COMBINING GRAPH-CUT TECHNIQUE AND ANATOMICAL KNOWLEDGE FOR AUTOMATIC SEGMENTATION OF LUNGS AFFECTED BY DIFFUSE PARENCHYMAL DISEASE IN HRCT IMAGES

2011 ◽  
Vol 11 (04) ◽  
pp. 509-529 ◽  
Author(s):  
LAURENT MASSOPTIER ◽  
AVISHKAR MISRA ◽  
ARCOT SOWMYA ◽  
SERGIO CASCIARO
Keyword(s):  
Lung Disease ◽  
Automatic Segmentation ◽  
Lung Parenchyma ◽  
Graph Cut ◽  
Dice Similarity Coefficient ◽  
High Resolution Computed Tomography ◽  
Segmentation Evaluation ◽  
Lung Segmentation ◽  
Disease Patterns ◽  
Parenchymal Disease

Accurate and automated lung segmentation in high-resolution computed tomography (HRCT) is highly challenged by the presence of pathologies affecting lung parenchyma appearance and borders. The algorithm presented employs an anatomical model-driven approach and systematic incremental knowledge acquisition to produce coarse lung delineation, used as initialization for the graph-cut algorithm. The proposed method is evaluated on a 49 HRCT cases dataset including various lung disease patterns. The accuracy of the method is assessed using dice similarity coefficient (DSC) and shape differentiation metrics (d mean , d rms ), by comparing the outputs of automatic lung segmentations and manual ones. The proposed automatic method demonstrates high segmentation accuracy ( DSC = 96.64%, d mean = 1.75 mm, d rms = 3.27 mm) with low variation that depends on the lung disease pattern. It also presents good improvement over the initial lung segmentation (Δ DSC = 4.74%, Δd mean = -3.67 mm, Δd rms = -6.25 mm), including impressive amelioration (maximum values of Δ DSC = 58.22% and Δd mean = -78.66 mm) when the anatomy-driven algorithm reaches its limit. Segmentation evaluation shows that the method can accurately segment lungs even in the presence of disease patterns, with some limitations in the apices and bases of lungs. Therefore, the developed automatic segmentation method is a good candidate for the first stage of a computer-aided diagnosis system for diffuse lung diseases.

scholarly journals Automatic Lung Segmentation Based on Texture and Deep Features of HRCT Images with Interstitial Lung Disease

2019 ◽  
Vol 2019 ◽  
pp. 1-8 ◽  
Author(s):  
Ting Pang ◽  
Shaoyong Guo ◽  
Xinwang Zhang ◽  
Lijie Zhao
Keyword(s):  
Interstitial Lung Disease ◽  
Lung Disease ◽  
Automatic Segmentation ◽  
Wiener Filter ◽  
Dice Similarity Coefficient ◽  
High Resolution Computed Tomography ◽  
Lung Segmentation ◽  
Training Performance ◽  
Lower Accuracy ◽  
Occurrence Matrix

Lung segmentation in high-resolution computed tomography (HRCT) images is necessary before the computer-aided diagnosis (CAD) of interstitial lung disease (ILD). Traditional methods are less intelligent and have lower accuracy of segmentation. This paper develops a novel automatic segmentation model using radiomics with a combination of hand-crafted features and deep features. The study uses ILD Database-MedGIFT from 128 patients with 108 annotated image series and selects 1946 regions of interest (ROI) of lung tissue patterns for training and testing. First, images are denoised by Wiener filter. Then, segmentation is performed by fusion of features that are extracted from the gray-level co-occurrence matrix (GLCM) which is a classic texture analysis method and U-Net which is a standard convolutional neural network (CNN). The final experiment result for segmentation in terms of dice similarity coefficient (DSC) is 89.42%, which is comparable to the state-of-the-art methods. The training performance shows the effectiveness for a combination of texture and deep radiomics features in lung segmentation.

scholarly journals Enhanced Classification of Interstitial Lung Disease Patterns in HRCT Images Using Differential Lacunarity

