Shape-Based Computer-Aided Detection of Lung Nodules in Thoracic CT Images

Ye, Xujiong; Lin, Xinyu; Dehmeshki, Jamshid; Slabaugh, Greg; Beddoe, Gareth

doi:10.1109/tbme.2009.2017027

Cited by 249 publications

(53 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although CT imaging provides excellent spatial and contrast resolution, a technique to quantify surface morphology has not been previously available, in part due to the complexity of the task compared with volume estimation. Curvature-based descriptors can be used to quantify surface shape (14) and texture, as previously described in colonic polyp detection at CT colonography (15, 16) and lung nodule detection in thoracic CT (17, 18). However, these studies focused on the overall geometric shape of the object surface, rather than its fine details.…”

Section: Discussionmentioning

confidence: 99%

Automated Assessment of Renal Cortical Surface Roughness From Computerized Tomography Images and Its Association with Age

et al. 2014

View full text Add to dashboard Cite

Rationale and Objectives Nephrosclerosis occurs with aging and is characterized by increased kidney sub-capsular surface irregularities at autopsy. Assessments of cortical roughness in-vivo could provide an important measure of nephrosclerosis. The purpose of this study was to develop and validate an image-processing algorithm for quantifying renal cortical surface roughness in-vivo and determine its association with age. Materials and methods Renal cortical surface roughness was measured on contrast-enhanced abdominal CT images of potential living kidney donors. A roughness index was calculated based on geometric curvature of each kidney from 3D images, and compared with visual observation scores. Cortical roughness was compared between the oldest and youngest donors, and its interaction with cortical volume and age assessed. Results The developed quantitative roughness index identified significant differences in kidneys with visual surface roughness scores of 0 (minimal), 1 (mild), and 2 (moderate) (p<0.001) in a random sample of 200 potential kidney donors. Cortical roughness was significantly higher in the 94 oldest (64–75y) versus 91 youngest (18–25y) potential kidney donors (p<0.001). Lower cortical volume was associated with older age but not with roughness (r=−0.03, p=0.75). The association of oldest age group with roughness (OR=1.8 per SD of roughness index) remained significant after adjustment for total cortex volume (OR=2.0 per SD of roughness index). Conclusion A new algorithm to measure renal cortical surface roughness from CT scans detected rougher surface in older compared to younger kidneys, independent of cortical volume loss. This novel index may allow quantitative evaluation of nephrosclerosis in vivo using contrast-enhanced CT.

show abstract

Section: Discussionmentioning

confidence: 99%

Automated Assessment of Renal Cortical Surface Roughness From Computerized Tomography Images and Its Association with Age

et al. 2014

View full text Add to dashboard Cite

show abstract

“…The pulmonary nodule recognition involves nodule candidate detection [12] and false-positive reduction [13]. The traditional approaches of false-positive reduction have successive steps: feature extraction [14, 15] and classifier model construction [10, 16]. The most effective features which can be used for classification for lung CT images are, for example, shape, intensity, texture, geometric, gradient, and wavelet.…”

Section: Related Workmentioning

confidence: 99%

Pulmonary Nodule Classification with Deep Convolutional Neural Networks on Computed Tomography Images

Cao

Zhao

et al. 2016

Computational and Mathematical Methods in Medicine

154

View full text Add to dashboard Cite

Computer aided detection (CAD) systems can assist radiologists by offering a second opinion on early diagnosis of lung cancer. Classification and feature representation play critical roles in false-positive reduction (FPR) in lung nodule CAD. We design a deep convolutional neural networks method for nodule classification, which has an advantage of autolearning representation and strong generalization ability. A specified network structure for nodule images is proposed to solve the recognition of three types of nodules, that is, solid, semisolid, and ground glass opacity (GGO). Deep convolutional neural networks are trained by 62,492 regions-of-interest (ROIs) samples including 40,772 nodules and 21,720 nonnodules from the Lung Image Database Consortium (LIDC) database. Experimental results demonstrate the effectiveness of the proposed method in terms of sensitivity and overall accuracy and that it consistently outperforms the competing methods.

show abstract

“…Xujiong Ye et al [7] presented a CAD method for detecting both solid nodules and ground-glass opacity (GGO) nodules in the lung image. A fuzzy based thresholding approach is used to segment the lung region.…”

Section: Related Workmentioning

confidence: 99%

FEMD Algorithm for Effective Segmentation of CT Lung Images

FaizalKhan¹,

Qathiri²

2015

IJCA

View full text Add to dashboard Cite

Medical Image segmentation is the most important step in extracting information from medical images. Segmentation of pulmonary Chest Computed Tomography (CT) images is a precursor to most of the pulmonary image analysis schemes. The purpose of lung image segmentation is to separate the voxels corresponding to lung tissue from the anatomy of the surrounding. In this paper, an automated image segmentation method has been proposed inorder to segment the region of interest present in the CT Lung slices. The proposed approach utilizes Fuzzy logic with Earth Mover's Distance (FEMD) based refinement methods. The final segmented output is further refined by morphological based operators. The performance of the proposed method is compared with various segmentation methods such as Canny Sobel and Prewitt and we have obtained an average segmentation accuracy of 79.4091% for segmenting CT lung images.

show abstract

Shape-Based Computer-Aided Detection of Lung Nodules in Thoracic CT Images

Cited by 249 publications

References 32 publications

Automated Assessment of Renal Cortical Surface Roughness From Computerized Tomography Images and Its Association with Age

Automated Assessment of Renal Cortical Surface Roughness From Computerized Tomography Images and Its Association with Age

Pulmonary Nodule Classification with Deep Convolutional Neural Networks on Computed Tomography Images

FEMD Algorithm for Effective Segmentation of CT Lung Images

Contact Info

Product

Resources

About