Automatic Lung Segmentation Research Articles

Automatic Lung Segmentation in CT Images with Accurate Handling of the Hilar Region

A fully automated and three-dimensional (3D) segmentation method for the identification of the pulmonary parenchyma in thorax X-ray computed tomography (CT) datasets is proposed. It is meant to be used as pre-processing step in the computer-assisted detection (CAD) system for malignant lung nodule detection that is being developed by the Medical Applications in a Grid Infrastructure Connection (MAGIC-5) Project. In this new approach the segmentation of the external airways (trachea and bronchi), is obtained by 3D region growing with wavefront simulation and suitable stop conditions, thus allowing an accurate handling of the hilar region, notoriously difficult to be segmented. Particular attention was also devoted to checking and solving the problem of the apparent 'fusion' between the lungs, caused by partial-volume effects, while 3D morphology operations ensure the accurate inclusion of all the nodules (internal, pleural, and vascular) in the segmented volume. The new algorithm was initially developed and tested on a dataset of 130 CT scans from the Italung-CT trial, and was then applied to the ANODE09-competition images (55 scans) and to the LIDC database (84 scans), giving very satisfactory results. In particular, the lung contour was adequately located in 96% of the CT scans, with incorrect segmentation of the external airways in the remaining cases. Segmentation metrics were calculated that quantitatively express the consistency between automatic and manual segmentations: the mean overlap degree of the segmentation masks is 0.96 ± 0.02, and the mean and the maximum distance between the mask borders (averaged on the whole dataset) are 0.74 ± 0.05 and 4.5 ± 1.5, respectively, which confirms that the automatic segmentations quite correctly reproduce the borders traced by the radiologist. Moreover, no tissue containing internal and pleural nodules was removed in the segmentation process, so that this method proved to be fit for the use in the framework of a CAD system. Finally, in the comparison with a two-dimensional segmentation procedure, inter-slice smoothness was calculated, showing that the masks created by the 3D algorithm are significantly smoother than those calculated by the 2D-only procedure.

Automatic lung segmentation from thoracic computed tomography scans using a hybrid approach with error detection

Lung segmentation is a prerequisite for automated analysis of chest CT scans. Conventional lung segmentation methods rely on large attenuation differences between lung parenchyma and surrounding tissue. These methods fail in scans where dense abnormalities are present, which often occurs in clinical data. Some methods to handle these situations have been proposed, but they are too time consuming or too specialized to be used in clinical practice. In this article, a new hybrid lung segmentation method is presented that automatically detects failures of a conventional algorithm and, when needed, resorts to a more complex algorithm, which is expected to produce better results in abnormal cases. In a large quantitative evaluation on a database of 150 scans from different sources, the hybrid method is shown to perform substantially better than a conventional approach at a relatively low increase in computational cost.

Automatic segmentation and recognition of lungs and lesion from CT scans of thorax

In this study, a fully automated texture-based segmentation and recognition system for lesion and lungs from CT of thorax is presented. For the segmentation part, we have extracted texture features by Gabor filtering the images, and, then combined these features to segment the target volume by using Fuzzy C Means (FCM) clustering. Since clustering is sensitive to initialization of cluster prototypes, optimal initialization of the cluster prototypes was done by using a Genetic Algorithm. For the recognition stage, we have used cortex like mechanism for extracting statistical features in addition to shape-based features. The segmented regions showed a high degree of imbalance between positive and negative samples, so we employed over and under sampling for balancing the data. Finally, the balanced and normalized data was subjected to Support Vector Machine (SimpleSVM) for training and testing. Results reveal an accuracy of delineation to be 94.06%, 94.32% and 89.04% for left lung, right lung and lesion, respectively. Average sensitivity of the SVM classifier was seen to be 89.48%.

Automated lung segmentation algorithm for CAD system of thoracic CT

Objective To design and test the accuracy and efficiency of our lung segmentation algorithm on thoracic CT image in computer-aided diagnostic (CAD) system, especially on the segmentation between left and right lungs. Methods We put forward the base frame of our lung segmentation firstly. Then, using optimal thresholding and mathematical morphologic methods, we acquired the rough image of lung segmentation. Finally, we presented a fast self-fit segmentation refinement algorithm, adapting to the unsuccessful left-right lung segmentation of thredsholding. Then our algorithm was used to CT scan images of 30 patients and the results were compared with those made by experts. Results Experiments on clinical 2-D pulmonary images showed the results of our algorithm were very close to the expert's manual outlines, and it was very effective for the separation of left and right lungs with a successful segmentation ratio 94.8%. Conclusion It is a practicable fast lung segmentation algorithm for CAD system on thoracic CT image.

