Method for segmenting chest CT image data using an anatomical model: preliminary results

A Review of Deep Learning Techniques for Lung Cancer Screening and Diagnosis Based on CT Images

One of the most common and deadly diseases in the world is lung cancer. Only early identification of lung cancer can increase a patient's probability of survival. A frequently used modality for the screening and diagnosis of lung cancer is computed tomography (CT) imaging, which provides a detailed scan of the lung. In line with the advancement of computer-assisted systems, deep learning techniques have been extensively explored to help in interpreting the CT images for lung cancer identification. Hence, the goal of this review is to provide a detailed review of the deep learning techniques that were developed for screening and diagnosing lung cancer. This review covers an overview of deep learning (DL) techniques, the suggested DL techniques for lung cancer applications, and the novelties of the reviewed methods. This review focuses on two main methodologies of deep learning in screening and diagnosing lung cancer, which are classification and segmentation methodologies. The advantages and shortcomings of current deep learning models will also be discussed. The resultant analysis demonstrates that there is a significant potential for deep learning methods to provide precise and effective computer-assisted lung cancer screening and diagnosis using CT scans. At the end of this review, a list of potential future works regarding improving the application of deep learning is provided to spearhead the advancement of computer-assisted lung cancer diagnosis systems.

Automated Endotracheal Tube Placement Check Using Semantically Embedded Deep Neural Networks

RATIONALE AND OBJECTIVES

To develop artificial intelligence (AI) system that assists in checking endotracheal tube (ETT) placement on chest X-rays (CXRs) and evaluate whether it can move into clinical validation as a quality improvement tool.

MATERIALS AND METHODS

A retrospective data set including 2000 de-identified images from intensive care unit patients was split into 1488 for training and 512 for testing. AI was developed to automatically identify the ETT, trachea, and carina using semantically embedded neural networks that combine a declarative knowledge base with deep neural networks. To check the ETT tip placement, a "safe zone" was computed as the region inside the trachea and 3-7 cm above the carina. Two AI outputs were evaluated: (1) ETT overlay, (2) ETT misplacement alert messages. Clinically relevant performance metrics were compared against prespecified thresholds of >85% overlay accuracy and positive predictive value (PPV) > 30% and negative predictive value NPV > 95% for alerts to move into clinical validation.

RESULTS

An ETT was present in 285 of 512 test cases. The AI detected 95% (271/285) of ETTs, 233 (86%) of these with accurate tip localization. The system (correctly) did not generate an ETT overlay in 221/227 CXRs where the tube was absent for an overall overlay accuracy of 89% (454/512). The alert messages indicating that either the ETT was misplaced or not detected had a PPV of 83% (265/320) and NPV of 98% (188/192).

CONCLUSION

The chest X-ray AI met prespecified performance thresholds to move into clinical validation.

Pulmonary nodules detection assistant platform: An effective computer aided system for early pulmonary nodules detection in physical examination

BACKGROUND AND OBJECTIVE

Early detection of the pulmonary nodule from physical examination low-dose computer tomography (LDCT) images is an effective measure to reduce the mortality rate of lung cancer. Although there are many computer aided diagnosis (CAD) methods used for detecting pulmonary nodules, there are few CAD systems for small pulmonary nodule detection with a large amount of physical examination LDCT images.

METHODS

In this work, we designed a CAD system called Pulmonary Nodules Detection Assistant Platform for early pulmonary nodules detection and classification based on the physical examination LDCT images. Based on the preprocessed physical examination CT images, the three-dimensional (3D) CNN-based model is presented to detect candidate pulmonary nodules and output detection results with quantitative parameters, the 3D ResNet is used to classify the detected nodules into intrapulmonary nodules and pleural nodules to reduce the physician workloads, and the Fully Connected Neural Network (FCNN) is used to classify ground-glass opacity (GGO) nodules and non-GGO nodules to help doctor pay more attention to those suspected early lung cancer nodules.

RESULTS

Experiments are performed on our 1000 samples of physical examinations (LNPE1000) with an average diameter of 5.3 mm and LUNA16 dataset with an average diameter of 8.31 mm, which show that the designed CAD system is automatic and efficient for detecting smaller and larger nodules from different datasets, especially for the detection of smaller nodules with diameter between 3 mm and 6 mm in physical examinations. The accuracy of pulmonary nodule detection reaches 0.879 with an average of 1 false positive per CT in LNPE1000 dataset, which is comparable to the experienced physicians. The classification accuracy reaches 0.911 between intrapulmonary and pleural nodules, and 0.950 between GGO and non-GGO nodules, respectively.

CONCLUSION

Experimental results show that the proposed pulmonary nodule detection model is robust for different datasets, which can successfully detect smaller and larger nodules in CT images obtained by physical examination. The interactive platform of the designed CAD system has been on trial in a hospital by combining with manual reading, which helps doctors analyze clinical data dynamically and improves the nodule detection efficiency in physical examination applications.

Deep Residual Separable Convolutional Neural Network for lung tumor segmentation

Lung cancer is one of the deadliest types of cancers. Computed Tomography (CT) is a widely used technique to detect tumors present inside the lungs. Delineation of such tumors is particularly essential for analysis and treatment purposes. With the advancement in hardware technologies, Machine Learning and Deep Learning methods are outperforming the traditional methods in the field of medical imaging. In order to delineate lung cancer tumors, we have proposed a deep learning-based methodology which includes a maximum intensity projection based pre-processing method, two novel deep learning networks and an ensemble strategy. The two proposed networks named Deep Residual Separable Convolutional Neural Network 1 and 2 (DRS-CNN1 and DRS-CNN2) achieved better performance over the state-of-the-art U-net network and other segmentation networks. For fair comparison, we have evaluated the performances of all networks on Medical Segmentation Decathlon (MSD) and StructSeg 2019 datasets. The DRS-CNN2 achieved a mean Dice Similarity Coefficient (DSC) of 0.649, mean 95 Hausdorff Distance (HD95) of 18.26, mean Sensitivity 0.737 and a mean Precision of 0.765 on independent test sets.

