Two-dimensional multi-criterion segmentation of pulmonary nodules on helical CT images

A novel higher order appearance texture analysis to diagnose lung cancer based on a modified local ternary pattern

BACKGROUND AND OBJECTIVE

Lung cancer is an important cause of death and morbidity around the world. Two of the primary computed tomography (CT) imaging markers that can be used to differentiate malignant and benign lung nodules are the inhomogeneity of the nodules' texture and nodular morphology. The objective of this paper is to present a new model that can capture the inhomogeneity of the detected lung nodules as well as their morphology.

METHODS

We modified the local ternary pattern to use three different levels (instead of two) and a new pattern identification algorithm to capture the nodule's inhomogeneity and morphology in a more accurate and flexible way. This modification aims to address the wide Hounsfield unit value range of the detected nodules which decreases the ability of the traditional local binary/ternary pattern to accurately classify nodules' inhomogeneity. The cut-off values defining these three levels of the novel technique are estimated empirically from the training data. Subsequently, the extracted imaging markers are fed to a hyper-tuned stacked generalization-based classification architecture to classify the nodules as malignant or benign. The proposed system was evaluated on in vivo datasets of 679 CT scans (364 malignant nodules and 315 benign nodules) from the benchmark Lung Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI) and an external dataset of 100 CT scans (50 malignant and 50 benign). The performance of the classifier was quantitatively assessed using a Leave-one-out cross-validation approach and externally validated using the unseen external dataset based on sensitivity, specificity, and accuracy.

RESULTS

The overall accuracy of the system is 96.17% with 97.14% sensitivity and 95.33% specificity. The area under the receiver-operating characteristic curve was 0.98, which highlights the robustness of the system. Using the unseen external dataset for validating the system led to consistent results showing the generalization abilities of the proposed approach. Moreover, applying the original local binary/ternary pattern or using other classification structures achieved inferior performance when compared against the proposed approach.

CONCLUSIONS

These experimental results demonstrate the feasibility of the proposed model as a novel tool to assist physicians and radiologists for lung nodules' early assessment based on the new comprehensive imaging markers.

A Feature-based Affine Registration Method for Capturing Background Lung Tissue Deformation for Ground Glass Nodule Tracking

Lung nodule tracking assessment relies on cross-sectional measurements of the largest lesion profile depicted in initial and follow-up computed tomography (CT) images. However, apparent changes in nodule size assessed via simple image-based measurements may also be compromised by the effect of the background lung tissue deformation on the GGN between the initial and follow-up images, leading to erroneous conclusions about nodule changes due to disease. To compensate for the lung deformation and enable consistent nodule tracking, here we propose a feature-based affine registration method and study its performance vis-a-vis several other registration methods. We implement and test each registration method using both a lung- and a lesion-centered region of interest on ten patient CT datasets featuring twelve nodules, including both benign and malignant GGO lesions containing pure GGNs, part-solid, or solid nodules. We evaluate each registration method according to the target registration error (TRE) computed across 30 - 50 homologous fiducial landmarks surrounding the lesions and selected by expert radiologists in both the initial and follow-up patient CT images. Our results show that the proposed feature-based affine lesion-centered registration yielded a 1.1 ± 1.2 mm TRE, while a Symmetric Normalization deformable registration yielded a 1.2 ± 1.2 mm TRE, and a least-square fit registration of the 30-50 validation fiducial landmark set yielded a 1.5 ± 1.2 mm TRE. Although the deformable registration yielded a slightly higher registration accuracy than the feature-based affine registration, it is significantly more computationally efficient, eliminates the need for ambiguous segmentation of GGNs featuring ill-defined borders, and reduces the susceptibility of artificial deformations introduced by the deformable registration, which may lead to increased similarity between the registered initial and follow-up images, over-compensating for the background lung tissue deformation, and, in turn, compromising the true disease-induced nodule change assessment. We also assessed the registration qualitatively, by visual inspection of the subtraction images, and conducted a pilot pre-clinical study that showed the proposed feature-based lesion-centered affine registration effectively compensates for the background lung tissue deformation between the initial and follow-up images and also serves as a reliable baseline registration method prior to assessing lung nodule changes due to disease.

A review of automatic lung tumour segmentation in the era of 4DCT

Aim

To review the literature on auto-contouring methods of lung tumour volumes on four-dimensional computed tomography (4DCT).

Background

Manual delineation of lung tumour on 4DCT has been the gold standard in clinical practice. However, it is resource intensive due to the high volume of data which results in longer contouring duration and uncertainties in defining target. Auto-contouring may present as an attractive alternative by decreasing manual inputs required, thus improving the contouring process. This review aims to assess the accuracy, variability and contouring duration of automatic contouring compared with manual contouring in lung cancer on 4DCT datasets.

Materials and methods

A search and review of literature were conducted to identify studies regarding lung tumour contouring on 4DCT. Manual and auto-contours were assessed and compared based on accuracy, variability and contouring duration.

Results

Thirteen studies were included in this review and their results were compared. Accuracy of auto-contours was found to be comparable to manual contours. Auto-contouring resulted in lesser inter-observer variation when compared to manual contouring, however there was no significant reduction in intra-observer variability. Additionally, contouring duration was reduced with auto-contouring although long computation time could present as a bottleneck.

Conclusion

Auto-contouring is reliable and efficient, producing accurate contours with better consistency compared to manual contours. However, manual inputs would still be required both before and after auto-propagation.

