Glaucomaé uma doença que danifica o nervoóptico. Elaé considerada a segunda principal causa de cegueira no mundo. Vários sistemas de diagnóstico automático têm sido propostos. No entanto, esses sistemas não foram capazes de lidar com uma grande diversidade de imagens. Portanto, tais métodos não são viáveis para uso em programas de triagem. Realizamos um extenso estudo para definir o melhor conjunto de atributos para a representação da imagem. No total, avaliamos 16.469 características. Nossa abordagem de detecção de glaucoma usa descritores de textura e Redes Neurais Convolucionais (CNNs). Avaliamos nossa proposta em um total de 873 imagens de quatro bancos de dados públicos e concluímos que a junção da GLCM e CNNs prétreinadas juntamente com o uso do classificador Random Forest são promissores na detecção desta patologia, obtendo uma acurácia de 93,35% e umíndice Kappa considerado Excelente.
A utilização de técnicas de processamento digital de imagens (PDI) é destaque no cenário médico para o diagnóstico automático de patologias. Na área oftalmológica o glaucoma é a segunda principal causa da perda de visão no mundo e não possui cura. Atualmente, existem tratamentos para prevenir a perda da visão, contudo a doença deve ser descoberta nos estágios iniciais. O objetivo principal deste artigo é revisar as metodologias e técnicas de segmentação dos limites do disco óptico e escavação. Essas regiões são utilizadas para o cálculo de métricas para classi cação do glaucoma e auxílio aos pro ssionais da área. Os trabalhos mais recentes publicados na área foram classi cados em cinco grupos de acordo com a principal técnica de PDI aplicada: agrupamento, superpixel, contorno ativo, morfologia matemática e redes neurais convolucionais. Além disso, foi realizado um levantamento das principais bases de imagens e métricas de avaliação utilizadas.Palavras-Chave: Algoritmos de Agrupamento; Contorno Ativo; Glaucoma; Morfologia matemática; Redes neurais convolucionais; Superpixel. Abstract The use of digital image processing techniques (DIP) is highlighted in the medical scenario for automatic diagnosis of pathologies. In the ophthalmologic area, glaucoma is the second leading cause of vision loss in the world
Leukemia is a disease that has no defined etiology and affects the production of white blood cells in the bone marrow. Young cells or blasts are produced abnormally, replacing normal blood cells (white, red blood cells, and platelets). Consequently, the person suffers problems in transporting oxygen and infections combat. Acute leukemia is a particular type of leukemia that causes abnormal cell growth in a short period, requiring a quick start of treatment. Classifying the types of acute leukemia in blood slide images is a vital process, and a system of assisting doctors in selecting treatment becomes necessary. This paper presents an ensemble approach using four convolutional neural networks (CNNs) -Alert Net-RWD, Resnet50, InceptionV3, and Xception. These CNNs, individually, demonstrated that are efficient in differentiating between the two types of acute leukemia -Acute Lymphoid Leukemia (ALL) and Acute Myeloid Leukemia (AML) -and Healthy Blood Slides (HBS). We verified that the union of these four well-known CNNs improve the hit rates of current techniques from the literature. The experiments were carried out using 18 data sets with 3,293 images, and the proposed CNNs ensemble achieved an accuracy of 96.17%, and precision of 96.38%.
Leukemia is a disease that has no defined etiology and affects the production of white blood cells in the bone marrow. Young cells or blasts are produced abnormally, replacing normal blood cells (white, red blood cells, and platelets). Consequently, the person suffers problems in transporting oxygen and infections combat. Acute leukemia is a particular type of leukemia that causes abnormal cell growth in a short period, requiring a quick start of treatment. Classifying the types of acute leukemia in blood slide images is a vital process, and a system of assisting doctors in selecting treatment becomes necessary. This paper presents an ensemble approach using four convolutional neural networks (CNNs) -Alert Net-RWD, Resnet50, InceptionV3, and Xception. These CNNs, individually, demonstrated that are efficient in differentiating between the two types of acute leukemia -Acute Lymphoid Leukemia (ALL) and Acute Myeloid Leukemia (AML) -and Healthy Blood Slides (HBS). We verified that the union of these four well-known CNNs improve the hit rates of current techniques from the literature. The experiments were carried out using 18 data sets with 3,293 images, and the proposed CNNs ensemble achieved an accuracy of 96.17%, and precision of 96.38%.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.