Precise Segmentation of COVID-19 Infected Lung from CT Images Based on Adaptive First-Order Appearance Model with Morphological/Anatomical Constraints

A new segmentation technique is introduced for delineating the lung region in 3D computed tomography (CT) images. To accurately model the distribution of Hounsfield scale values within both chest and lung regions, a new probabilistic model is developed that depends on a linear combination of Gaussian (LCG). Moreover, we modified the conventional expectation-maximization (EM) algorithm to be run in a sequential way to estimate both the dominant Gaussian components (one for the lung region and one for the chest region) and the subdominant Gaussian components, which are used to refine the final estimated joint density. To estimate the marginal density from the mixed density, a modified k-means clustering approach is employed to classify the Gaussian subdominant components to determine which components belong properly to a lung and which components belong to a chest. The initial segmentation, based on the LCG-model, is then refined by the imposition of 3D morphological constraints based on a 3D Markov-Gibbs random field (MGRF) with analytically estimated potentials. The proposed approach was tested on CT data from 32 coronavirus disease 2019 (COVID-19) patients. Segmentation quality was quantitatively evaluated using four metrics: Dice similarity coefficient (DSC), overlap coefficient, 95th-percentile bidirectional Hausdorff distance (BHD), and absolute lung volume difference (ALVD), and it achieved 95.67±1.83%, 91.76±3.29%, 4.86±5.01, and 2.93±2.39, respectively. The reported results showed the capability of the proposed approach to accurately segment healthy lung tissues in addition to pathological lung tissues caused by COVID-19, outperforming four current, state-of-the-art deep learning-based lung segmentation approaches.

Multi-resolution convolutional neural networks for fully automated segmentation of acutely injured lungs in multiple species

Segmentation of lungs with acute respiratory distress syndrome (ARDS) is a challenging task due to diffuse opacification in dependent regions which results in little to no contrast at the lung boundary. For segmentation of severely injured lungs, local intensity and texture information, as well as global contextual information, are important factors for consistent inclusion of intrapulmonary structures. In this study, we propose a deep learning framework which uses a novel multi-resolution convolutional neural network (ConvNet) for automated segmentation of lungs in multiple mammalian species with injury models similar to ARDS. The multi-resolution model eliminates the need to tradeoff between high-resolution and global context by using a cascade of low-resolution to high-resolution networks. Transfer learning is used to accommodate the limited number of training datasets. The model was initially pre-trained on human CT images, and subsequently fine-tuned on canine, porcine, and ovine CT images with lung injuries similar to ARDS. The multi-resolution model was compared to both high-resolution and low-resolution networks alone. The multi-resolution model outperformed both the low- and high-resolution models, achieving an overall mean Jacaard index of 0.963 ± 0.025 compared to 0.919 ± 0.027 and 0.950 ± 0.036, respectively, for the animal dataset (N=287). The multi-resolution model achieves an overall average symmetric surface distance of 0.438 ± 0.315 mm, compared to 0.971 ± 0.368 mm and 0.657 ± 0.519 mm for the low-resolution and high-resolution models, respectively. We conclude that the multi-resolution model produces accurate segmentations in severely injured lungs, which is attributed to the inclusion of both local and global features.

An unsupervised semi-automated pulmonary nodule segmentation method based on enhanced region growing

Background

Nowadays, computer technology is getting popular for clinical aided diagnosis, especially in the direction of medical images. It makes physician diagnosis of lung nodules more efficient by providing them with reliable and accurate segmentation.

Methods

A region growing based semi-automated pulmonary nodule segmentation algorithm (ReGANS) was developed with three improvements: an automatic threshold calculation method, a lesion area pre-projection method, and an optimized region growing method. The algorithm can quickly and accurately segment a whole lung nodule in a set of computed tomography (CT) images based on an initial manual point.

Results

The average time taken for ReGANS to segment 1 pulmonary nodule was 0.83s, and the probability rand index (PRI), global consistency error (GCE), and variation of information (VoI) from a comparison between the algorithm and the radiologist's 2 manual results were 0.93, 0.06, and 0.3 for the boundary range (BR), and 0.86, 0.06, 0.3 for the precise range (PR). The number of images covered by one pulmonary nodule in a CT image set was also evaluated to compare the segmentation algorithm with the radiologist's results, with an error rate of 15%. At the same time, the results were verified in multiple data sets to validate the robustness.

Conclusions

Compared with other algorithms, ReGANS can segment the lung nodule image region more quickly and more precisely. The experimental results show that ReGANS can assist medical imaging diagnosis and has good clinical application value. It also provides a faster and more convenient method for pre-data preparation of intelligent algorithms.

Automatic Lung Segmentation With Juxta-Pleural Nodule Identification Using Active Contour Model and Bayesian Approach

OBJECTIVE

chest computed tomography (CT) images and their quantitative analyses have become increasingly important for a variety of purposes, including lung parenchyma density analysis, airway analysis, diaphragm mechanics analysis, and nodule detection for cancer screening. Lung segmentation is an important prerequisite step for automatic image analysis. We propose a novel lung segmentation method to minimize the juxta-pleural nodule issue, a notorious challenge in the applications.

METHOD

we initially used the Chan-Vese (CV) model for active lung contours and adopted a Bayesian approach based on the CV model results, which predicts the lung image based on the segmented lung contour in the previous frame image or neighboring upper frame image. Among the resultant juxta-pleural nodule candidates, false positives were eliminated through concave points detection and circle/ellipse Hough transform. Finally, the lung contour was modified by adding the final nodule candidates to the area of the CV model results.

RESULTS

to evaluate the proposed method, we collected chest CT digital imaging and communications in medicine images of 84 anonymous subjects, including 42 subjects with juxta-pleural nodules. There were 16 873 images in total. Among the images, 314 included juxta-pleural nodules. Our method exhibited a disc similarity coefficient of 0.9809, modified hausdorff distance of 0.4806, sensitivity of 0.9785, specificity of 0.9981, accuracy of 0.9964, and juxta-pleural nodule detection rate of 96%. It outperformed existing methods, such as the CV model used alone, the normalized CV model, and the snake algorithm. Clinical impact: the high accuracy with the juxta-pleural nodule detection in the lung segmentation can be beneficial for any computer aided diagnosis system that uses lung segmentation as an initial step.