Vessel suppressed chest Computed Tomography for semi-automated volumetric measurements of solid pulmonary nodules

OBJECTIVE

To evaluate whether vessel-suppressed computed tomography (VSCT) can be reliably used for semi-automated volumetric measurements of solid pulmonary nodules, as compared to standard CT (SCT) MATERIAL AND METHODS: Ninety-three SCT were elaborated by dedicated software (ClearRead CT, Riverain Technologies, Miamisburg, OH, USA), that allows subtracting vessels from lung parenchyma. Semi-automated volumetric measurements of 65 solid nodules were compared between SCT and VSCT. The measurements were repeated by two readers. For each solid nodule, volume measured on SCT by Reader 1 and Reader 2 was averaged and the average volume between readers acted as standard of reference value. Concordance between measurements was assessed using Lin's Concordance Correlation Coefficient (CCC). Limits of agreement (LoA) between readers and CT datasets were evaluated.

RESULTS

Standard of reference nodule volume ranged from 13 to 366 mm³. The mean overestimation between readers was 3 mm³ and 2.9 mm³ on SCT and VSCT, respectively. Semi-automated volumetric measurements on VSCT showed substantial agreement with the standard of reference (Lin's CCC = 0.990 for Reader 1; 0.985 for Reader 2). The upper and lower LoA between readers' measurements were (16.3, -22.4 mm³) and (15.5, -21.4 mm³) for SCT and VSCT, respectively.

CONCLUSIONS

VSCT datasets are feasible for the measurements of solid nodules, showing an almost perfect concordance between readers and with measurements on SCT.

The influence of inspiratory effort and emphysema on pulmonary nodule volumetry reproducibility

AIM

To evaluate the impact of inspiratory effort and emphysema on reproducibility of pulmonary nodule volumetry.

MATERIALS AND METHODS

Eighty-eight nodules in 24 patients with emphysema were studied retrospectively. All patients had undergone volumetric inspiratory and end-expiratory thoracic computed tomography (CT) for consideration of bronchoscopic lung volume reduction. Inspiratory and expiratory nodule volumes were measured using commercially available software. Local emphysema extent was established by analysing a segmentation area extended circumferentially around each nodule (quantified as percent of lung with density of -950 HU or less). Lung volumes were established using the same software. Differences in inspiratory and expiratory nodule volumes were illustrated using the Bland-Altman test. The influences of percentage reduction in lung volume at expiration, local emphysema extent, and nodule size on nodule volume variability were tested with multiple linear regression.

RESULTS

The majority of nodules (59/88 [67%]) showed an increased volume at expiration. Mean difference in nodule volume between expiration and inspiration was +7.5% (95% confidence interval: -24.1, 39.1%). No relationships were demonstrated between nodule volume variability and emphysema extent, degree of expiration, or nodule size.

CONCLUSION

Expiration causes a modest increase in volumetry-derived nodule volumes; however, the effect is unpredictable. Local emphysema extent had no significant effect on volume variability in the present cohort.

Potential lung nodules identification for characterization by variable multistep threshold and shape indices from CT images

Computed tomography (CT) is an important imaging modality. Physicians, surgeons, and oncologists prefer CT scan for diagnosis of lung cancer. However, some nodules are missed in CT scan. Computer aided diagnosis methods are useful for radiologists for detection of these nodules and early diagnosis of lung cancer. Early detection of malignant nodule is helpful for treatment. Computer aided diagnosis of lung cancer involves lung segmentation, potential nodules identification, features extraction from the potential nodules, and classification of the nodules. In this paper, we are presenting an automatic method for detection and segmentation of lung nodules from CT scan for subsequent features extraction and classification. Contribution of the work is the detection and segmentation of small sized nodules, low and high contrast nodules, nodules attached with vasculature, nodules attached to pleura membrane, and nodules in close vicinity of the diaphragm and lung wall in one-go. The particular techniques of the method are multistep threshold for the nodule detection and shape index threshold for false positive reduction. We used 60 CT scans of “Lung Image Database Consortium-Image Database Resource Initiative” taken by GE medical systems LightSpeed16 scanner as dataset and correctly detected 92% nodules. The results are reproducible.

Computerized segmentation of pulmonary nodules depicted in CT examinations using freehand sketches

PURPOSE

To aid a consistent segmentation of pulmonary nodules, the authors describe a novel computerized scheme that utilizes a freehand sketching technique and an improved break-and-repair strategy.

METHODS

This developed scheme consists of two primary parts. The first part is freehand sketch analysis, where the freehand sketching not only serves a natural way of specifying the location of a nodule, but also provides a mechanism for inferring adaptive information (e.g., the mass center, the density, and the size) in regard to the nodule. The second part is an improved break-and-repair strategy. The improvement avoids the time-consuming ray-triangle intersections using spherical bins and replaces the original global implicit surface reconstruction with a local implicit surface fitting and blending scheme. The performance of this scheme, including accuracy and consistence, was assessed using 50 CT examinations in the Lung Image Database Consortium (LIDC). For each of these examinations, a single nodule was selected under the aid of a publically available tool to assure these nodules were diverse in size, location, and density. Two radiologists were asked to use the developed tool to segment these nodules twice at different times (at least three months apart). A Hausdorff distance based method was used to assess the discrepancies (agreements) between the computerized results and the results by the four radiologists in the LIDC as well as the inter- and intrareader agreements in freehand sketching.

RESULTS

The maximum and mean discrepancies in boundary outlines between the computerized scheme and the radiologists were 2.73 ± 1.32 mm and 1.01 ± 0.47 mm, respectively. When the nodules were classified (binned) into different size ranges, the maximum errors ranged from 1.91 to 4.13 mm; but smaller nodules had larger percentage discrepancies in term of size. Under the aid of the developed scheme, the inter- and intrareader variability in averaged maximum discrepancy across all types of pulmonary nodules were consistently smaller than 0.15 ± 0.07 mm. The computational cost in time of segmenting a pulmonary nodule ranged from 0.4 to 2.3 s with an average of 1.1 s for a typical desktop computer.

CONCLUSIONS

The experiments showed that this scheme could achieve a reasonable performance in nodule segmentation and demonstrated the merits of incorporating freehand sketching into pulmonary nodule segmentation.