SU‐GG‐I‐96: Improvements for Fully Automatic Lung Segmentation From RT Planning Scans

Purpose: Segmentation of pulmonary X‐ray computed tomography (CT) images is a primordial step to most pulmonary image analysis applications. Many approaches have been presented in the past mainly focusing on lung extraction from high resolution CT data. Looking towards RT planning one has often to deal with low image quality, artifacts and pathological changes of the anatomy. We present a fully automatic method for delineation of the lungs and the trachea in three‐dimensional RT planning scans for thorax and breast patients. Method and Materials: The method consists of several steps. First, lung region is extracted automatically computing an optimal threshold method. Then, the trachea is separated from the lungs by fast marching. Next left and right lungs are separated in each slice by computing a minimal path in the area where separation is suspected from the information of the neighboring slices. Finally the 3D segmentations of the left and right lungs are refined running the random walker algorithm with foreground and background seeds automatically calculated from the intensity‐based pre‐segmentation. Results: Processing time is below one minute and does not require any user input. Several 3D CT Datasets from different patients were used to evaluate the methods. The algorithms can segment the trachea and separate the left and right lung. In comparison to previous publications stability could be improved. Still in some cases result are not satisfying, e.g. trachea segmentation reaches into the lungs and lesions or emphysema are not included in the lung segmentations. Conclusion: The presented algorithm allows for fully automatic lung segmentation. One major improvement could be to employ a shape model as recently published by other groups as a shape‐constrained segmentation method could help to overcome the observed problems with exclusion of pathologies.

Combining 2D wavelet edge highlighting and 3D thresholding for lung segmentation in thin-slice CT

The first step in lung analysis by CT is the identification of the lung border. To deal with the increased number of sections per scan in thin-slice multidetector CT, it has been crucial to develop accurate and automated lung segmentation algorithms. In this study, an automated method for lung segmentation of thin-slice CT data is presented. The method exploits the advantages of a two-dimensional wavelet edge-highlighting step in lung border delineation. Lung volume segmentation is achieved with three-dimensional (3D) grey level thresholding, using a minimum error technique. 3D thresholding, combined with the wavelet pre-processing step, successfully deals with lung border segmentation challenges, such as anterior or posterior junction lines and juxtapleural nodules. Finally, to deal with mediastinum border under-segmentation, 3D morphological closing with a spherical structural element is applied. The performance of the proposed method is quantitatively assessed on a dataset originating from the Lung Imaging Database Consortium (LIDC) by comparing automatically derived borders with the manually traced ones. Segmentation performance, averaged over left and right lung volumes, for lung volume overlap is 0.983+/-0.008, whereas for shape differentiation in terms of mean distance it is 0.770+/-0.251 mm (root mean square distance is 0.520+/-0.008 mm; maximum distance is 3.327+/-1.637 mm). The effect of the wavelet pre-processing step was assessed by comparing the proposed method with the 3D thresholding technique (applied on original volume data). This yielded statistically significant differences for all segmentation metrics (p<0.01). Results demonstrate an accurate method that could be used as a first step in computer lung analysis by CT.

515 POSTER Transcriptional profiling of palmerolide A, a putative inhibitor of V-ATPase, indicates perturbation of cholesterol biosynthesis