An Architecture for Computer-Aided Detection and Radiologic Measurement of Lung Nodules in Clinical Trials

Computer tomography (CT) imaging plays an important role in cancer detection and quantitative assessment in clinical trials. High-resolution imaging studies on large cohorts of patients generate vast data sets, which are infeasible to analyze through manual interpretation.

In this article we describe a comprehensive architecture for computer-aided detection (CAD) and surveillance on lung nodules in CT images. Central to this architecture are the analytic components: an automated nodule detection system, nodule tracking capabilities and volume measurement, which are integrated within a data management system that includes mechanisms for receiving and archiving images, a database for storing quantitative nodule measurements and visualization, and reporting tools.

We describe two studies to evaluate CAD technology within this architecture, and the potential application in large clinical trials. The first study involves performance assessment of an automated nodule detection system and its ability to increase radiologist sensitivity when used to provide a second opinion. The second study investigates nodule volume measurements on CT made using a semi-automated technique and shows that volumetric analysis yields significantly different tumor response classifications than a 2D diameter approach. These studies demonstrate the potential of automated CAD tools to assist in quantitative image analysis for clinical trials.

Accurate Lungs Segmentation on CT Chest Images by Adaptive Appearance-Guided Shape Modeling

To accurately segment pathological and healthy lungs for reliable computer-aided disease diagnostics, a stack of chest CT scans is modeled as a sample of a spatially inhomogeneous joint 3D Markov-Gibbs random field (MGRF) of voxel-wise lung and chest CT image signals (intensities). The proposed learnable MGRF integrates two visual appearance sub-models with an adaptive lung shape submodel. The first-order appearance submodel accounts for both the original CT image and its Gaussian scale space (GSS) filtered version to specify local and global signal properties, respectively. Each empirical marginal probability distribution of signals is closely approximated with a linear combination of discrete Gaussians (LCDG), containing two positive dominant and multiple sign-alternate subordinate DGs. The approximation is separated into two LCDGs to describe individually the lungs and their background, i.e., all other chest tissues. The second-order appearance submodel quantifies conditional pairwise intensity dependencies in the nearest voxel 26-neighborhood in both the original and GSS-filtered images. The shape submodel is built for a set of training data and is adapted during segmentation using both the lung and chest appearances. The accuracy of the proposed segmentation framework is quantitatively assessed using two public databases (ISBI VESSEL12 challenge and MICCAI LOLA11 challenge) and our own database with, respectively, 20, 55, and 30 CT images of various lung pathologies acquired with different scanners and protocols. Quantitative assessment of our framework in terms of Dice similarity coefficients, 95-percentile bidirectional Hausdorff distances, and percentage volume differences confirms the high accuracy of our model on both our database (98.4_±1.0%, 2.2_±1.0 mm, 0.42_±0.10%) and the VESSEL12 database (99.0_±0.5%, 2.1_±1.6 mm, 0.39_±0.20%), respectively. Similarly, the accuracy of our approach is further verified via a blind evaluation by the organizers of the LOLA11 competition, where an average overlap of 98.0% with the expert's segmentation is yielded on all 55 subjects with our framework being ranked first among all the state-of-the-art techniques compared.

A Rapid Segmentation-Insensitive "Digital Biopsy" Method for Radiomic Feature Extraction: Method and Pilot Study Using CT Images of Non-Small Cell Lung Cancer

Quantitative imaging approaches compute features within images' regions of interest. Segmentation is rarely completely automatic, requiring time-consuming editing by experts. We propose a new paradigm, called “digital biopsy,” that allows for the collection of intensity- and texture-based features from these regions at least 1 order of magnitude faster than the current manual or semiautomated methods. A radiologist reviewed automated segmentations of lung nodules from 100 preoperative volume computed tomography scans of patients with non–small cell lung cancer, and manually adjusted the nodule boundaries in each section, to be used as a reference standard, requiring up to 45 minutes per nodule. We also asked a different expert to generate a digital biopsy for each patient using a paintbrush tool to paint a contiguous region of each tumor over multiple cross-sections, a procedure that required an average of <3 minutes per nodule. We simulated additional digital biopsies using morphological procedures. Finally, we compared the features extracted from these digital biopsies with our reference standard using intraclass correlation coefficient (ICC) to characterize robustness. Comparing the reference standard segmentations to our digital biopsies, we found that 84/94 features had an ICC >0.7; comparing erosions and dilations, using a sphere of 1.5-mm radius, of our digital biopsies to the reference standard segmentations resulted in 41/94 and 53/94 features, respectively, with ICCs >0.7. We conclude that many intensity- and texture-based features remain consistent between the reference standard and our method while substantially reducing the amount of operator time required.

Automatic Approach for Lung Segmentation with Juxta-Pleural Nodules from Thoracic CT Based on Contour Tracing and Correction

This paper presents a fully automatic framework for lung segmentation, in which juxta-pleural nodule problem is brought into strong focus. The proposed scheme consists of three phases: skin boundary detection, rough segmentation of lung contour, and pulmonary parenchyma refinement. Firstly, chest skin boundary is extracted through image aligning, morphology operation, and connective region analysis. Secondly, diagonal-based border tracing is implemented for lung contour segmentation, with maximum cost path algorithm used for separating the left and right lungs. Finally, by arc-based border smoothing and concave-based border correction, the refined pulmonary parenchyma is obtained. The proposed scheme is evaluated on 45 volumes of chest scans, with volume difference (VD) 11.15 ± 69.63 cm³, volume overlap error (VOE) 3.5057 ± 1.3719%, average surface distance (ASD) 0.7917 ± 0.2741 mm, root mean square distance (RMSD) 1.6957 ± 0.6568 mm, maximum symmetric absolute surface distance (MSD) 21.3430 ± 8.1743 mm, and average time-cost 2 seconds per image. The preliminary results on accuracy and complexity prove that our scheme is a promising tool for lung segmentation with juxta-pleural nodules.