2015 ◽  
Vol 2015 ◽  
pp. 1-9 ◽  
Author(s):  
Verónica Vasconcelos ◽  
João Barroso ◽  
Luis Marques ◽  
José Silvestre Silva
Keyword(s):  
Interstitial Lung Disease ◽  
Lung Disease ◽  
Texture Analysis ◽  
Lung Parenchyma ◽  
Normal Lung ◽  
Support Vector ◽  
High Resolution Computed Tomography ◽  
Disease Patterns ◽  
Vector Machines ◽  
Texture Attributes

The analysis and interpretation of high-resolution computed tomography (HRCT) images of the chest in the presence of interstitial lung disease (ILD) is a time-consuming task which requires experience. In this paper, a computer-aided diagnosis (CAD) scheme is proposed to assist radiologists in the differentiation of lung patterns associated with ILD and healthy lung parenchyma. Regions of interest were described by a set of texture attributes extracted using differential lacunarity (DLac) and classical methods of statistical texture analysis. The proposed strategy to compute DLac allowed a multiscale texture analysis, while maintaining sensitivity to small details. Support Vector Machines were employed to distinguish between lung patterns. Training and model selection were performed over a stratified 10-fold cross-validation (CV). Dimensional reduction was made based on stepwise regression (F-test,pvalue < 0.01) during CV. An accuracy of 95.8 ± 2.2% in the differentiation of normal lung pattern from ILD patterns and an overall accuracy of 94.5 ± 2.1% in a multiclass scenario revealed the potential of the proposed CAD in clinical practice. Experimental results showed that the performance of the CAD was improved by combining multiscale DLac with classical statistical texture analysis.

scholarly journals High resolution computed tomography and chest radiography findings among interstitial lung disease patients

2019 ◽  
Vol 9 (4) ◽  
pp. 24-27
Author(s):  
Anusmriti Pal ◽  
Manoj Kumar Yadav ◽  
Chiranjibi Pant ◽  
Bishow Kumar Shrestha
Keyword(s):  
Computed Tomography ◽  
High Resolution ◽  
Interstitial Lung Disease ◽  
Lung Disease ◽  
Imaging Modality ◽  
Lung Parenchyma ◽  
Median Number ◽  
High Resolution Computed Tomography ◽  
Cross Sectional ◽  
Diffuse Parenchymal Lung Diseases

Background: Interstitial lung disease (ILD) is a heterogeneous group of diffuse parenchymal lung diseases, characterized by restrictive physiology, impaired gas exchange, pulmonary inflammation and fibrosis. Chest radiograph (CXR) may appear normal during initial course of the disease and may show few abnormalities. High resolution computed tomography (HRCT) chest is a most ac­curate non-invasive, high spatial resolution descriptive imaging modality for evaluation of lung parenchyma. It assesses presence, location, type and characterization of ILD in appropriate clinical setting. Our aim was to study radiological patterns and its distribution in CXR and HRCT chest of ILD patients. Methods: This was an observational, single centered, cross-sectional study conducted at author’s place over the period of 6 months starting from January 2018 using convenient sampling method. Data analysis was done using students t-test for comparison of means and chi-square test for proportions. Results: A total of 30 suspected or diagnosed patients of ILD were enrolled in our study and pat­terns found on CXR were correlated with that on HRCT chest. The number of findings in HRCT chest for a patient was significantly higher than CXR (Median number: 4 verses 2, P<0.001), commonest reticular opacity 50% in CXR and 56.6% HRCT. One subject had normal CXR. Conclusion: HRCT was superior to CXR in detection of all basic patterns and their distribution as­sociated with ILD as higher numbers of findings were detected by HRCT chest as compared to CXR. HRCT chest could characterize the abnormality and specify its location much more accurately.