A novel approach for lung nodules segmentation in chest CT using level sets

A new variational level set approach is proposed for lung nodule segmentation in lung CT scans. A general lung nodule shape model is proposed using implicit spaces as a signed distance function. The shape model is fused with the image intensity statistical information in a variational segmentation framework. The nodule shape model is mapped to the image domain by a global transformation that includes inhomogeneous scales, rotation, and translation parameters. A matching criteria between the shape model and the image implicit representations is employed to handle the alignment process. Transformation parameters evolve through gradient descent optimization to handle the shape alignment process and hence mark the boundaries of the nodule “head.” The embedding process takes into consideration the image intensity as well as prior shape information. A nonparametric density estimation approach is employed to handle the statistical intensity representation of the nodule and background regions. The proposed technique does not depend on nodule type or location. Exhaustive experimental and validation results are demonstrated on 742 nodules obtained from four different CT lung databases, illustrating the robustness of the approach.

Segmentation of lung lesions on CT scans using watershed, active contours, and Markov random field

PURPOSE

Lung lesions vary considerably in size, density, and shape, and can attach to surrounding anatomic structures such as chest wall or mediastinum. Automatic segmentation of the lesions poses a challenge. This work communicates a new three-dimensional algorithm for the segmentation of a wide variety of lesions, ranging from tumors found in patients with advanced lung cancer to small nodules detected in lung cancer screening programs.

METHODS

The authors' algorithm uniquely combines the image processing techniques of marker-controlled watershed, geometric active contours as well as Markov random field (MRF). The user of the algorithm manually selects a region of interest encompassing the lesion on a single slice and then the watershed method generates an initial surface of the lesion in three dimensions, which is refined by the active geometric contours. MRF improves the segmentation of ground glass opacity portions of part-solid lesions. The algorithm was tested on an anthropomorphic thorax phantom dataset and two publicly accessible clinical lung datasets. These clinical studies included a same-day repeat CT (prewalk and postwalk scans were performed within 15 min) dataset containing 32 lung lesions with one radiologist's delineated contours, and the first release of the Lung Image Database Consortium (LIDC) dataset containing 23 lung nodules with 6 radiologists' delineated contours. The phantom dataset contained 22 phantom nodules of known volumes that were inserted in a phantom thorax.

RESULTS

For the prewalk scans of the same-day repeat CT dataset and the LIDC dataset, the mean overlap ratios of lesion volumes generated by the computer algorithm and the radiologist(s) were 69% and 65%, respectively. For the two repeat CT scans, the intra-class correlation coefficient (ICC) was 0.998, indicating high reliability of the algorithm. The mean relative difference was -3% for the phantom dataset.

CONCLUSIONS

The performance of this new segmentation algorithm in delineating tumor contour and measuring tumor size illustrates its potential clinical value for assisting in noninvasive diagnosis of pulmonary nodules, therapy response assessment, and radiation treatment planning.

Computer-aided diagnosis systems for lung cancer: challenges and methodologies

This paper overviews one of the most important, interesting, and challenging problems in oncology, the problem of lung cancer diagnosis. Developing an effective computer-aided diagnosis (CAD) system for lung cancer is of great clinical importance and can increase the patient's chance of survival. For this reason, CAD systems for lung cancer have been investigated in a huge number of research studies. A typical CAD system for lung cancer diagnosis is composed of four main processing steps: segmentation of the lung fields, detection of nodules inside the lung fields, segmentation of the detected nodules, and diagnosis of the nodules as benign or malignant. This paper overviews the current state-of-the-art techniques that have been developed to implement each of these CAD processing steps. For each technique, various aspects of technical issues, implemented methodologies, training and testing databases, and validation methods, as well as achieved performances, are described. In addition, the paper addresses several challenges that researchers face in each implementation step and outlines the strengths and drawbacks of the existing approaches for lung cancer CAD systems.

Segmentation of juxtapleural pulmonary nodules using a robust surface estimate

An algorithm was developed to segment solid pulmonary nodules attached to the chest wall in computed tomography scans. The pleural surface was estimated and used to segment the nodule from the chest wall. To estimate the surface, a robust approach was used to identify points that lie on the pleural surface but not on the nodule. A 3D surface was estimated from the identified surface points. The segmentation performance of the algorithm was evaluated on a database of 150 solid juxtapleural pulmonary nodules. Segmented images were rated on a scale of 1 to 4 based on visual inspection, with 3 and 4 considered acceptable. This algorithm offers a large improvement in the success rate of juxtapleural nodule segmentation, successfully segmenting 98.0% of nodules compared to 81.3% for a previously published plane-fitting algorithm, which will provide for the development of more robust automated nodule measurement methods.

Elastic phantoms generated by microfluidics technology: validation of an imaged-based approach for accurate measurement of the growth rate of lung nodules

This paper focuses on validating our approach for monitoring the development of lung nodules detected in successive chest low-dose computed tomography (LDCT) scans of a patient. Our methodology for monitoring detected lung nodules includes 3D LDCT data registration, which is a non-rigid technique and involves two steps: (i) global target-to-prototype alignment of one scan to another using the experience gained from a prior appearance model, followed by (ii) local alignment to correct for intricate relative deformations. We propose a new approach for validating the accuracy of our algorithm for elastic lung phantoms constructed with state-of-the-art microfluidics technology and in vivo data. Fabricated from a flexible transparent polymer, i.e. polydimethylsiloxane (PDMS), the phantoms mimic the contractions and expansions of the lung and nodules as seen during normal breathing. The in vivo data in our study had been collected from a small control group of four subjects and a larger test group of 27 subjects with known ground truth (biopsy diagnosis. The growth rate and diagnostic results for both phantoms and in vivo data confirm the high accuracy of our algorithm.