Segmentation and Image Analysis of Abnormal Lungs at CT: Current Approaches, Challenges, and Future Trends

The computer-based process of identifying the boundaries of lung from surrounding thoracic tissue on computed tomographic (CT) images, which is called segmentation, is a vital first step in radiologic pulmonary image analysis. Many algorithms and software platforms provide image segmentation routines for quantification of lung abnormalities; however, nearly all of the current image segmentation approaches apply well only if the lungs exhibit minimal or no pathologic conditions. When moderate to high amounts of disease or abnormalities with a challenging shape or appearance exist in the lungs, computer-aided detection systems may be highly likely to fail to depict those abnormal regions because of inaccurate segmentation methods. In particular, abnormalities such as pleural effusions, consolidations, and masses often cause inaccurate lung segmentation, which greatly limits the use of image processing methods in clinical and research contexts. In this review, a critical summary of the current methods for lung segmentation on CT images is provided, with special emphasis on the accuracy and performance of the methods in cases with abnormalities and cases with exemplary pathologic findings. The currently available segmentation methods can be divided into five major classes: (a) thresholding-based, (b) region-based, (c) shape-based, (d) neighboring anatomy-guided, and (e) machine learning-based methods. The feasibility of each class and its shortcomings are explained and illustrated with the most common lung abnormalities observed on CT images. In an overview, practical applications and evolving technologies combining the presented approaches for the practicing radiologist are detailed.

Supervised recursive segmentation of volumetric CT images for 3D reconstruction of lung and vessel tree

Three dimensional reconstruction of lung and vessel tree has great significance to 3D observation and quantitative analysis for lung diseases. This paper presents non-sheltered 3D models of lung and vessel tree based on a supervised semi-3D lung tissues segmentation method. A recursive strategy based on geometric active contour is proposed instead of the "coarse-to-fine" framework in existing literature to extract lung tissues from the volumetric CT slices. In this model, the segmentation of the current slice is supervised by the result of the previous one slice due to the slight changes between adjacent slice of lung tissues. Through this mechanism, lung tissues in all the slices are segmented fast and accurately. The serious problems of left and right lungs fusion, caused by partial volume effects, and segmentation of pleural nodules can be settled meanwhile during the semi-3D process. The proposed scheme is evaluated by fifteen scans, from eight healthy participants and seven participants suffering from early-stage lung tumors. The results validate the good performance of the proposed method compared with the "coarse-to-fine" framework. The segmented datasets are utilized to reconstruct the non-sheltered 3D models of lung and vessel tree.

Accurate and efficient separation of left and right lungs from 3D CT scans: A generic hysteresis approach

Separation of left and right lungs from binary segmentation is often necessary for quantitative image-based pulmonary disease evaluation. In this article, we present a new fully automated approach for accurate, robust, and efficient lung separation using 3-D CT scans. Our method follows a hysteresis setting that utilizes information from both lung regions and background gaps. First, original segmentation is separated by subtracting the gaps between left and right lungs, which are enhanced with Hessian filtering. Second, the 2-D separation manifold in 3-D image space is estimated based on the distance information from the two subsets. Finally, the separation manifold is projected back to the original segmentation in order to produce the separated lungs through optimization for addressing minor local variations. An evaluation on over 400 human and 100 small animal 3-D CT images with various abnormalities is performed. The proposed scheme successfully separated all connections on the candidate CT images. Using hysteresis mechanism, each phase is performed robustly and 3-D information is utilized to achieve a generic, efficient, and accurate solution.

Thoracic cavity definition for 3D PET/CT analysis and visualization

X-ray computed tomography (CT) and positron emission tomography (PET) serve as the standard imaging modalities for lung-cancer management. CT gives anatomical details on diagnostic regions of interest (ROIs), while PET gives highly specific functional information. During the lung-cancer management process, a patient receives a co-registered whole-body PET/CT scan pair and a dedicated high-resolution chest CT scan. With these data, multimodal PET/CT ROI information can be gleaned to facilitate disease management. Effective image segmentation of the thoracic cavity, however, is needed to focus attention on the central chest. We present an automatic method for thoracic cavity segmentation from 3D CT scans. We then demonstrate how the method facilitates 3D ROI localization and visualization in patient multimodal imaging studies. Our segmentation method draws upon digital topological and morphological operations, active-contour analysis, and key organ landmarks. Using a large patient database, the method showed high agreement to ground-truth regions, with a mean coverage=99.2% and leakage=0.52%. Furthermore, it enabled extremely fast computation. For PET/CT lesion analysis, the segmentation method reduced ROI search space by 97.7% for a whole-body scan, or nearly 3 times greater than that achieved by a lung mask. Despite this reduction, we achieved 100% true-positive ROI detection, while also reducing the false-positive (FP) detection rate by >5 times over that achieved with a lung mask. Finally, the method greatly improved PET/CT visualization by eliminating false PET-avid obscurations arising from the heart, bones, and liver. In particular, PET MIP views and fused PET/CT renderings depicted unprecedented clarity of the lesions and neighboring anatomical structures truly relevant to lung-cancer assessment.

Automatic left and right lung separation using free-formed surface fitting on volumetric CT

This study presents a completely automated method for separating the left and right lungs using free-formed surface fitting on volumetric computed tomography (CT). The left and right lungs are roughly divided using iterative 3-dimensional morphological operator and a Hessian matrix analysis. A point set traversing between the initial left and right lungs is then detected with a Euclidean distance transform to determine the optimal separating surface, which is then modeled from the point set using a free-formed surface-fitting algorithm. Subsequently, the left and right lung volumes are smoothly and directly separated using the separating surface. The performance of the proposed method was estimated by comparison with that of a human expert on 44 CT examinations. For all data sets, averages of the root mean square surface distance, maximum surface distance, and volumetric overlap error between the results of the automatic and the manual methods were 0.032 mm, 2.418 mm, and 0.017 %, respectively. Our study showed the feasibility of automatically separating the left and right lungs by identifying the 3D continuous separating surface on volumetric chest CT images.

