A public image database to support research in computer aided diagnosis

A Point-Matching Method of Moment with Sparse Bayesian Learning Applied and Evaluated in Dynamic Lung Electrical Impedance Tomography

Dynamic lung imaging is a major application of Electrical Impedance Tomography (EIT) due to EIT's exceptional temporal resolution, low cost and absence of radiation. EIT however lacks in spatial resolution and the image reconstruction is very sensitive to mismatches between the actual object's and the reconstruction domain's geometries, as well as to the signal noise. The non-linear nature of the reconstruction problem may also be a concern, since the lungs' significant conductivity changes due to inhalation and exhalation. In this paper, a recently introduced method of moment is combined with a sparse Bayesian learning approach to address the non-linearity issue, provide robustness to the reconstruction problem and reduce image artefacts. To evaluate the proposed methodology, we construct three CT-based time-variant 3D thoracic structures including the basic thoracic tissues and considering 5 different breath states from end-expiration to end-inspiration. The Graz consensus reconstruction algorithm for EIT (GREIT), the correlation coefficient (CC), the root mean square error (RMSE) and the full-reference (FR) metrics are applied for the image quality assessment. Qualitative and quantitative comparison with traditional and more advanced reconstruction techniques reveals that the proposed method shows improved performance in the majority of cases and metrics. Finally, the approach is applied to single-breath online in-vivo data to qualitatively verify its applicability.

Modified Quality Threshold Clustering for Temporal Analysis and Classification of Lung Lesions

Lung cancer is the type of cancer that most often kills after the initial diagnosis. To aid the specialist in its diagnosis, temporal evaluation is a potential tool for analyzing indeterminate lesions, which may be benign or malignant, during treatment. With this goal in mind, a methodology is herein proposed for the analysis, quantification, and visualization of changes in lung lesions. This methodology uses a modified version of the quality threshold clustering algorithm to associate each voxel of the lesion to a cluster, and changes in the lesion over time are defined in terms of voxel moves to another cluster. In addition, statistical features are extracted for classification of the lesion as benign or malignant. To develop the proposed methodology, two databases of pulmonary lesions were used, one for malignant lesions in treatment (public) and the other for indeterminate cases (private). We determined that the density change percentage varied from 6.22% to 36.93% of lesion volume in the public database of malignant lesions under treatment and from 19.98% to 38.81% in the private database of lung nodules. Additionally, other inter-cluster density change measures were obtained. These measures indicate the degree of change in the clusters and how each of them is abundant in relation to volume. From the statistical analysis of regions in which the density changes occurred, we were able to discriminate lung lesions with an accuracy of 98.41%, demonstrating that these changes can indicate the true nature of the lesion. In addition to visualizing the density changes occurring in lesions over time, we quantified these changes and analyzed the entire set through volumetry, which is the technique most commonly used to analyze changes in pulmonary lesions.

Large-scale image region documentation for fully automated image biomarker algorithm development and evaluation

With the advent of fully automated image analysis and modern machine learning methods, there is a need for very large image datasets having documented segmentations for both computer algorithm training and evaluation. This paper presents a method and implementation for facilitating such datasets that addresses the critical issue of size scaling for algorithm validation and evaluation; current evaluation methods that are usually used in academic studies do not scale to large datasets. This method includes protocols for the documentation of many regions in very large image datasets; the documentation may be incrementally updated by new image data and by improved algorithm outcomes. This method has been used for 5 years in the context of chest health biomarkers from low-dose chest CT images that are now being used with increasing frequency in lung cancer screening practice. The lung scans are segmented into over 100 different anatomical regions, and the method has been applied to a dataset of over 20,000 chest CT images. Using this framework, the computer algorithms have been developed to achieve over 90% acceptable image segmentation on the complete dataset.

Voxel-based comparative analysis of lung lesions in CT for therapeutic purposes

Lung cancer remains as one of the most incident types of cancer throughout the world. Temporal evaluation has become a very useful tool when one wishes to analyze some malignancy-indicating behavior. The objective of the present work is to detect changes in the local densities of lung lesions over time (follow-up analysis). From the detected changes, local information as well as extent region of changes can complement the studies regarding the malignant or benign nature of the lesion. Based on this idea, we attempt to use techniques that allow the observation of changes in the lesion over time, based on remote sensing techniques which highlight changes occurring in the environment. The techniques used were the image differencing, image rationing, median filtering, image regression and the fuzzy XOR operator. Based on the global measurement of change percentage in the density, we found density variations which were considered significant in a range from 2.22 to 36.57 % of the volume of the lesion. The results achieved are promising since, besides the visual aspects of the changes in density of the lung lesion over time, we managed to quantify these changes and compare them by volumetric analysis, a more commonly used technique for analysis of changes in lung lesions.

Automated 3D closed surface segmentation: application to vertebral body segmentation in CT images

PURPOSE

A fully automated segmentation algorithm, progressive surface resolution (PSR), is presented in this paper to determine the closed surface of approximately convex blob-like structures that are common in biomedical imaging. The PSR algorithm was applied to the cortical surface segmentation of 460 vertebral bodies on 46 low-dose chest CT images, which can be potentially used for automated bone mineral density measurement and compression fracture detection.