Quantitative lung computed tomographic (CT) analysis yields objective data regarding lung aeration but is currently not used in clinical routine primarily because of the labor-intensive process of manual CT segmentation. Automatic lung segmentation could help to shorten processing times significantly. In this study, we assessed bias and precision of lung CT analysis using automatic segmentation compared with manual segmentation. In this monocentric clinical study, 10 mechanically ventilated patients with mild to moderate acute respiratory distress syndrome were included who had received lung CT scans at 5- and 45-mbar airway pressure during a prior study. Lung segmentations were performed both automatically using a computerized algorithm and manually. Automatic segmentation yielded similar lung volumes compared with manual segmentation with clinically minor differences both at 5 and 45 mbar. At 5 mbar, results were as follows: overdistended lung 49.58 mL (manual, SD 77.37 mL) and 50.41 mL (automatic, SD 77.3 mL), P = .028; normally aerated lung 2142.17 mL (manual, SD 1131.48 mL) and 2156.68 mL (automatic, SD 1134.53 mL), P = .1038; and poorly aerated lung 631.68 mL (manual, SD 196.76 mL) and 646.32 mL (automatic, SD 169.63 mL), P = .3794. At 45 mbar, values were as follows: overdistended lung 612.85 mL (manual, SD 449.55 mL) and 615.49 mL (automatic, SD 451.03 mL), P = .078; normally aerated lung 3890.12 mL (manual, SD 1134.14 mL) and 3907.65 mL (automatic, SD 1133.62 mL), P = .027; and poorly aerated lung 413.35 mL (manual, SD 57.66 mL) and 469.58 mL (automatic, SD 70.14 mL), P = .007. Bland-Altman analyses revealed the following mean biases and limits of agreement at 5 mbar for automatic vs manual segmentation: overdistended lung +0.848 mL (±2.062 mL), normally aerated +14.51 mL (±49.71 mL), and poorly aerated +14.64 mL (±98.16 mL). At 45 mbar, results were as follows: overdistended +2.639 mL (±8.231 mL), normally aerated 17.53 mL (±41.41 mL), and poorly aerated 56.23 mL (±100.67 mL). Automatic single CT image and whole lung segmentation were faster than manual segmentation (0.17 vs 125.35 seconds [P < .0001] and 10.46 vs 7739.45 seconds [P < .0001]). Automatic lung CT segmentation allows fast analysis of aerated lung regions. A reduction of processing times by more than 99% allows the use of quantitative CT at the bedside.

Automated lung segmentation of diseased and artifact-corrupted magnetic resonance sectionsa)

Segmentation of the lungs within magnetic resonance (MR) scans is a necessary step in the computer-based analysis of thoracic MR images. This process is often confounded by image acquisition artifacts and disease-induced morphological deformation. We have developed an automated method for lung segmentation that is insensitive to these complications. The automated method was applied to 23 thoracic MR scans (413 sections) obtained from 10 patients. Two radiologists manually outlined the lung regions in a random sample of 101 sections (n=202 lungs), and the extent to which disease or artifact confounded lung border visualization was evaluated. Accuracy of lung regions extracted by the automated segmentation method was quantified by comparison with the radiologist-defined lung regions using an area overlap measure (AOM) that ranged from 0 (disjoint lung regions) to 1 (complete overlap). The AOM between each observer and the automated method was 0.82 when averaged over all lungs. The average AOM in the lung bases, where lung segmentation is most difficult, was 0.73.

Smoothing Lung Segmentation Surfaces in Three-dimensional X-ray CT Images Using Anatomic Guidance 1

Automatic lung segmentation in volumetric computed tomography (CT) images has been extensively investigated, and several methods have been proposed. Most methods distinguish the lung parenchyma from the surrounding anatomy based on the difference in CT attenuation values. This leads to an irregular and inconsistent lung boundary for the regions near the mediastinum, which can cause inconsistent boundaries both across subjects and within subjects scanned at different intervals of time. Processes like lung image registration and lung atlas construction can be affected by such inconsistencies. Therefore there is a need for a more consistent lung surface near the mediastinum. This paper presents a fully automatic method for the three-dimensional smoothing of the lung boundary using information from the segmented human airway tree. First, using the segmented airway tree, we define a bounding box around the mediastinum for each lung, within which all operations are performed. We then define all generations of the airway tree distal to the right and left main stem bronchi to be part of the respective lungs and exclude all other segments. Finally, we perform a fast morphological closing with an ellipsoidal kernel to smooth the surface of the lung. This method has been tested by processing the segmented lungs from eight normal datasets. The mean value of the magnitude of curvedness of the lung contours, averaged over all the datasets, is 0.13 postsmoothing, compared with 4.91 before smoothing. The accuracy of the lung contours after smoothing is assessed by comparing the automatic results to manually traced smooth lung borders by a human analyst. Averaged over all volumes, the root mean square difference between human and computer borders is 0.8691 mm after smoothing, compared with 1.3012 mm before smoothing. The mean similarity index, which is an area overlap measure based on the kappa statistic, is 0.9958 (SD 0.0032), after smoothing. We have described a novel scheme for smoothing the lung contour around the mediastinum. The method is based on using anatomic information from the segmented airway tree. The validation results show that there is good agreement between manual and computer results. Because there are no accepted criteria for defining the lung boundary near the mediastinum, we believe our method of defining the boundary based on the structure of the airway tree provides a good basis for three-dimensional smoothing.