Methodology for automatic detection of lung nodules in computerized tomography images

Lung cancer is a disease with significant prevalence in several countries around the world. Its difficult treatment and rapid progression make the mortality rates among people affected by this illness to be very high. Aiming to offer a computational alternative for helping in detection of nodules, serving as a second opinion to the specialists, this work proposes a totally automatic methodology based on successive detection refining stages. The automated lung nodules detection scheme consists of six stages: thorax extraction, lung extraction, lung reconstruction, structures extraction, tubular structures elimination, and false positive reduction. In the thorax extraction stage all the artifacts external to the patient's body are discarded. Lung extraction stage is responsible for the identification of the lung parenchyma. The objective of the lung reconstruction stage is to prevent incorrect elimination of portions belonging to the parenchyma. Structures extraction stage comprises the selection of dense structures from inside the lung parenchyma. The next stage, tubular structures elimination eliminates a great part of the pulmonary trees. Finally, the false positive stage selects only structures with great probability to be nodule. Each of the several stages has very specific objectives in detection of particular cases of lung nodules, ensuring good matching rates even in difficult detection situations. We use 33 exams with diversified diagnosis and slices numbers for validating the methodology. We obtained a false positive per exam rate of 0.42 and false negative rate of 0.15. The total classification sensitivity obtained, measured out of the nodule candidates, was 84.84%. The specificity achieved was 96.15% and the total accuracy of the method was 95.21%.

Evaluating variability in tumor measurements from same-day repeat CT scans of patients with non-small cell lung cancer

PURPOSE

To evaluate the variability of tumor unidimensional, bidimensional, and volumetric measurements on same-day repeat computed tomographic (CT) scans in patients with non-small cell lung cancer.

MATERIALS AND METHODS

This HIPAA-compliant study was approved by the institutional review board, with informed patient consent. Thirty-two patients with non-small cell lung cancer, each of whom underwent two CT scans of the chest within 15 minutes by using the same imaging protocol, were included in this study. Three radiologists independently measured the two greatest diameters of each lesion on both scans and, during another session, measured the same tumors on the first scan. In a separate analysis, computer software was applied to assist in the calculation of the two greatest diameters and the volume of each lesion on both scans. Concordance correlation coefficients (CCCs) and Bland-Altman plots were used to assess the agreements between the measurements of the two repeat scans (reproducibility) and between the two repeat readings of the same scan (repeatability).

RESULTS

The reproducibility and repeatability of the three radiologists' measurements were high (all CCCs, >or=0.96). The reproducibility of the computer-aided measurements was even higher (all CCCs, 1.00). The 95% limits of agreements for the computer-aided unidimensional, bidimensional, and volumetric measurements on two repeat scans were (-7.3%, 6.2%), (-17.6%, 19.8%), and (-12.1%, 13.4%), respectively.

CONCLUSION

Chest CT scans are well reproducible. Changes in unidimensional lesion size of 8% or greater exceed the measurement variability of the computer method and can be considered significant when estimating the outcome of therapy in a patient.

Segmentation of lung nodules in computed tomography images using dynamic programming and multidirection fusion techniques

RATIONALE AND OBJECTIVES

The aim of this study was to develop a novel algorithm for segmenting lung nodules on three-dimensional (3D) computed tomographic images to improve the performance of computer-aided diagnosis (CAD) systems.

MATERIALS AND METHODS

The database used in this study consists of two data sets obtained from the Lung Imaging Database Consortium. The first data set, containing 23 nodules (22% irregular nodules, 13% nonsolid nodules, 17% nodules attached to other structures), was used for training. The second data set, containing 64 nodules (37% irregular nodules, 40% nonsolid nodules, 62% nodules attached to other structures), was used for testing. Two key techniques were developed in the segmentation algorithm: (1) a 3D extended dynamic programming model, with a newly defined internal cost function based on the information between adjacent slices, allowing parameters to be adapted to each slice, and (2) a multidirection fusion technique, which makes use of the complementary relationships among different directions to improve the final segmentation accuracy. The performance of this approach was evaluated by the overlap criterion, complemented by the true-positive fraction and the false-positive fraction criteria.

RESULTS

The mean values of the overlap, true-positive fraction, and false-positive fraction for the first data set achieved using the segmentation scheme were 66%, 75%, and 15%, respectively, and the corresponding values for the second data set were 58%, 71%, and 22%, respectively.

CONCLUSION

The experimental results indicate that this segmentation scheme can achieve better performance for nodule segmentation than two existing algorithms reported in the literature. The proposed 3D extended dynamic programming model is an effective way to segment sequential images of lung nodules. The proposed multidirection fusion technique is capable of reducing segmentation errors especially for no-nodule and near-end slices, thus resulting in better overall performance.

An automated CT based lung nodule detection scheme using geometric analysis of signed distance field

The authors present a new computerized scheme to automatically detect lung nodules depicted on computed tomography (CT) images. The procedure is performed in the signed distance field of the CT images. To obtain an accurate signed distance field, CT images are first interpolated linearly along the axial direction to form an isotropic data set. Then a lung segmentation strategy is applied to smooth the lung border aiming to include as many juxtapleural nodules as possible while minimizing over segmentations of the lung regions. Potential nodule regions are then detected by locating local maximas of signed distances in each subvolume with values and the sizes larger than the smallest nodule of interest in the three-dimensional space. Finally, all detected candidates are scored by computing the similarity distance of their medial axis-like shapes obtained through a progressive clustering strategy combined with a marching cube algorithm from a sphere based shape. A free-response receiver operating characteristics curve is computed to assess the scheme performance. A performance test on 52 low-dose CT screening examinations that depict 184 verified lung nodules showed that during the initial stage the scheme achieved an asymptotic maximum sensitivity of 95.1% (175/184) with an average of 1200 suspicious voxels per CT examination. The nine missed nodules included two small solid nodules (with a diameter < or =3.1 mm) and seven nonsolid nodules. The final performance level after the similarity scoring stage was an absolute sensitivity level, namely, including the nine missed during the initial stage, of 81.5% (150/184) with 6.5 false-positive identifications per CT examination. This preliminary study demonstrates the feasibility of applying a simple and robust geometric model using the signed distance field to identify suspicious lung nodules. In the authors' data set the sensitivity of this scheme is not affected by nodule size. In addition to potentially being a stand alone approach, the signed distance field based method can be easily implemented as an initial filtering step in other computer-aided detection schemes.

