Automatic Lung Segmentation of Volumetric Low-Dose CT Scans Using Graph Cuts

Ali, Asem; Farag, Aly A.

doi:10.1007/978-3-540-89639-5_25

Cited by 33 publications

(37 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are a variety of methods for extracting lung volume from a pulmonary CT scan. In order to segment lung, approaches such as global thresholding [5,6], optimal thresholding [7,8], three-dimensional (3-D)-adaptive fuzzy thresholding [9], rule-based region growing [10], connected component labeling [11], graph-cut algorithm [12], and hybrid segmentation [13] have generally been used. After initial segmentation, lung volume has been extracted using several methods, and the extracted lung volume needed to be refined to include juxta-pleural nodules.…”

Section: Introductionmentioning

confidence: 99%

Automated Pulmonary Nodule Detection System in Computed Tomography Images: A Hierarchical Block Classification Approach

Choi

2013

Entropy

View full text Add to dashboard Cite

A computer-aided detection (CAD) system is helpful for radiologists to detect pulmonary nodules at an early stage. In this paper, we propose a novel pulmonary nodule detection method based on hierarchical block classification. The proposed CAD system consists of three steps. In the first step, input computed tomography images are split into three-dimensional block images, and we apply entropy analysis on the block images to select informative blocks. In the second step, the selected block images are segmented and adjusted for detecting nodule candidates. In the last step, we classify the nodule candidate images into nodules and non-nodules. We extract feature vectors of the objects in the selected blocks. Lastly, the support vector machine is applied to classify the extracted feature vectors. Performance of the proposed system is evaluated on the Lung Image Database Consortium database. The proposed method has reduced the false positives in the nodule candidates significantly. It achieved 95.28% sensitivity with only 2.27 false positives per scan.

show abstract

Section: Introductionmentioning

confidence: 99%

Automated Pulmonary Nodule Detection System in Computed Tomography Images: A Hierarchical Block Classification Approach

Choi

2013

Entropy

View full text Add to dashboard Cite

show abstract

“…This method is evaluated against two sophisticated approaches developed in the CVIP Lab based on the level sets [15] and random fields [16], and is shown to provide comparable performance yet is faster and requires minimal human intervention.…”

Section: Image Segmentationmentioning

confidence: 99%

“…The author of this thesis has been involved in developing a simplified statistical approach (e.g., Elhabian, Farag et al 2008 [14]) that exploits the major chrematistics of the lung CT scans as we studied in Chapter 2. This chapter will focus on this approach and will briefly discuss the sophisticated algorithms of Farag and Abdelmunim [15], 2004 and Ali and Farag, 2008 [16].…”

Section: E Summarymentioning

confidence: 99%

“…Graph Cuts-Based Approach [16] Ali and Farag [16] developed a statistical based segmentation algorithm for lung segmentation using composite random field modeling and employed the graph cuts approach to carry out the optimization of the resulting objective function which provides maximum a posterior (MAP) estimation of the desired segmentation.…”

Section: Algorithmmentioning

confidence: 99%

“…In summary, vis-à-vis CT image modeling we consider two aspects: a) correction of noise and haziness -this is accomplished by optimal filtering; and b) modeling the probability density function, that characterizes the gray scale intensity distribution, in order to serve the purpose of isolating the lung tissues from the thorax -this is accomplished using the linear model with focus on the special case of Gaussian kernels in which the Expectation-Maximization (EM) is the vehicle to carryout the estimation process (e.g., [14] [16]). In all the developments to follow we shall avoid immersing into the ocean of the scientific literature that deals with image modeling as this indeed would drift us into unchartered territories beyond the focus of this thesis.…”

Section: A Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Lung nodule modeling and detection for computerized image analysis of low dose CT imaging of the chest.