METHODS

The target surface is realized by a closed triangular mesh, which thereby guarantees the enclosure. The surface vertices of the triangular mesh representation are constrained along radial trajectories that are uniformly distributed in 3D angle space. The segmentation is accomplished by determining for each radial trajectory the location of its intersection with the target surface. The surface is first initialized based on an input high confidence boundary image and then resolved progressively based on a dynamic attraction map in an order of decreasing degree of evidence regarding the target surface location.

RESULTS

For the visual evaluation, the algorithm achieved acceptable segmentation for 99.35 % vertebral bodies. Quantitative evaluation was performed on 46 vertebral bodies and achieved overall mean Dice coefficient of 0.939 (with max [Formula: see text] 0.957, min [Formula: see text] 0.906 and standard deviation [Formula: see text] 0.011) using manual annotations as the ground truth.

CONCLUSIONS

Both visual and quantitative evaluations demonstrate encouraging performance of the PSR algorithm. This novel surface resolution strategy provides uniform angular resolution for the segmented surface with computation complexity and runtime that are linearly constrained by the total number of vertices of the triangular mesh representation.

Local noise estimation in low-dose chest CT images

PURPOSE

Image noise in computed tomography (CT) images may have significant local variation due to tissue properties, dose, and location of the X-ray source. We developed and tested an automated tissue-based estimator method for estimating local noise in CT images.

METHOD

An automated TBE method for estimating the local noise in CT image in 3 steps was developed: (1) Partition the image into homogeneous and transition regions, (2) For each pixel in the homogeneous regions, compute the standard deviation in a 15 x 15 x 1 voxel local region using only pixels from the same homogeneous region, and (3) Interpolate the noise estimate from the homogeneous regions in the transition regions. Noise-aware fat segmentation was implemented. Experiments were conducted on the anthropomorphic phantom and in vivo low-dose chest CT scans to validate the TBE, characterize the magnitude of local noise variation, and determine the sensitivity of noise estimates to the size of the region in which noise is computed. The TBE was tested on all scans from the Early Lung Cancer Action Program public database. The TBE was evaluated quantitatively on the phantom data and qualitatively on the in vivo data.

RESULTS

The results show that noise can vary locally by over 200 Hounsfield units on low-dose in vivo chest CT scans and that the TBE can characterize these noise variations within 5 %. The new fat segmentation algorithm successfully improved segmentation on all 50 scans tested.

CONCLUSION

The TBE provides a means to estimate noise for image quality monitoring, optimization of denoising algorithms, and improvement of segmentation algorithms. The TBE was shown to accurately characterize the large local noise variations that occur due to changes in material, dose, and X-ray source location.

The Lung Image Database Consortium (LIDC) and Image Database Resource Initiative (IDRI): a completed reference database of lung nodules on CT scans

PURPOSE

The development of computer-aided diagnostic (CAD) methods for lung nodule detection, classification, and quantitative assessment can be facilitated through a well-characterized repository of computed tomography (CT) scans. The Lung Image Database Consortium (LIDC) and Image Database Resource Initiative (IDRI) completed such a database, establishing a publicly available reference for the medical imaging research community. Initiated by the National Cancer Institute (NCI), further advanced by the Foundation for the National Institutes of Health (FNIH), and accompanied by the Food and Drug Administration (FDA) through active participation, this public-private partnership demonstrates the success of a consortium founded on a consensus-based process.

METHODS

Seven academic centers and eight medical imaging companies collaborated to identify, address, and resolve challenging organizational, technical, and clinical issues to provide a solid foundation for a robust database. The LIDC/IDRI Database contains 1018 cases, each of which includes images from a clinical thoracic CT scan and an associated XML file that records the results of a two-phase image annotation process performed by four experienced thoracic radiologists. In the initial blinded-read phase, each radiologist independently reviewed each CT scan and marked lesions belonging to one of three categories ("nodule > or =3 mm," "nodule <3 mm," and "non-nodule > or =3 mm"). In the subsequent unblinded-read phase, each radiologist independently reviewed their own marks along with the anonymized marks of the three other radiologists to render a final opinion. The goal of this process was to identify as completely as possible all lung nodules in each CT scan without requiring forced consensus.

RESULTS

The Database contains 7371 lesions marked "nodule" by at least one radiologist. 2669 of these lesions were marked "nodule > or =3 mm" by at least one radiologist, of which 928 (34.7%) received such marks from all four radiologists. These 2669 lesions include nodule outlines and subjective nodule characteristic ratings.

CONCLUSIONS

The LIDC/IDRI Database is expected to provide an essential medical imaging research resource to spur CAD development, validation, and dissemination in clinical practice.

In the spotlight: biomedical imaging

This article resumes on a selected set of topics and collects promising and recent research advances in the field of multimodal temporal data analysis, high-field magnetic resonance spectroscopy, trends in computer-aided diagnosis and advances in cardiac diagnostic imaging. The first section briefly points to promising work on statistical models for tracking, detection, and segmentation in multimodal temporal imagery. Section III gives a brief snapshot of slice selective free induction decay (FID) acquisition for 7 tesla high-field MR imaging. Section IV outlines highlights in comparative validation of computer-aided diagnosis and associated image analysis algorithms spanning a variety of application domains from the heart to the eye. Lastly, Section V describes advances in the analysis of real-time three-dimensional (3-D) echocardiography for computing myocardial strain.