Adaptive border marching algorithm: automatic lung segmentation on chest CT images

Segmentation of the lungs in chest-computed tomography (CT) is often performed as a preprocessing step in lung imaging. This task is complicated especially in presence of disease. This paper presents a lung segmentation algorithm called adaptive border marching (ABM). Its novelty lies in the fact that it smoothes the lung border in a geometric way and can be used to reliably include juxtapleural nodules while minimizing oversegmentation of adjacent regions such as the abdomen and mediastinum. Our experiments using 20 datasets demonstrate that this computational geometry algorithm can re-include all juxtapleural nodules and achieve an average oversegmentation ratio of 0.43% and an average under-segmentation ratio of 1.63% relative to an expert determined reference standard. The segmentation time of a typical case is under 1min on a typical PC. As compared to other available methods, ABM is more robust, more efficient and more straightforward to implement, and once the chest CT images are input, there is no further interaction needed from users. The clinical impact of this method is in potentially avoiding false negative CAD findings due to juxtapleural nodules and improving volumetry and doubling time accuracy.

Segmentation of pulmonary nodules in thoracic CT scans: a region growing approach

This paper presents an efficient algorithm for segmenting different types of pulmonary nodules including high and low contrast nodules, nodules with vasculature attachment, and nodules in the close vicinity of the lung wall or diaphragm. The algorithm performs an adaptive sphericity oriented contrast region growing on the fuzzy connectivity map of the object of interest. This region growing is operated within a volumetric mask which is created by first applying a local adaptive segmentation algorithm that identifies foreground and background regions within a certain window size. The foreground objects are then filled to remove any holes, and a spatial connectivity map is generated to create a 3-D mask. The mask is then enlarged to contain the background while excluding unwanted foreground regions. Apart from generating a confined search volume, the mask is also used to estimate the parameters for the subsequent region growing, as well as for repositioning the seed point in order to ensure reproducibility. The method was run on 815 pulmonary nodules. By using randomly placed seed points, the approach was shown to be fully reproducible. As for acceptability, the segmentation results were visually inspected by a qualified radiologist to search for any gross miss-segmentation. 84% of the first results of the segmentation were accepted by the radiologist while for the remaining 16% nodules, alternative segmentation solutions that were provided by the method were selected.

Prospective assessment of discontinuation and reinitiation of erlotinib or gefitinib in patients with acquired resistance to erlotinib or gefitinib followed by the addition of everolimus

PURPOSE

Ten percent of U.S. patients with non-small cell lung cancer experience partial radiographic responses to erlotinib or gefitinib. Despite initial regressions, these patients develop acquired resistance to erlotinib or gefitinib. In these patients, we sought to assess changes in tumor metabolism and size after stopping and restarting erlotinib or gefitinib and to determine the effect of adding everolimus.

EXPERIMENTAL DESIGN

Patients with non-small cell lung cancer and acquired resistance to erlotinib or gefitinib were eligible. Patients had 18-fluoro-2-deoxy-d-glucose-positron emission tomography/computed tomography and computed tomography scans at baseline, 3 weeks after stopping erlotinib or gefitinib, and 3 weeks after restarting erlotinib or gefitinib. Three weeks after restarting erlotinib or gefitinib, everolimus was added to treatment.

RESULTS

Ten patients completed all four planned studies. Three weeks after stopping erlotinib or gefitinib, there was a median 18% increase in SUV(max) and 9% increase in tumor diameter. Three weeks after restarting erlotinib or gefitinib, there was a median 4% decrease in SUV(max) and 1% decrease in tumor diameter. No partial responses (0 of 10; 95% confidence interval, 0-31%) were seen with the addition of everolimus to erlotinib or gefitinib.

CONCLUSIONS

In patients who develop acquired resistance, stopping erlotinib or gefitinib results in symptomatic progression, increase in SUV(max), and increase in tumor size. Symptoms improve and SUV(max) decreases after restarting erlotinib or gefitinib, suggesting that some tumor cells remain sensitive to epidermal growth factor receptor blockade. No responses were observed with combined everolimus and erlotinib or gefitinib. We recommend a randomized trial to assess the value of continuing erlotinib or gefitinib after development of acquired resistance.

Computer-assisted lung nodule volumetry from multi-detector row CT: Influence of image reconstruction parameters

PURPOSE

To investigate differences in volumetric measurement of pulmonary nodules caused by changing the reconstruction parameters for multi-detector row CT.

MATERIALS AND METHODS

Thirty-nine pulmonary nodules less than 2 cm in diameter were examined by multi-slice CT. All nodules were solid, and located in the peripheral part of the lungs. The resultant 48 parameters images were reconstructed by changing slice thickness (1.25, 2.5, 3.75, or 5 mm), field of view (FOV: 10, 20, or 30 cm), algorithm (high-spatial frequency algorithm or low-spatial frequency algorithm) and reconstruction interval (reconstruction with 50% overlapping of the reconstructed slices or non-overlapping reconstruction). Volumetric measurements were calculated using commercially available software. The differences between nodule volumes were analyzed by the Kruskal-Wallis test and the Wilcoxon Signed-Ranks test.

