Segmentation of abnormal cells by using level set model

Haj-Hassan, Hawraa; Chaddad, Ahmad; Tanougast, Camel; Harkouss, Youssef

doi:10.1109/codit.2014.6996994

Cited by 2 publications

(2 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To evaluate the performance of our segmentation model, we used manually annotated ground truth provided by the CHU Nancy-Brabois Hospital. The Jaccard similarity coefficient (JSC), dice similarity coefficient (DSC),[ 22 23 ] false positive rate (FPR),[ 24 ] and false negative rate (FNR)[ 25 ] were considered as performance metrics. The JSC and DSC metrics evaluate the degree of the correspondence between two segmentations (i.e., segmentation output and ground truth) and are defined as:…”

Section: Ethodsmentioning

confidence: 99%

Classifications of Multispectral Colorectal Cancer Tissues Using Convolution Neural Network

Haj-Hassan

Chaddad

Harkouss

et al. 2017

Journal of Pathology Informatics

Self Cite

View full text Add to dashboard Cite

Background:Colorectal cancer (CRC) is the third most common cancer among men and women. Its diagnosis in early stages, typically done through the analysis of colon biopsy images, can greatly improve the chances of a successful treatment. This paper proposes to use convolution neural networks (CNNs) to predict three tissue types related to the progression of CRC: benign hyperplasia (BH), intraepithelial neoplasia (IN), and carcinoma (Ca).Methods:Multispectral biopsy images of thirty CRC patients were retrospectively analyzed. Images of tissue samples were divided into three groups, based on their type (10 BH, 10 IN, and 10 Ca). An active contour model was used to segment image regions containing pathological tissues. Tissue samples were classified using a CNN containing convolution, max-pooling, and fully-connected layers. Available tissue samples were split into a training set, for learning the CNN parameters, and test set, for evaluating its performance.Results:An accuracy of 99.17% was obtained from segmented image regions, outperforming existing approaches based on traditional feature extraction, and classification techniques.Conclusions:Experimental results demonstrate the effectiveness of CNN for the classification of CRC tissue types, in particular when using presegmented regions of interest.

show abstract

Section: Ethodsmentioning

confidence: 99%

Classifications of Multispectral Colorectal Cancer Tissues Using Convolution Neural Network

Haj-Hassan

Chaddad

Harkouss

et al. 2017

Journal of Pathology Informatics

Self Cite

View full text Add to dashboard Cite

show abstract

“…2014, Ahmad Chaddad et.al presented the level set method [8] used in medical image segmentation. It is employed and implemented using real data of carcinoma cancer cells.…”

Section: Literature Reviewmentioning

confidence: 99%

Untitled

2018

IJET

View full text Add to dashboard Cite

Segmentation means dividing an image into connected regions. It is the basic procedure to enhance the image. This process accommodate to identify and envisage objects in an image. In medical imaging, this algorithm is useful to elevate organ, cells or structures in the body. But there are many difficulties occurred in the segmentation process, they may be noise, variation of contrast, motion blurring artifacts. There are many segmentation methods have been proposed from the past years, but we used in this paper the segmentation algorithm based on level-set. The application in this work is to assess the performance of the six level-set algorithms on a given MRI image of Brain segmentation in 2D. The initialization implemented in this process is common for all the algorithms and the reference contour chosen for the computation of Dice criterion. MATLAB tool based application is used to evaluate the performance of various level-set algorithms in the segmentation of an image particularly on medical images. Classical methods such as Chan & Vese and Shi algorithms only evolve on their narrow band and are region based. We propose the recent methods such as Li and Lankton algorithms which are localized region based. Analyse and compare various level set algorithms on medical images in terms of Dice criterion, computational time , PSNR, Hausdorff distance and the mean sum of squared distance(MSSD).

show abstract

Segmentation of abnormal cells by using level set model

Cited by 2 publications

References 26 publications

Classifications of Multispectral Colorectal Cancer Tissues Using Convolution Neural Network

Classifications of Multispectral Colorectal Cancer Tissues Using Convolution Neural Network

Untitled

Contact Info

Product

Resources

About