Review of automatic pulmonary lobe segmentation methods from CT

The computational detection of pulmonary lobes from CT images is a challenging segmentation problem with important respiratory health care applications, including surgical planning and regional image analysis. Several authors have proposed automated algorithms and we present a methodological review. These algorithms share a number of common stages and we consider each stage in turn, comparing the methods applied by each author and discussing their relative strengths. No standard method has yet emerged and none of the published methods have been demonstrated across a full range of clinical pathologies and imaging protocols. We discuss how improved methods could be developed by combining different approaches, and we use this to propose a workflow for the development of new algorithms.

An automated lung segmentation approach using bidirectional chain codes to improve nodule detection accuracy

Computer-aided detection and diagnosis (CAD) has been widely investigated to improve radiologists׳ diagnostic accuracy in detecting and characterizing lung disease, as well as to assist with the processing of increasingly sizable volumes of imaging. Lung segmentation is a requisite preprocessing step for most CAD schemes. This paper proposes a parameter-free lung segmentation algorithm with the aim of improving lung nodule detection accuracy, focusing on juxtapleural nodules. A bidirectional chain coding method combined with a support vector machine (SVM) classifier is used to selectively smooth the lung border while minimizing the over-segmentation of adjacent regions. This automated method was tested on 233 computed tomography (CT) studies from the lung imaging database consortium (LIDC), representing 403 juxtapleural nodules. The approach obtained a 92.6% re-inclusion rate. Segmentation accuracy was further validated on 10 randomly selected CT series, finding a 0.3% average over-segmentation ratio and 2.4% under-segmentation rate when compared to manually segmented reference standards done by an expert.

A generic approach to pathological lung segmentation

In this study, we propose a novel pathological lung segmentation method that takes into account neighbor prior constraints and a novel pathology recognition system. Our proposed framework has two stages; during stage one, we adapted the fuzzy connectedness (FC) image segmentation algorithm to perform initial lung parenchyma extraction. In parallel, we estimate the lung volume using rib-cage information without explicitly delineating lungs. This rudimentary, but intelligent lung volume estimation system allows comparison of volume differences between rib cage and FC based lung volume measurements. Significant volume difference indicates the presence of pathology, which invokes the second stage of the proposed framework for the refinement of segmented lung. In stage two, texture-based features are utilized to detect abnormal imaging patterns (consolidations, ground glass, interstitial thickening, tree-inbud, honeycombing, nodules, and micro-nodules) that might have been missed during the first stage of the algorithm. This refinement stage is further completed by a novel neighboring anatomy-guided segmentation approach to include abnormalities with weak textures, and pleura regions. We evaluated the accuracy and efficiency of the proposed method on more than 400 CT scans with the presence of a wide spectrum of abnormalities. To our best of knowledge, this is the first study to evaluate all abnormal imaging patterns in a single segmentation framework. The quantitative results show that our pathological lung segmentation method improves on current standards because of its high sensitivity and specificity and may have considerable potential to enhance the performance of routine clinical tasks.

Automated segmentation of pulmonary structures in thoracic computed tomography scans: a review

Computed tomography (CT) is the modality of choice for imaging the lungs in vivo. Sub-millimeter isotropic images of the lungs can be obtained within seconds, allowing the detection of small lesions and detailed analysis of disease processes. The high resolution of thoracic CT and the high prevalence of lung diseases require a high degree of automation in the analysis pipeline. The automated segmentation of pulmonary structures in thoracic CT has been an important research topic for over a decade now. This systematic review provides an overview of current literature. We discuss segmentation methods for the lungs, the pulmonary vasculature, the airways, including airway tree construction and airway wall segmentation, the fissures, the lobes and the pulmonary segments. For each topic, the current state of the art is summarized, and topics for future research are identified.

Contour propagation using feature-based deformable registration for lung cancer

Accurate target delineation of CT image is a critical step in radiotherapy treatment planning. This paper describes a novel strategy for automatic contour propagation, based on deformable registration, for CT images of lung cancer. The proposed strategy starts with a manual-delineated contour in one slice of a 3D CT image. By means of feature-based deformable registration, the initial contour in other slices of the image can be propagated automatically, and then refined by active contour approach. Three algorithms are employed in the strategy: the Speeded-Up Robust Features (SURF), Thin-Plate Spline (TPS), and an adapted active contour (Snake), used to refine and modify the initial contours. Five pulmonary cancer cases with about 400 slices and 1000 contours have been used to verify the proposed strategy. Experiments demonstrate that the proposed strategy can improve the segmentation performance in the pulmonary CT images. Jaccard similarity (JS) mean is about 0.88 and the maximum of Hausdorff distance (HD) is about 90%. In addition, delineation time has been considerably reduced. The proposed feature-based deformable registration method in the automatic contour propagation improves the delineation efficiency significantly.

Automated pulmonary nodule detection based on three-dimensional shape-based feature descriptor

Computer-aided detection (CAD) can help radiologists to detect pulmonary nodules at an early stage. In pulmonary nodule CAD systems, feature extraction is very important for describing the characteristics of nodule candidates. In this paper, we propose a novel three-dimensional shape-based feature descriptor to detect pulmonary nodules in CT scans. After lung volume segmentation, nodule candidates are detected using multi-scale dot enhancement filtering in the segmented lung volume. Next, we extract feature descriptors from the detected nodule candidates, and these are refined using an iterative wall elimination method. Finally, a support vector machine-based classifier is trained to classify nodules and non-nodules. The performance of the proposed system is evaluated on Lung Image Database Consortium data. The proposed method significantly reduces the number of false positives in nodule candidates. This method achieves 97.5% sensitivity, with only 6.76 false positives per scan.

Illustration of the obstacles in computerized lung segmentation using examples

PURPOSE

Automated lung volume segmentation is often a preprocessing step in quantitative lung computed tomography (CT) image analysis. The objective of this study is to identify the obstacles in computerized lung volume segmentation and illustrate those explicitly using real examples. Awareness of these "difficult" cases may be helpful for the development of a robust and consistent lung segmentation algorithm.

