Association of Grad-CAM, LIME and Multidimensional Fractal Techniques for the Classification of H&E Images

Thales Lopes, Guilherme Roberto, Carlos Soares, Thaína Tosta, Adriano Silva, Adriano Loyola, Sérgio Cardoso, Paulo R. de Faria, Marcelo Z. do Nascimento, Leandro Neves

2024

Abstract

In this work, a method based on the use of explainable artificial intelligence techniques with multiscale and multidimensional fractal techniques is presented in order to investigate histological images stained with Hematoxylin-Eosin. The CNN GoogLeNet neural activation patterns were explored, obtained from the gradient-weighted class activation mapping and locally-interpretable model-agnostic explanation techniques. The feature vectors were generated with multiscale and multidimensional fractal techniques, specifically fractal dimension, lacunarity and percolation. The features were evaluated by ranking each entry, using the ReliefF algorithm. The discriminative power of each solution was defined via classifiers with different heuristics. The best results were obtained from LIME, with a significant increase in accuracy and AUC rates when compared to those provided by GoogLeNet. The details presented here can contribute to the development of models aimed at the classification of histological images.

Paper Citation

in Harvard Style

Lopes T., Roberto G., Soares C., Tosta T., Silva A., Loyola A., Cardoso S., R. de Faria P., Z. do Nascimento M. and Neves L. (2024). Association of Grad-CAM, LIME and Multidimensional Fractal Techniques for the Classification of H&E Images. In Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP; ISBN 978-989-758-679-8, SciTePress, pages 441-447. DOI: 10.5220/0012358200003660

in Bibtex Style

@conference{visapp24,
author={Thales Lopes and Guilherme Roberto and Carlos Soares and Thaína Tosta and Adriano Silva and Adriano Loyola and Sérgio Cardoso and Paulo R. de Faria and Marcelo Z. do Nascimento and Leandro Neves},
title={Association of Grad-CAM, LIME and Multidimensional Fractal Techniques for the Classification of H&E Images},
booktitle={Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP},
year={2024},
pages={441-447},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012358200003660},
isbn={978-989-758-679-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP
TI - Association of Grad-CAM, LIME and Multidimensional Fractal Techniques for the Classification of H&E Images
SN - 978-989-758-679-8
AU - Lopes T.
AU - Roberto G.
AU - Soares C.
AU - Tosta T.
AU - Silva A.
AU - Loyola A.
AU - Cardoso S.
AU - R. de Faria P.
AU - Z. do Nascimento M.
AU - Neves L.
PY - 2024
SP - 441
EP - 447
DO - 10.5220/0012358200003660
PB - SciTePress