Evaluation of Semiautomated Measurements of Mesothelioma Tumor Thickness on CT Scans 1

To evaluate the clinical acceptability of semiautomated methods for the measurement of mesothelioma tumor thickness in computed tomography (CT) scans. A computer interface was developed to allow the acquisition of semiautomated mesothelioma tumor thickness measurements, which require the manual selection of a point along the outer margin of the tumor in a CT section. After application of an automated lung segmentation method, the computer automatically identifies a corresponding point along the inner margin of the tumor (as represented by the lung boundary), constructs a line segment between the manually selected outer tumor margin point and the computer-determined inner tumor margin point, and computes tumor thickness as the length of this line segment. Three radiologists and oncologists independently reviewed line segments representing the semiautomated measurements generated by three different algorithms at 134 measurement sites in the CT scans of 22 mesothelioma patients. The observers either accepted a measurement line segment or modified it through the interface. Differences between the initial semiautomated measurements and the measurements as modified by the observers were analyzed. The frequency with which observers accepted the semiautomated measurements without modification was as high as 86%. Of all measurements across all observers and methods (1,206 measurements), 89% were changed by 2 mm or less. We have developed semiautomated methods to measure mesothelioma tumor thickness. The potential of these methods has been demonstrated through an observer study. We expect these methods to become important tools for the efficient quantification of tumor extent.

Automated lung segmentation for thoracic CT: Impact on computer-aided diagnosis 1

Automated lung segmentation in thoracic computed tomography scans is essential for the development of computer-aided diagnostic (CAD) methods. A core segmentation method may be developed for general application; however, modifications may be required for specific clinical tasks. An automated lung segmentation method has been applied (1) as preprocessing for automated lung nodule detection and (2) as the foundation for computer-assisted measurements of pleural mesothelioma tumor thickness. The core method uses gray-level thresholding to segment the lungs within each computed tomography section. The segmentation is revised through separation of right and left lungs along the anterior junction line, elimination of the trachea and main bronchi from the lung segmentation regions, and suppression of the diaphragm. Segmentation modifications required for nodule detection include a rolling ball algorithm to include juxtapleural nodules and morphologic erosion to eliminate partial volume pixels at the boundary of the segmentation regions. For automated lung nodule detection, 4 of 82 actual nodules (4.9%) were excluded from the lung segmentation regions when the core segmentation method was modified compared with 14 nodules (17.1%) excluded without modifications. The computer-assisted quantification of mesothelioma method achieved a correlation coefficient of 0.990 with 134 manual measurements when the core segmentation method was used alone; correlation was reduced to 0.977 when the segmentation modifications, as adapted for the lung nodule detection task, were applied to the mesothelioma measurement task. Different CAD applications impose different requirements on the automated lung segmentation process. The specific approach to lung segmentation must be adapted to the particular CAD task.

Automated lung segmentation in X-ray computed tomography: development and evaluation of a heuristic threshold-based scheme 1