RESULTS

The diameter of the nodules was 8.7+/-2.7 mm on average, ranging from 4.3 to 16.4mm. Pulmonary nodule volume did not change significantly with changes in slice thickness or FOV (p>0.05), but was significantly larger with the high-spatial frequency algorithm than the low-spatial frequency algorithm (p<0.05), except for one reconstruction parameter. The volumes determined by non-overlapping reconstruction were significantly larger than those of overlapping reconstruction (p<0.05), except for a 1.25 mm thickness with 10 cm FOV with the high-spatial frequency algorithm, and 5mm thickness. The maximum difference in measured volume was 16% on average between the 1.25 mm slice thickness/10 cm FOV/high-spatial frequency algorithm parameters and overlapping reconstruction.

CONCLUSION

Volumetric measurements of pulmonary nodules differ with changes in the reconstruction parameters, with a tendency toward larger volumes in high-spatial frequency algorithm and non-overlapping reconstruction compared to the low-spatial frequency algorithm and overlapping reconstruction.

Lung cancer: computerized quantification of tumor response--initial results

PURPOSE

To prospectively quantify tumor response or progression in patients with lung cancer by using thin-section computed tomography (CT) and a semiautomated algorithm to calculate tumor volume and other parameter values.

MATERIALS AND METHODS

This HIPAA-compliant study was institutional review board approved; informed patient consent was waived. CT scans of 15 measurable non-small cell lung cancers (in five men and 10 women; mean age, 64 years; range, 38-78 years) before and after gefitinib treatment were analyzed. A semiautomated three-dimensional lung cancer segmentation algorithm was developed and applied to each tumor at baseline and follow-up. The computer calculated the greatest diameter (unidimensional measurement), the product of the greatest diameter and the greatest perpendicular diameter (bidimensional measurement), and the volume of each tumor. Exact McNemar tests were used to analyze differences in the percentage change calculated with different measurement techniques.

RESULTS

The computer accurately segmented 14 of the 15 tumors. One paramediastinal tumor required manual separation from the mediastinum. Eleven (73%) of the 15 patients had an absolute change in tumor volume of at least 20%, compared with one (7%) and four (27%) patients who had similar changes in unscaled unidimensional (P < .01) and bidimensional (P = .04) tumor measurements, respectively. Seven (47%) patients had an absolute change in tumor volume of at least 30%. In contrast, at unscaled analysis, no patients at unidimensional measurement (P = .02) and two (13%) patients at bidimensional measurement (P = .06) had a change of at least 30%.

CONCLUSION

Compared with the unidimensional and bidimensional techniques, semiautomated tumor segmentation enabled the identification of a larger number of patients with absolute changes in tumor volume of at least 20% and 30%.

Effect of blood vessels on measurement of nodule volume in a chest phantom

PURPOSE

To identify, by using a chest phantom, whether vessels that contact lung nodules measuring less than 5 mm in diameter will affect nodule volume assessment.

MATERIALS AND METHODS

Forty synthetic nodules (20 with ground-glass attenuation and 20 with solid attenuation) that measured less than 5 mm in diameter were placed into a chest phantom either adjacent to (n = 30) or isolated from (n = 10) synthetic vessels. Nodules were imaged by using low-dose (20 mAs) and diagnostic (120 mAs) multi-detector row computed tomography (CT). Nodules that were known to lie in direct contact with vessels were confirmed by visual inspection. Nontargeted 1.25 x 1.00-mm sections were analyzed with a three-dimensional computer-assisted method for measuring nodule volume. A mixed-model analysis of variance was used to examine the influence of several factors (eg, the presence of adjacent vessels; tube current-time product; and nodule attenuation, diameter, and location) on measurement error.

RESULTS

The mean absolute error (MAE) for all nodules adjacent to vessels was 2.3 mm(3), which was higher than the MAE for isolated nodules (1.9 mm(3)) (P < .001). This difference proved significant only for diagnostic CT (2.2 mm(3) for nodules adjacent to vessels vs 1.3 mm(3) for nodules isolated from vessels) (P < .05). A larger MAE was noted for nodules with ground-glass attenuation (2.3 mm(3)) versus those with solid attenuation (2.0 mm(3)), for increasing nodule volume (1.66 mm(3) for nodules smaller than 20 mm(3) vs 2.83 mm(3) for nodules larger than 40 mm(3)), and for posterior nodule location (P < .05).

CONCLUSION

The presence of a vessel led to a small yet significant increase in volume error on diagnostic-quality images. This represents less than one-third of the overall error, even for nodules larger than 40 mm(3) or approximately 4 mm in diameter. This increase, however, may be more important for smaller nodules with errors of less than 3 mm(3).

Computer analysis of computed tomography scans of the lung: a survey

Current computed tomography (CT) technology allows for near isotropic, submillimeter resolution acquisition of the complete chest in a single breath hold. These thin-slice chest scans have become indispensable in thoracic radiology, but have also substantially increased the data load for radiologists. Automating the analysis of such data is, therefore, a necessity and this has created a rapidly developing research area in medical imaging. This paper presents a review of the literature on computer analysis of the lungs in CT scans and addresses segmentation of various pulmonary structures, registration of chest scans, and applications aimed at detection, classification and quantification of chest abnormalities. In addition, research trends and challenges are identified and directions for future research are discussed.

Volumetric measurement of synthetic lung nodules with multi-detector row CT: effect of various image reconstruction parameters and segmentation thresholds on measurement accuracy

PURPOSE

To evaluate the effect of various multi-detector row computed tomographic (CT) reconstruction parameters and nodule segmentation thresholds on the accuracy of volumetric measurement of synthetic lung nodules.

MATERIALS AND METHODS

Synthetic lung nodules of four different diameters (3.2, 4.8, 6.4, and 12.7 mm) were scanned with multi-detector row CT. Images were reconstructed at various section thicknesses (0.75, 1.0, 2.0, 3.0, and 5.0 mm), fields of view (30, 20, and 10 cm), and reconstruction intervals (0.5, 1.0, and 2.0 mm). The nodules were segmented from the simulated background lung region by using four segmentation thresholds (-300, -400, -500, and -600 HU), and their volumes were estimated and compared with a reference standard (measurements according to fluid displacement) by computing the absolute percentage error (APE). APE was regressed against nodule size, and multivariate analysis of variance (MANOVA) was performed with APE as the dependent variable and with four within-subject factors (field of view, reconstruction interval, threshold, and section thickness).

