Publicação: A Hermite polynomial algorithm for detection of lesions in lymphoma images
Carregando...
Data
Orientador
Coorientador
Pós-graduação
Curso de graduação
Título da Revista
ISSN da Revista
Título de Volume
Editor
Tipo
Artigo
Direito de acesso
Resumo
There are different types of lesions that can be investigated with the hematoxylin–eosin staining protocol. Lymphoma is a type of malignant disease which affects one of the highest white blood cell populations responsible for the immunological defence system. There are lymphoma sub-types that can have similar features, which make their diagnoses a difficult task. In this study, we investigated algorithms based on multiscale and multidimensional fractal geometry with colour models for classification of lymphoma images. Fractal features were extracted from the colour models and separate channels from these models. These features were concatenated to form feature vectors. Finally, we investigated the Hermite polynomial classifier and machine learning algorithms in order to evaluate the performance of the proposed approach. We employed the tenfold cross-validation method and evaluated the lesion sub-types with the binary and multiclass classifications. The separated colour channels obtained from histological images achieved relevant values for the binary and multiclass classifications, with an accuracy rating between 91 and 97%. These results can contribute to the detection and classification of the lesions by supporting specialists in clinical practices.
Descrição
Palavras-chave
Classification, Colour fractal, Colour spaces, Hermite polynomial, Lymphoma
Idioma
Inglês
Como citar
Pattern Analysis and Applications, v. 24, n. 2, p. 523-535, 2021.