Rationale and Objectives. To develop and evaluate a reliable, fully-automated lung segmentation scheme for application in X-ray computed tomography. Materials and Methods. The automated scheme was heuristically developed using a slice-based, pixel-value threshold and two sets of classification rules. Features used in the rules include size, circularity, and location. The segmentation scheme operates slice-by-slice and performs three key operations: (1) image preprocessing to remove background pixels, (2) computation and application of a pixel-value threshold to identify lung tissue, and (3) refinement of the initial segmented regions to prune incorrectly detected airways and separate fused right and left lungs. Results. The performance of the automated segmentation scheme was evaluated using 101 computed tomography cases (91 thick slice, 10 thin slice scans). The 91 thick cases were pre- and post-surgery from 50 patients and were not independent. The automated scheme successfully segmented 94.0% of the 2,969 thick slice images and 97.6% of the 1,161 thin slice images. The mean difference of the total lung volumes calculated by the automated scheme and functional residual capacity plus 60% inspiratory capacity was −24.7 ± 508.1 mL. The mean differences of the total lung volumes calculated by the automated scheme and an established, commonly used semi-automated scheme were 95.2 ± 52.5 mL and −27.7 ± 66.9 mL for the thick and thin slice cases, respectively. Conclusion. This simple, fully-automated lung segmentation scheme provides an objective tool to facilitate lung segmentation from computed tomography scans.

Po-topic V-03: Automated lung segmentation in thoracic CT scans: applications for computer-aided diagnosis research

Po-topic V-03: Automated lung segmentation in thoracic CT scans: applications for computer-aided diagnosis research

Automated lung segmentation and computer-aided diagnosis for thoracic CT scans

Automated segmentation of the lungs in thoracic computed tomography (CT) scans represents an essential process in the development of computer-aided diagnostic (CAD) methods and computer-assisted quantification techniques. A core segmentation process may be developed for general application; however, modifications may be required for specific clinical tasks. We have developed such an automated lung segmentation method based on gray-level thresholding techniques and have applied this method (1) as a pre-processing step for automated lung nodule detection and (2) as the foundation for a computer-assisted technique to measure the extent of pleural mesothelioma. In the automated detection of lung nodules, we have developed a method that has achieved 71% nodule detection sensitivity with an average of 0.4 false-positive detections per section on a database of 38 CT scans. Our method for the computer-assisted quantification of mesothelioma achieved a correlation coefficient of 0.97 with the average manual measurements of four observers based on 134 measurement sites in 22 CT scans. Important differences exist in the specific approaches to automated lung segmentation required for these two clinical tasks.

A software tool for automatic image-based ventilation analysis using dynamic chest CT-scanning in healthy and in ARDS lungs

Density measurements in dynamic CT image series of the lungs allow one to quantify ventilated, hyperinflated, and atelectatic pulmonary compartments with high temporal resolution. Fast automatic segmentation of lung parenchyma and a subsequent evaluation of it's respective density values are a prerequisite for any clinical application of this technique. For automatic lung segmentation in thoracic CT scans, an algorithm was developed which uses (a) different density masks, and (b) anatomic knowledge to differentiate heart, diaphragm and chest wall from ventilated and atelectatic lung parenchyma. With Animal Care Committee approval, the automated technique was tested in 8 anaesthetized ventilated pigs undergoing dynamic CT before and after induction of lavage-ARDS. Images were acquired in one supradiaphragmatic, cross-sectional slice (temporal resolution of 100 ms; slice thickness of 1 mm, high resolution reconstruction algorithm). In 120 CT images the total pixel number and the calculated MLD from the automatically segmentated lung were compared to the values obtained from an interactive lung segmentation. The software tool was able to read all image series (DICOM standard). Automatic and interactive segmentation were in high agreement (R(2) = 0.99 for the total number of pixels and the MLD). Originally, the most frequent error was misclassification of atelectasis as extrapulmonary solid tissue. An automatic software tool is presented for lung segmentation in healthy lungs and in ARDS. Aerated lung and atelectasis were identified with high accuracy. This post-processing tool allows for a quantitative, CT based assessment of ventilation and recruitment processes in the lung. Thus, it may help to optimize ventilation patterns in patients with ARDS.

Automatic lung segmentation for accurate quantitation of volumetric X-ray CT images.