METHODS

We collected a large diverse dataset consisting of 2768 chest CT examinations acquired on 2292 subjects from various sources. These examinations cover a wide range of diseases, including lung cancer, chronic obstructive pulmonary disease, human immunodeficiency virus, pulmonary embolism, pneumonia, asthma, and interstitial lung disease (ILD). The CT acquisition protocols, including dose, scanners, and reconstruction kernels, vary significantly. After the application of a "neutral" thresholding-based approach to the collected CT examinations in a batch manner, the failed cases were subjectively identified and classified into different subgroups.

RESULTS

Totally, 121 failed examinations are identified, corresponding to a failure ratio of 4.4%. These failed cases are summarized as 11 different subgroups, which is further classified into 3 broad categories: (1) failure caused by diseases, (2) failure caused by anatomy variability, and (3) failure caused by external factors. The failure percentages in these categories are 62.0%, 32.2%, and 5.8%, respectively.

CONCLUSIONS

The presence of specific lung diseases (e.g., pulmonary nodules, ILD, and pneumonia) is the primary issue in computerized lung segmentation. The segmentation failures caused by external factors and anatomy variety are relatively low but unavoidable in practice. It is desirable to develop robust schemes to handle these issues in a single pass when a large number of CT examinations need to be analyzed.

Potential of computer-aided diagnosis to improve CT lung cancer screening

The development of low-dose spiral computed tomography (CT) has rekindled hope that effective lung cancer screening might yet be found. Screening is justified when there is evidence that it will extend lives at reasonable cost and acceptable levels of risk. A screening test should detect all extant cancers while avoiding unnecessary workups. Thus optimal screening modalities have both high sensitivity and specificity. Due to the present state of technology, radiologists must opt to increase sensitivity and rely on follow-up diagnostic procedures to rule out the incurred false positives. There is evidence in published reports that computer-aided diagnosis technology may help radiologists alter the benefit-cost calculus of CT sensitivity and specificity in lung cancer screening protocols. This review will provide insight into the current discussion of the effectiveness of lung cancer screening and assesses the potential of state-of-the-art computer-aided design developments.

A novel segmentation approach for improving diagnostic accuracy of CAD systems for detecting lung cancer from chest computed tomography images

Segmentation of lung tissue is an important and challenging task in any computer aided diagnosis system. The accuracy of the segmentation subsystem determines the performance of the other subsystems in any computer aided diagnosis system based on image analysis. We propose a novel technique for segmentation of lung tissue from computed tomography of the chest. Manual segmentation of lung parenchyma becomes difficult with an enormous volume of images. The goal of this work is to present an automated approach to segmentation of lung parenchyma from the rest of the chest CT image. The approach involves the conventional optimal thresholding technique and operations based on convex edge and centroid properties of the lung region. The segmentation technique proposed in this article can be used to preprocess lung images given to a computer aided diagnosis system for diagnosis of lung disorders. This improves the diagnostic performance of the system. This has been tested by using it in a computer aided diagnosis system that was used for detection of lung cancer from chest computed tomography images. The results obtained show that the lungs can be correctly segmented even in the presence of peripheral pathology bearing regions; pathology bearing regions that could not be detected using a CAD system that applies optimal thresholding could be detected using a CAD system using out proposed approach for segmentation of lungs.

GROUND-GLASS OPACITY DETECTION BY USING CORRELATION BETWEEN SUCCESSIVE SLICE IMAGES

Medical imaging systems such as computed tomography, magnetic resonance imaging provided a high resolution image for powerful diagnostic tool in visual inspection fields by physician. Especially MDCT image can be used to obtain detailed images of the pulmonary anatomy, including pulmonary diseases such as the pulmonary nodules, the pulmonary vein, etc. In the medical image processing technique, segmentation is a difficult task because surrounding soft tissues and organs have similar CT values and sometimes contact with each other. We propose a new technique for automatic segmentation of lung regions and its classification for ground-glass opacity from the extracted lung regions by computer based on a set of the thorax CT images. In this paper, we segment the lung region for extraction of the region of interest employing binarization and labeling process from the inputted each slices images. The region having the largest area is regarded as the tentative lung regions. Furthermore, the ground-glass opacity is classified by correlation distribution on the slice to slice from the extracted lung region with respect to the thorax CT images. Experiment is performed employing twenty six thorax CT image sets and 96% of recognition rates were achieved. Obtained results are shown along with a discussion.

A fast and accurate automatic lung segmentation and volumetry method for MR data used in epidemiological studies

In modern epidemiological population-based studies a huge amount of magnetic resonance imaging (MRI) data is analysed. This requires reliable automatic methods for organ extraction. In the current paper, we propose a fast and accurate automatic method for lung segmentation and volumetry. Our approach follows a "coarse-to-fine" segmentation strategy. First, we extract the lungs and trachea excluding the main pulmonary vessels. This step is executed very fast and allows for measuring the volume of both structures. Thereafter, we start a refinement procedure that consists of three main stages: trachea extraction, lung separation, and filling the cavities on the final lung masks. After the trachea extraction step the volumes of both lungs without the main vessels can be measured. The final segmentation step results in the volumes of the left and right lungs including the vessels. The method has been tested by processing MR datasets from ten healthy participants. We compare our results with manually produced masks and obtain high agreement between the expert reading and our method: the True Positive Volume Fraction is more than 95%. The proposed automatic approach is fast and accurate enough to be applied in clinical routine for processing of thousands of participants.

Reproducibility of volume and densitometric measures of emphysema on repeat computed tomography with an interval of 1 week

OBJECTIVES

The reproducibilities of CT lung volume and densitometric measures of emphysema were assessed over 1 week. The influence of breathhold on reproducibility was assessed.

METHODS

HRCT was performed on 44 subjects at inspiration on two visits with a 7-day interval. CT lung volume, relative area below -950HU (RA950-raw), and 15th percentile density (PD15-raw) were computed. Volume correction was used to obtain RA950-adj and PD15-adj. Reproducibilities between visits were assessed using concordance correlation coefficient (CCC) and repeatability coefficient (RC). Reproducibilities were compared between raw and adjusted measures. Differences between visits were computed for volume and density measures. Correlations were computed for density differences versus volume difference. Subgroup analysis was performed using a 0.25 L volume difference threshold.