RESULTS

The MANOVA demonstrated statistically significant effects for threshold (P = .02), section thickness (P < .01), and interaction of threshold and section thickness (P = .04). The regression of mean APE values on nodule size indicates that APE progressively increases with decreasing synthetic nodule size (R2 = 0.99, P < .01).

CONCLUSION

For accurate measurement of lung nodule volume, it is critical to select a section thickness and/or segmentation threshold appropriate for the size of a nodule.

Pulmonary metastases: effect of CT section thickness on measurement--initial experience

PURPOSE

To assess the effect of commonly used computed tomographic (CT) section thicknesses on metastatic tumor measurements calculated with unidimensional, bidimensional, area, and volumetric methods.

MATERIALS AND METHODS

Analysis and data collection were approved by the Institutional Review Board, with waived informed patient consent. Forty-two pulmonary metastases in 10 patients (three men and seven women; age range, 43-83 years; mean age, 65.4 years) were analyzed on CT scans obtained with 3.75-, 5.0-, and 7.5-mm section thicknesses. The lesions were automatically delineated by using a three-dimensional multicriteria segmentation algorithm. Unidimensional (the largest diameter), bidimensional (the product of the two maximal perpendicular diameters), maximal cross-sectional area, and volumetric measurements were automatically obtained for each pulmonary lesion on each section thickness. Means and variances were calculated, and the differences across the three section thicknesses for each of the four measurements were studied by using linear mixed-effects models. The Levene test was used to study the equality of variances.

RESULTS

Differences in the means for unidimensional, bidimensional, and area measurements were significant between a section thickness of 3.75 and 5.0 mm (unidimensional, P=.05; bidimensional, P=.05; area, P=.01) and 3.75 and 7.5 mm (unidimensional, P=.06; bidimensional, P=.03; area, P=.02), but not 5.0 and 7.5 mm. There was a significant difference in volumetric measurement as section thickness decreased from 7.5 to 5.0 mm (P <.001) and from 7.5 to 3.75 mm (P <.001). Although there was a slight trend for differences in the variances across section thickness for each measurement, none of the differences were significant.

CONCLUSION

Volumetric tumor measurements change with a reduction in section thickness from 7.5 to 5.0 and 3.75 mm. For unidimensional measurement, no change was found when thickness decreased from 7.5 to 5.0 mm.

Aspects of computer-aided detection (CAD) and volumetry of pulmonary nodules using multislice CT

With the superb spatial resolution of modern multislice CT scanners and their ability to complete a thoracic scan within one breath-hold, software algorithms for computer-aided detection (CAD) of pulmonary nodules are now reaching high sensitivity levels at moderate false positive rates. A number of pilot studies have shown that CAD modules can successfully find overlooked pulmonary nodules and serve as a powerful tool for diagnostic quality assurance. Equally important are tools for fast and accurate three-dimensional volume measurement of detected nodules. These allow monitoring of nodule growth between follow-up examinations for differential diagnosis and response to oncological therapy. Owing to decreasing partial volume effect, nodule volumetry is more accurate with high resolution CT data. Several studies have shown the feasibility and robustness of automated matching of corresponding nodule pairs between follow-up examinations. Fast and automated growth rate monitoring with only few reader interactions also adds to diagnostic quality assurance.

Segmentation of nodules on chest computed tomography for growth assessment

Several segmentation methods to evaluate growth of small isolated pulmonary nodules on chest computed tomography (CT) are presented. The segmentation methods are based on adaptively thresholding attenuation levels and use measures of nodule shape. The segmentation methods were first tested on a realistic chest phantom to evaluate their performance with respect to specific nodule characteristics. The segmentation methods were also tested on sequential CT scans of patients. The methods' estimation of nodule growth were compared to the volume change calculated by a chest radiologist. The best method segmented nodules on average 43% smaller or larger than the actual nodule when errors were computed across all nodule variations on the phantom. Some methods achieved smaller errors when examined with respect to certain nodule properties. In particular, on the phantom individual methods segmented solid nodules to within 23% of their actual size and nodules with 60.7 mm3 volumes to within 14%. On the clinical data, none of the methods examined showed a statistically significant difference in growth estimation from the radiologist.

Population screening for lung cancer

The feasibility of diagnosing small stage 1 lung cancers using low-dose chest computed tomography in asymptomatic at-risk individuals has been demonstrated in multiple studies. However, it has yet to be proved that the introduction of a chest computed tomography screening programme would do more good than harm at an acceptable cost.

Characterization of small nodules by automatic segmentation of X-ray computed tomography images

OBJECTIVE

To characterize the ability of an automatic lung nodule segmentation algorithm to measure small nodule dimensions and growth rates.

METHODS

A phantom of 20 sets of 6 balls each (11 different nylon balls and 9 acrylic balls) of 1 to 9.5 mm in diameter, in foam, was imaged using x-ray computed tomography with slice thicknesses of 5, 2.5, and 1.25 mm, pitches of 3 and 6, and standard and lung resolution. Measurements consisted of volume and maximum in-plane cross-sectional areas and their derived maximum and effective diameters. Growth rates were simulated using pairs of groups of balls.

RESULTS

Volume measurements overestimate volume, more so for thicker slices. For the largest balls, the error is 60% for 5-mm slices and 20% for 1.25-mm slices. Effective diameter calculated from volume better approximates actual diameter. For area measurements, errors are 0% to 5% for the largest balls, and the effective and actual diameters are closely matched.