Segmentation of pulmonary X-ray computed tomography (CT) images is a precursor to most pulmonary image analysis applications. This paper presents a fully automatic method for identifying the lungs in three-dimensional (3-D) pulmonary X-ray CT images. The method has three main steps. First, the lung region is extracted from the CT images by gray-level thresholding. Then, the left and right lungs are separated by identifying the anterior and posterior junctions by dynamic programming. Finally, a sequence of morphological operations is used to smooth the irregular boundary along the mediastinum in order to obtain results consistent with those obtained by manual analysis, in which only the most central pulmonary arteries are excluded from the lung region. The method has been tested by processing 3-D CT data sets from eight normal subjects, each imaged three times at biweekly intervals with lungs at 90% vital capacity. We present results by comparing our automatic method to manually traced borders from two image analysts. Averaged over all volumes, the root mean square difference between the computer and human analysis is 0.8 pixels (0.54 mm). The mean intrasubject change in tissue content over the three scans was 2.75% +/- 2.29% (mean +/- standard deviation).

Automated lung segmentation in digital lateral chest radiographs.

We are developing a fully automated computerized scheme for segmenting the lung fields in digital lateral chest radiographs. Existing computer-aided diagnostic (CAD) schemes and automated lung segmentation methods have focused exclusively on the posteroanterior view, despite the diagnostic importance of the lateral view. Information from the lateral radiograph is routinely incorporated by radiologists in their decision-making process, and thus computer analysis of lateral images may potentially add another dimension to current CAD schemes. Automated analysis of the lung fields in lateral images will necessarily require accurate segmentation. Our scheme employs an initial procedure to eliminate external and subcutaneous pixels. Global gray-level histogram analysis then allows for the identification of a range of gray-level thresholds. An iterative gray-level thresholding scheme is implemented using this range of thresholds to construct a series of binary images in which contiguous regions are identified and geometrically analyzed. Regions determined to be outside the lungs are prevented from contributing to binary images at later iterations. Adaptive local gray-level thresholding is applied along the initial contour that results from the global thresholding procedure to extend the contour closer to the true lung borders. This local thresholding method uses regions of interest of various dimensions, depending on the enclosed anatomy. Smoothing and polynomial curve fitting complete the segmentation. A database of 100 normal and 100 abnormal lateral images was analyzed. Quantitative comparison of computer-segmented lung regions with lung regions manually delineated by two radiologists indicated that 83% and 84% of normal and abnormal images, respectively, displayed segmentation contours within three standard deviations of the mean inter-radiologist contour degree-of-overlap value.

Automated lung segmentation in digitized posteroanterior chest radiographs

The authors developed and tested a gray-level thresholding-based approach to automated lung segmentation in digitized posteroanterior chest radiographs. Gray-level histogram analysis was initially performed to establish a range of thresholds for use during an iterative global gray-level thresholding technique. Local gray-level threshold analysis was then performed on the output of global thresholding. The resulting contours were subjected to several smoothing processes, including a rolling-ball technique. The final contours closely approximated the boundaries of the aerated lung regions. The method was applied to a database of 600 posteroanterior chest images. Radiologists rated the accuracy and completeness of the contours with a five-point scale. Results of the subjective rating evaluation indicated that this method was accurate, with 79% of the assigned ratings reflecting moderately or highly accurate segmentation and only 8% of the ratings indicating moderately or highly inaccurate segmentation. This gray-level thresholding-based approach provides accurate automated lung segmentation in digital posteroanterior chest radiographs.

Automated lung segmentation in digital posteroanterior and lateral chest radiographs: Applications in diagnostic radiology and nuclear medicine

Automated segmentation of anatomic regions in digital radiographs is a key element of computer‐aided diagnostic (CAD) methods. Computerized methods for automatically segmenting the aerated lung fields from digital posteroanterior (PA) and lateral chest radiographs have been developed. The methods are based on gray‐level thresholding techniques, including iterative global gray‐level thresholding and local gray‐level thresholding, thus reducing the need for a priori assumptions regarding overall lung morphology. Consequently, this method is particularly applicable to the detection of abnormal asymmetry in PA radiographs, where pathological processes may produce gross alterations in the appearance of radiographic lung anatomy. An anatomic region of particular interest is the costophrenic angle. The method described lends itself to the automated assessment of costophrenic angle blunting, for which a technique has been developed. Another application of this method crosses into the field of nuclear medicine. Radionuclide lung scans are typically used in conjunction with chest radiographs for diagnostic purposes. An automated method for registering digital PA and lateral chest radiographs with the perfusion and ventilation images of a lung scan has been developed. These computerized techniques may be incorporated into future CAD schemes and may potentially aid radiologists in the interpretation and assessment of medical images.