Farag¹

View full text Add to dashboard Cite

"Lung nodule modeling and detection for computerized image analysis of low dose CT imaging of the chest." (2009 The classification stage matches, according to a particular approach, the characteristics (e.g., shape, texture and spatial distribution) of the detected nodules to common characteristics (again, shape, texture and spatial distribution) of nodules with known pathologies (confirmed by biopsies). This thesis focuses on the first step; i.e., nodule detection.Specifically, the thesis addresses three issues: a) understanding the CT data of typical low dose CT (LDCT) scanning of the chest, and devising an image processing approach to reduce the inherent artifacts in the scans; b) devising an image segmentation approach to isolate the lung tissues from the rest of the chest and thoracic regions in the CT scans;and c) devising a nodule modeling methodology to enhance the detection rate and lend benefits for the ultimate step in computerized image analysis of LDCT of the lungs, namely associating a pathology to the detected nodule.The methodology for reducing the noise artifacts is based on noise analysis and examination of typical LDCT scans that may be gathered on a repetitive fashion; since, a reduction in the resolution is inevitable to avoid excessive radiation. Two optimal filtering methods are tested on samples of the ELCAP screening data; the Weiner and the vi Anisotropic Diffusion Filters. Preference is given to the Anisotropic Diffusion Filter, which can be implemented on 7x7 blocks/windows of the CT data.The methodology for lung segmentation is based on the inherent characteristics of the LDCT scans, shown as distinct bi-modal gray scale histogram. A linear model is used to describe the histogram (the joint probability density function of the lungs and nonlungs tissues) by a linear combination of weighted kernels. The Gaussian kernels were chosen, and the classic Expectation-Maximization (EM) algorithm was employed to estimate the marginal probability densities of the lungs and non-lungs tissues, and select an optimal segmentation threshold. The segmentation is further enhanced using standard shape analysis based on mathematical morphology, which improves the continuity of the outer and inner borders of the lung tissues. This approach (a preliminary version of it appeared in [14]) is found to be adequate for lung segmentation as compared to more sophisticated approaches developed at the CVIP Lab (e.g., [15][16]) and elsewhere.The methodology developed for nodule modeling is based on understanding the physical characteristics of the nodules in LDCT scans, as identified by human experts.An empirical model is introduced for the probability density of the image intensity (or Hounsfield units) versus the radial distance measured from the centroid -center of massof typical nodules. This probability density showed that the nodule spatial support is within a circle/square of size 10 pixels; i.e., limited to 5 mm in length; which is within the range that the radiologist specify to be of concern. This probabil...

show abstract

Superpixel with nanoscale imaging and boosted deep convolutional neural network concept for lung tumor classification

Rani

Jawhar

2020

Int J Imaging Syst Tech

View full text Add to dashboard Cite

Lung tumor is a complex illness caused by irregular lung cell growth. Earlier tumor detection is a key factor in effective treatment planning. When assessing the lung computed tomography, the doctor has many difficulties when determining the precise tumor boundaries. By offering the radiologist a second opinion and helping to improve the sensitivity and accuracy of tumor detection, the use of computer‐aided diagnosis could be near as effective. In this research article, the proposed Lung Tumor Detection Algorithm consists of four phases: image acquisition, preprocessing, segmentation, and classification. The Advance Target Map Superpixel‐based Region Segmentation Algorithm is proposed for segmentation purposes, and then the tumor region is measured using the nanoimaging theory. Using the concept of boosted deep convolutional neural network yields 97.3% precision, image recognition can be achieved. In the types of literature with the current method, which shows the study's proposed efficacy, the implementation of the proposed approach is found dramatically.

show abstract

Automatic Lung Segmentation of Volumetric Low-Dose CT Scans Using Graph Cuts

Cited by 33 publications

References 10 publications

Automated Pulmonary Nodule Detection System in Computed Tomography Images: A Hierarchical Block Classification Approach

Automated Pulmonary Nodule Detection System in Computed Tomography Images: A Hierarchical Block Classification Approach

Lung nodule modeling and detection for computerized image analysis of low dose CT imaging of the chest.

Superpixel with nanoscale imaging and boosted deep convolutional neural network concept for lung tumor classification

Contact Info

Product

Resources

About