a Transformer model: ViT-Base-16. Our results show that data augmentation has little impact, but hyperparameter tuning and advanced modeling improve performance. CNNs, particularly VGG-16 and ResNet-50, outperform ViT-Base-16 and EfficientNetV2L, likely due to underfitting from limited data. XAI methods like LIME and SHAP further reveal that better-performing models visualize tumors more effectively. These findings suggest that CNNs with shallower architectures are more effective for small datasets and can support medical decision-making.
title Explainability of Deep Neural Networks for Brain Tumor Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2410.07613