CONCLUSIONS

Below 5 mm in diameter, changes in volume should reach 100% for reliable indication of growth. Above 6 mm, the threshold for detecting change is on the order of 25% growth. Even under ideal conditions, results indicate the need for caution when making a diagnosis of malignancy on the basis of volume change.

Effect of Varying CT Section Width on Volumetric Measurement of Lung Tumors and Application of Compensatory Equations

PURPOSE

To determine how volume measurements of simulated and clinical lung tumors at standard computed tomographic (CT) lung window and level settings vary with section width and to derive and apply compensatory equations.

MATERIALS AND METHODS

Spherical simulated tumors of varying diameters were imaged with varying CT section widths, the images were displayed on a workstation, the cross-sectional area of the tumor on each section was measured by using elliptical and perimeter methods, and the areas were integrated to compute tumor volume. The actual and measured tumor volumes for differing section widths and tumor diameters were compared, and compensatory equations were derived. The equations were applied to contemporaneous chest CT images obtained in patients with stage I lung cancer, and the difference between thick- and thin-section-derived volumes before and after application of the equations was determined.

RESULTS

All simulated tumor volumes were overestimated 11%-278%; overestimation varied directly with section width and inversely with tumor diameter. With both measurement methods, mean thin-section volumes of clinical tumors in 55 patients were significantly smaller (P <.01) than mean thick-section volumes: Mean elliptical measurements were 15,025 mm3 (thin) and 18,037 mm3 (thick), with a 20.0% difference; mean perimeter measurements were 16,164 mm3 (thin) and 20,718 mm3 (thick), with a 22.2% difference. The thin-section-to-thick-section volume difference was larger for the smallest tumors. Thin-section volumes were smaller than thick-section volumes in 53 patients with the elliptical method and in 51 patients with the perimeter method. Applying the equations decreased the difference between thick- and thin-section volumes in 37 (67%) of the 55 patients with the elliptical method and in 41 (74%) patients with the perimeter method. The mean thin-section-to-thick-section volume difference became nonsignificant with the perimeter method but remained significant with the elliptical method.

CONCLUSION

Measured lung tumor volumes vary significantly with varying CT section width; overestimation varies directly with section width and inversely with tumor size. Compensatory equations that are somewhat effective in reducing these effects can be derived.

Three-dimensional segmentation and growth-rate estimation of small pulmonary nodules in helical CT images

Small pulmonary nodules are a common radiographic finding that presents an important diagnostic challenge in contemporary medicine. While pulmonary nodules are the major radiographic indicator of lung cancer, they may also be signs of a variety of benign conditions. Measurement of nodule growth rate over time has been shown to be the most promising tool in distinguishing malignant from nonmalignant pulmonary nodules. In this paper, we describe three-dimensional (3-D) methods for the segmentation, analysis, and characterization of small pulmonary nodules imaged using computed tomography (CT). Methods for the isotropic resampling of anisotropic CT data are discussed. 3-D intensity and morphology-based segmentation algorithms are discussed for several classes of nodules. New models and methods for volumetric growth characterization based on longitudinal CT studies are developed. The results of segmentation and growth characterization methods based on in vivo studies are described. The methods presented are promising in their ability to distinguish malignant from nonmalignant pulmonary nodules and represent the first such system in clinical use.

Lung nodule detection and characterization with multislice CT

The ability to identify and characterize pulmonary nodules has been dramatically increased by the introduction of multislice CT (MSCT) technology. Using high-resolution sections, MSCT allows considerable improvement in assessing nodule morphology, enhancement patterns, and growth. MSCT also has facilitated the development and potential of clinical application of computer-assisted diagnosis.

Medical image segmentation with knowledge-guided robust active contours

Medical image segmentation techniques typically require some form of expert human supervision to provide accurate and consistent identification of anatomic structures of interest. A novel segmentation technique was developed that combines a knowledge-based segmentation system with a sophisticated active contour model. This approach exploits the guidance of a higher-level process to robustly perform the segmentation of various anatomic structures. The user need not provide initial contour placement, and the high-level process carries out the required parameter optimization automatically. Knowledge about the anatomic structures to be segmented is defined statistically in terms of probability density functions of parameters such as location, size, and image intensity (eg, computed tomographic [CT] attenuation value). Preliminary results suggest that the performance of the algorithm at chest and abdominal CT is comparable to that of more traditional segmentation techniques like region growing and morphologic operators. In some cases, the active contour-based technique may outperform standard segmentation methods due to its capacity to fully enforce the available a priori knowledge concerning the anatomic structure of interest. The active contour algorithm is particularly suitable for integration with high-level image understanding frameworks, providing a robust and easily controlled low-level segmentation tool. Further study is required to determine whether the proposed algorithm is indeed capable of providing consistently superior segmentation.

Broadband ultrasound attenuation imaging: algorithm development and clinical assessment of a region growing technique

This paper presents a computerized method for the selection of an irregular region of interest (ROI) in broadband ultrasound attenuation (BUA) images. A region growing algorithm searches an initial region in the posterior part of the calcaneus until the pixel with the lowest attenuation value is found; this is the starting seed. Then, the algorithm evaluates the values of the eight pixels neighbouring the starting seed. Pixels that have the closest value to the starting seed are accepted. This procedure is the first processing level. The procedure is repeated for the group of pixels neighbouring those accepted from the previous processing level. The algorithm ceases when the number of accepted pixels reaches a user-specified number. The clinical part of this study compares measurements of BUA at an automatic ROI implemented on a quantitative ultrasound imaging device, defined as the circular region of lowest attenuation in the posterior part of the calcaneus, and at irregular ROIs of various sizes generated by the algorithm developed in this study. The algorithm was applied to BUA images obtained from 24 post-menopausal women with hip fractures and 26 age-matched healthy female subjects. The use of the irregular ROI with a size of 2400 pixels is proposed because that region yielded better clinical results compared to irregular ROIs with different size and the circular automatic ROI.