RESULTS

High CCC were observed for all measures in full group (CCC > 0.97). Reproducibilities of volume (RC = 0.67 L), RA950-raw (RC = 2.3%), and PD15-raw (RC = 10.6HU) were observed. Volume correction significantly improved PD15 (RC = 3.6HU) but not RA950 (RC = 1.7%). RA950-raw and PD15-raw had significantly better RC in <0.25 L subgroup than ≥0.25 L. Significant correlations with volume were observed for RA950-raw and PD15-raw (R (2) > 0.71), but not RA950-adj or PD15-adj (R (2) < 0.11).

CONCLUSIONS

Good breathhold and RA950 reproducibilities were achieved. PD15 was less reproducible but improved with volume correction or superior breathhold reproduction.

KEY POINTS

• Good breath-hold reproducibility is achievable between multiple CT examinations. • Reproducibility of densitometric measures may be improved by statistical volume correction. • Volume correction may result in decreased signal. • Densitometric reproducibility may also be improved by achieving good breath-hold reproduction. • Careful consideration of signal and noise is necessary in reproducibility assessment.

Automated segmentation of micro-CT images of bone formation in calcium phosphate scaffolds

In this work, we develop and validate an automated micro-computed tomography (micro-CT) image segmentation algorithm that accurately and efficiently segments bone, calcium phosphate (CaP)-based bone scaffold, and soft tissue. The algorithm enables quantitative evaluation of bone growth in CaP scaffolds in our study that includes many samples (100+) and large data sets (900 images per sample). The use of micro-CT for such applications is otherwise limited because the similarity in X-ray attenuation for the two materials makes them indistinguishable. Destructive characterization using histological techniques and scanning electron microscopy (SEM) has been the standard for CaP scaffolds, but these methods are cumbersome, inaccurate, and yield only 2D information. The proposed algorithm exploits scaffold periodicity and combines signal analysis, edge detection, and knowledge of three-dimensional spatial relationships between bone, CaP scaffold, and soft tissue to achieve fast and accurate segmentation. Application of this algorithm can lead to a new understanding of the role of CaP and scaffold internal structure on patterns and rates of bone growth.

Separation of left and right lungs using 3-dimensional information of sequential computed tomography images and a guided dynamic programming algorithm

OBJECTIVE

This article presents a new computerized scheme that aims to accurately and robustly separate left and right lungs on computed tomography (CT) examinations.

METHODS

We developed and tested a method to separate the left and right lungs using sequential CT information and a guided dynamic programming algorithm using adaptively and automatically selected start point and end point with especially severe and multiple connections.

RESULTS

The scheme successfully identified and separated all 827 connections on the total 4034 CT images in an independent testing data set of CT examinations. The proposed scheme separated multiple connections regardless of their locations, and the guided dynamic programming algorithm reduced the computation time to approximately 4.6% in comparison with the traditional dynamic programming and avoided the permeation of the separation boundary into normal lung tissue.

CONCLUSIONS

The proposed method is able to robustly and accurately disconnect all connections between left and right lungs, and the guided dynamic programming algorithm is able to remove redundant processing.

Automatic definition of the central-chest lymph-node stations

PURPOSE

Lung cancer remains the leading cause of cancer death in the United States. Central to the lung-cancer diagnosis and staging process is the assessment of the central-chest lymph nodes. This assessment requires two steps: (1) examination of the lymph-node stations and identification of diagnostically important nodes in a three-dimensional (3D) multidetector computed tomography (MDCT) chest scan; (2) tissue sampling of the identified nodes. We describe a computer-based system for automatically defining the central-chest lymph-node stations in a 3D MDCT chest scan.

METHODS

Automated methods first construct a 3D chest model, consisting of the airway tree, aorta, pulmonary artery, and other anatomical structures. Subsequent automated analysis then defines the 3D regional nodal stations, as specified by the internationally standardized TNM lung-cancer staging system. This analysis involves extracting over 140 pertinent anatomical landmarks from structures contained in the 3D chest model. Next, the physician uses data mining tools within the system to interactively select diagnostically important lymph nodes contained in the regional nodal stations.

RESULTS

Results from a ground-truth database of unlabeled lymph nodes identified in 32 MDCT scans verify the system's performance. The system automatically defined 3D regional nodal stations that correctly labeled 96% of the database's lymph nodes, with 93% of the stations correctly labeling 100% of their constituent nodes.

CONCLUSIONS

The system accurately defines the regional nodal stations in a given high-resolution 3D MDCT chest scan and eases a physician's burden for analyzing a given MDCT scan for lymph-node station assessment. It also shows potential as an aid for preplanning lung-cancer staging procedures.

A randomized study of endobronchial valves for advanced emphysema

BACKGROUND

Endobronchial valves that allow air to escape from a pulmonary lobe but not enter it can induce a reduction in lobar volume that may thereby improve lung function and exercise tolerance in patients with pulmonary hyperinflation related to advanced emphysema.

METHODS

We compared the safety and efficacy of endobronchial-valve therapy in patients with heterogeneous emphysema versus standard medical care. Efficacy end points were percent changes in the forced expiratory volume in 1 second (FEV1) and the 6-minute walk test on intention-to-treat analysis. We assessed safety on the basis of the rate of a composite of six major complications.

RESULTS

Of 321 enrolled patients, 220 were randomly assigned to receive endobronchial valves (EBV group) and 101 to receive standard medical care (control group). At 6 months, there was an increase of 4.3% in the FEV1 in the EBV group (an increase of 1.0 percentage point in the percent of the predicted value), as compared with a decrease of 2.5% in the control group (a decrease of 0.9 percentage point in the percent of the predicted value). Thus, there was a mean between-group difference of 6.8% in the FEV1 (P=0.005). Roughly similar between-group differences were observed for the 6-minute walk test. At 12 months, the rate of the complications composite was 10.3% in the EBV group versus 4.6% in the control group (P=0.17). At 90 days, in the EBV group, as compared with the control group, there were increased rates of exacerbation of chronic obstructive pulmonary disease (COPD) requiring hospitalization (7.9% vs. 1.1%, P=0.03) and hemoptysis (6.1% vs. 0%, P=0.01). The rate of pneumonia in the target lobe in the EBV group was 4.2% at 12 months. Greater radiographic evidence of emphysema heterogeneity and fissure completeness was associated with an enhanced response to treatment.

