Improving the non-extensive medical image segmentation based on Tsallis entropy

Rodrigues, Paulo S.; Giraldi, Gilson A.

doi:10.1007/s10044-011-0225-y

Cited by 25 publications

(15 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…On the other hand, we get interesting enhancement effects of values nearby β by using another sigmoid-like function defined as Following the idea of non-extensive systems for natural as well as medical images [22,[27][28][29], in this work, we propose an extended version of Equations (9) and (10), called here as q-sigmoid functions, which are defined based on the q-exponential function, given by Equation 7in the forms bellow.…”

Section: Proposed Q-sigmoid Functionsmentioning

confidence: 99%

A q-Extension of Sigmoid Functions and the Application for Enhancement of Ultrasound Images

Rodrigues

Wachs-Lopes

Santos

et al. 2019

Entropy

View full text Add to dashboard Cite

This paper proposes the q-sigmoid functions, which are variations of the sigmoid expressions and an analysis of their application to the process of enhancing regions of interest in digital images. These new functions are based on the non-extensive Tsallis statistics, arising in the field of statistical mechanics through the use of q-exponential functions. The potential of q-sigmoids for image processing is demonstrated in tasks of region enhancement in ultrasound images which are highly affected by speckle noise. Before demonstrating the results in real images, we study the asymptotic behavior of these functions and the effect of the obtained expressions when processing synthetic images. In both experiments, the q-sigmoids overcame the original sigmoid functions, as well as two other well-known methods for the enhancement of regions of interest: slicing and histogram equalization. These results show that q-sigmoids can be used as a preprocessing step in pipelines including segmentation as demonstrated for the Otsu algorithm and deep learning approaches for further feature extractions and analyses.

show abstract

Section: Proposed Q-sigmoid Functionsmentioning

confidence: 99%

A q-Extension of Sigmoid Functions and the Application for Enhancement of Ultrasound Images

Rodrigues

Wachs-Lopes

Santos

et al. 2019

Entropy

View full text Add to dashboard Cite

show abstract

“…A ideia por trás da utilização da função generalizada da sigmoideé o parâmetro q, que permite adaptar a topologia do filtro para cada classe de aplicações, como em [Rodrigues and Giraldi 2011].…”

Section: Metodologiaunclassified

Estudo da ativação de regiões encefálicas em imagens fMRI

Santos

Rodrigues

2019

Anais Estendidos Do Simpósio Brasileiro De Computação Aplicada À Saúde (SBCAS 2019)

View full text Add to dashboard Cite

Resumo. Atualmente, 93 % das regiões subcorticais são desconhecidas dos neurocientistas e, portanto, classificadas como terra incognita. Apesar dos recentes avanços nas tecnologias de captura de imagens médicas, a qualidade das imagens coletadas in vivo aindaé baixa em relaçãoàs imagens histológicas. Por outro lado, aárea de estatística não-extensiva tem avançado e apresentados bons resultados sobretudo em processamento de imagens. Esse trabalho propõe a utilização do filtro sigmoidal baseado em entropia não-extensiva em imagens de ressonância magnética funcional (fMRI), com o intuito de realçar e mapear regiões encefálicas ativadas durante a execução de tarefas específicas.Abstract. Nowadays, 93 % of subcortical regions are unknown to neuroscientists and, therefore, are classified as terra incognita. Despite the recent advances in medical imaging technologies, the quality of images collected in vivo is still low when compared to the histological images. On the other hand, the area of non-extensive statistics has advanced and presented good results, mainly in image processing. This work proposes the use of the sigmoid fitler with nonextensive entropy in functional magnetic resonance imaging (fMRI), with the purpose of highlighting and mapping brain regions activated during the execution of specific tasks. Introdução e Trabalhos relacionadosApesar dos recentes avanços na neurociência, sobretudo no entendimento das regiões encefálicas e suas correlações, ainda restam diversos desafios pendentes. As interações entre regiões corticais e subcorticais, por exemplo, ainda não são completamente compreendidas pelos cientistas [Wager et al. 2008]. Portanto, desconhece-se a origem e motivos de progressão de doenças neurológicas que acometem milhões de indivíduos anualmente.Até 1940, por exemplo, acreditava-se que surtos de convulsão epiléptica ocorriam por disrupções nas interações convencionais apenas entre regiões corticais. Contudo, atualmente, evidências sugerem a participação de regiões subcorticais em surtos de epilepsia [Norden and Blumenfeld 2002]. Um exemplo de regiãoé o Thalamus, região subcortical responsável, dentre outras coisas, pela regulação do sono. Também, o Thalamusé uma das regiões responsáveis por descargas de espícula ou onda que causam distúrbios de epilepsia durante o sono [Norden and Blumenfeld 2002]. Portanto, o estudo de regiões subcorticais por meio de imagens de ressonância magnética pode auxiliar no diagnóstico preventivo e tratamento de doenças neurológicas.Entretanto, imagens coletadas in vivo têm qualidade inferioràs imagens coletadas em estado post-mortem. Apesar dos recentes avanços em técnicas de imageamento

show abstract

“…13 In recent development of nonextensive entropy, also called Tsallis entropy, some researchers proposed several Tsallis entropy thresholding methods. [14][15][16][17][18][19][20] Although such an entropy thresholding seems promising, it also suffers from one drawback. It does not take into account the image spatial correlation.…”

Section: Introductionmentioning

confidence: 99%

Tsallis cross-entropy based framework for image segmentation with histogram thresholding

Nie

2015

J. Electron. Imaging

View full text Add to dashboard Cite

Various techniques have previously been proposed for thresholding of images to separate objects from the background. Although these thresholding techniques have been proven effective on particular types of images, none of them is able to produce consistently good results on a wide range of existing images. The nonextensive cross-entropy (also known as Tsallis cross-entropy) is introduced to determine the optimal threshold value. The new thresholding scheme aims to minimize the Tsallis cross-entropy between the original image and the thresholded image. The effectiveness of the proposed scheme is demonstrated by using examples from the synthetic images, natural scene images, and an image dataset that includes nondestructive testing images and document images, on the basis of comparison with the traditional cross-entropy, Otsu's, minimum error thresholding, and two state-of-the-art methods. Furthermore, a tunable parameter q of Tsallis cross-entropy in the presented scheme gives the proposed methods the potential to handle the different image segmentation tasks.

show abstract

Improving the non-extensive medical image segmentation based on Tsallis entropy

Cited by 25 publications

References 16 publications

A q-Extension of Sigmoid Functions and the Application for Enhancement of Ultrasound Images

A q-Extension of Sigmoid Functions and the Application for Enhancement of Ultrasound Images

Estudo da ativação de regiões encefálicas em imagens fMRI

Tsallis cross-entropy based framework for image segmentation with histogram thresholding

Contact Info

Product

Resources

About