High-resolution CT-derived measures of lung density are valid indexes of interstitial lung disease

1994 ◽  
Vol 76 (1) ◽  
pp. 271-277 ◽  
Author(s):  
P. G. Hartley ◽  
J. R. Galvin ◽  
G. W. Hunninghake ◽  
J. A. Merchant ◽  
S. J. Yagla ◽  
...  
Keyword(s):  
High Resolution ◽  
Interstitial Lung Disease ◽  
Lung Disease ◽  
Lung Parenchyma ◽  
Study Groups ◽  
Computer Assisted ◽  
Gray Scale ◽  
High Resolution Computed Tomography ◽  
Lung Density ◽  
The Mean

To assess the validity of computer-assisted methods in analyzing the lung parenchyma imaged with high-resolution computed tomography (HRCT), we compared computer-derived estimates of lung density to other, more traditional, measures of parenchymal injury in 24 subjects with idiopathic pulmonary fibrosis (IPF) and 60 subjects with extensive occupational exposure to asbestos. Gray scale density histograms were constructed from the HRCT images. The gray scale histogram of both study groups was of a skewed unimodal distribution. However, compared with the asbestos-exposed subjects, the patients with IPF had a gray scale distribution that was significantly shifted to the right (greater density) and flatter. In a multivariate analysis, after controlling for age and cigarette smoking, we found that the mean and median gray scale densities were independently associated with the presence of moderate-to-severe dyspnea, a higher International Labour Office chest X-ray category, a lower forced vital capacity, and a higher concentration of macrophages and eosinophils in the bronchoalveolar lavage fluid. These factors accounted for > 70% of the variance of the mean and median gray scale densities. Interestingly, no differences in gray scale density measures were noted between patients with IPF and patients with asbestosis when these other factors were taken into account. Our results suggest that computer-derived density analysis of the lung parenchyma on the HRCT scan is a valid, clinically meaningful, and objective measure of interstitial lung disease.

scholarly journals Rheumatoid Arthritis Related Interstitial Lung Disease: Patterns of High-resolution Computed Tomography

Cureus ◽  
2020 ◽  
Author(s):  
Mahesh Gautam ◽  
Mah Jabeen Masood ◽  
Sadaf Arooj ◽  
Mufazzal-e-Haque Mahmud ◽  
Muhammad Umer Mukhtar
Keyword(s):  
Rheumatoid Arthritis ◽  
Computed Tomography ◽  
High Resolution ◽  
Interstitial Lung Disease ◽  
Lung Disease ◽  
High Resolution Computed Tomography ◽  
Disease Patterns

scholarly journals Pulmonary abnormalities on high-resolution computed tomography in patients with long standing rheumatoid arthritis

2021 ◽  
Vol 30 (1) ◽  
pp. 15-20
Author(s):  
Iulia Andronache ◽  
◽  
Cristina Suta ◽  
Sabina Ciocodei ◽  
Ionut Bulbuc ◽  
...  
Keyword(s):  
Rheumatoid Arthritis ◽  
Computed Tomography ◽  
High Resolution ◽  
Lung Disease ◽  
Lung Parenchyma ◽  
Pulmonary Involvement ◽  
Wall Thickening ◽  
High Resolution Computed Tomography ◽  
Systemic Inflammatory Disease ◽  
Organ Manifestations

Background. Rheumatoid arthritis (RA) is a systemic inflammatory disease, associated with a number of extra-articular organ manifestations. Pulmonary involvement is a frequent and severe extraarticular manifestations of rheumatoid arthritis. RA can affect lung parenchyma, airways and pleura. Objectives. To identify RA-related lung disease on chest computed tomography (CT). Material and methods. We performed high-resolution computed tomography (HRCT) on a total of 92 patients with longstanding RA. Results. The subjects were predominantely female (79.3%), the age at entry was 63.77 ±11.56 years, and 42.9% had a history of smoking. Disease duration was 15.00±11.55 years. Pulmonary CT abnormalities were found in 71 of the 92 patients (77.2%). The most common HRCT anomalies were: linear attenuation (reticulation) (52.11%), bronchiectasis andbronchial wall thickening (45%), nodular attenuation (39.43%) and pleural involvement (pleural effusion or thickening) (39.43%). Conclusions. We conclude that RA-related lung disease was commonly detected on chest CT imaging in longstanding RA patients.

Sign in / Sign up

Export Citation Format

Share Document