CONCLUSIONS

Endobronchial-valve treatment for advanced heterogeneous emphysema induced modest improvements in lung function, exercise tolerance, and symptoms at the cost of more frequent exacerbations of COPD, pneumonia, and hemoptysis after implantation. (Funded by Pulmonx; ClinicalTrials.gov number, NCT00129584.)

Robust 3-D airway tree segmentation for image-guided peripheral bronchoscopy

A vital task in the planning of peripheral bronchoscopy is the segmentation of the airway tree from a 3-D multidetector computed tomography chest scan. Unfortunately, existing methods typically do not sufficiently extract the necessary peripheral airways needed to plan a procedure. We present a robust method that draws upon both local and global information. The method begins with a conservative segmentation of the major airways. Follow-on stages then exhaustively search for additional candidate airway locations. Finally, a graph-based optimization method counterbalances both the benefit and cost of retaining candidate airway locations for the final segmentation. Results demonstrate that the proposed method typically extracts 2-3 more generations of airways than several other methods, and that the extracted airway trees enable image-guided bronchoscopy deeper into the human lung periphery than past studies.

Reproducibility of lung and lobar volume measurements using computed tomography

RATIONALE AND OBJECTIVES

Lung and lobar volume measurements from computed tomographic (CT) imaging are being used in clinical trials to assess new minimally invasive emphysema treatments aiming to reduce lung volumes. Establishing the reproducibility of lung volume measurements is important if they are to be accepted as treatment planning and outcome variables. The aims of this study were to (1) investigate the correlation between lung volumes assessed on CT imaging and on pulmonary function testing (PFT), (2) compare the two methods' reproducibility, and (3) assess the reproducibility of CT lobar volumes.

MATERIALS AND METHODS

CT imaging and body plethysmography were performed at baseline and after a 9-month interval in multicenter emphysema treatment trials. Lung volumes were measured at total lung capacity (TLC) and at residual volume (RV). Lobar volumes were measured on CT imaging using a semiautomated technique. The correlations between CT and PFT volumes were computed for 486 subjects at baseline. Reproducibility was assessed in terms of the intraclass correlation coefficient (ICC) for 126 subjects from the control group at TLC and 120 subjects at RV.

RESULTS

Correlations between CT and PFT lung volumes were 0.86 at TLC and 0.67 at RV. At TLC, the ICCs were 0.943 for CT imaging and 0.814 for PFT. At RV, the ICCs were 0.886 for CT imaging and 0.683 for PFT. CT lobar volumes showed good reproducibility (all P values < .05).

CONCLUSION

CT lung and lobar volume measurements could be captured in a multicenter trial setting with high reproducibility and were highly correlated with those obtained on PFT. CT imaging showed significantly better reproducibility than PFT between interval lung volume measurements, offering the potential for designing emphysema treatment trials involving fewer subjects.

A new computationally efficient CAD system for pulmonary nodule detection in CT imagery

Early detection of lung nodules is extremely important for the diagnosis and clinical management of lung cancer. In this paper, a novel computer aided detection (CAD) system for the detection of pulmonary nodules in thoracic computed tomography (CT) imagery is presented. The paper describes the architecture of the CAD system and assesses its performance on a publicly available database to serve as a benchmark for future research efforts. Training and tuning of all modules in our CAD system is done using a separate and independent dataset provided courtesy of the University of Texas Medical Branch (UTMB). The publicly available testing dataset is that created by the Lung Image Database Consortium (LIDC). The LIDC data used here is comprised of 84 CT scans containing 143 nodules ranging from 3 to 30mm in effective size that are manually segmented at least by one of the four radiologists. The CAD system uses a fully automated lung segmentation algorithm to define the boundaries of the lung regions. It combines intensity thresholding with morphological processing to detect and segment nodule candidates simultaneously. A set of 245 features is computed for each segmented nodule candidate. A sequential forward selection process is used to determine the optimum subset of features for two distinct classifiers, a Fisher Linear Discriminant (FLD) classifier and a quadratic classifier. A performance comparison between the two classifiers is presented, and based on this, the FLD classifier is selected for the CAD system. With an average of 517.5 nodule candidates per case/scan (517.5+/-72.9), the proposed front-end detector/segmentor is able to detect 92.8% of all the nodules in the LIDC/testing dataset (based on merged ground truth). The mean overlap between the nodule regions delineated by three or more radiologists and the ones segmented by the proposed segmentation algorithm is approximately 63%. Overall, with a specificity of 3 false positives (FPs) per case/patient on average, the CAD system is able to correctly identify 80.4% of the nodules (115/143) using 40 selected features. A 7-fold cross-validation performance analysis using the LIDC database only shows CAD sensitivity of 82.66% with an average of 3 FPs per CT scan/case.

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.

Controlling familywise error rate for matched subspace detection in dynamic FDG PET

Detection of small lesions in fluorodeoxyglucose (FDG) positron emission tomography (PET) is limited by image resolution and low signal to noise ratio. We have previously described a matched subspace detection method that uses the time activity curve to distinguish tumors from background in dynamic FDG PET. Applying this algorithm on a voxel by voxel basis throughout the dynamic image produces a test statistic image or "map" which on thresholding indicates the potential locations of secondary or metastatic tumors. In this paper, we describe a thresholding method that controls familywise error rate (FWER) for the matched subspace detection statistical map. The method involves three steps. First, the PET image is segmented into several homogeneous regions. Then, the statistical map is normalized to a zero mean unit variance Gaussian random field. Finally, the images are thresholded at a fixed FWER by estimating their spatial smoothness and applying a random field theory maximum statistic approach. We evaluate this thresholding method using digital phantoms generated from clinical dynamic images. We also present an application of the proposed approach to clinical PET data from a breast cancer patient with metastatic disease.