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

• Autor(a) principal: Lopes, Thales R. S. [UNESP]
• Data de Publicação: 2024
• Outros Autores: Roberto, Guilherme F., Soares, Carlos, Tosta, Thaína A. A., Silva, Adriano B., Loyola, Adriano M., Cardoso, Sérgio V., de Faria, Paulo R., do Nascimento, Marcelo Z., Neves, Leandro A. [UNESP]
• Tipo de documento: Artigo de conferência
• Idioma: eng
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://dx.doi.org/10.5220/0012358200003660; https://hdl.handle.net/11449/309503
• topic: Deep Learning; Explainable Artificial Intelligence; Fractal Features; Histological Images

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.