Cupping artifact correction and automated classification for high-resolution dedicated breast CT images

Cone-beam breast CT using an offset detector: effect of detector offset and image reconstruction algorithm

Objective.A dedicated cone-beam breast computed tomography (BCT) using a high-resolution, low-noise detector operating in offset-detector geometry has been developed. This study investigates the effects of varying detector offsets and image reconstruction algorithms to determine the appropriate combination of detector offset and reconstruction algorithm.Approach.Projection datasets (300 projections in 360°) of 30 breasts containing calcified lesions that were acquired using a prototype cone-beam BCT system comprising a 40 × 30 cm flat-panel detector with 1024 × 768 detector pixels were reconstructed using Feldkamp-Davis-Kress (FDK) algorithm and served as the reference. The projection datasets were retrospectively truncated to emulate cone-beam datasets with sinograms of 768×768 and 640×768 detector pixels, corresponding to 5 cm and 7.5 cm lateral offsets, respectively. These datasets were reconstructed using the FDK algorithm with appropriate weights and an ASD-POCS-based Fast, total variation-Regularized, Iterative, Statistical reconstruction Technique (FRIST), resulting in a total of 4 offset-detector reconstructions (2 detector offsets × 2 reconstruction methods). Signal difference-to-noise ratio (SDNR), variance, and full-width at half-maximum (FWHM) of calcifications in two orthogonal directions were determined from all reconstructions. All quantitative measurements were performed on images in units of linear attenuation coefficient (1/cm).Results.The FWHM of calcifications did not differ (P > 0.262) among reconstruction algorithms and detector formats, implying comparable spatial resolution. For a chosen detector offset, the FRIST algorithm outperformed FDK in terms of variance and SDNR (P < 0.0001). For a given reconstruction method, the 5 cm offset provided better results.Significance.This study indicates the feasibility of using the compressed sensing-based, FRIST algorithm to reconstruct sinograms from offset-detectors. Among the reconstruction methods and detector offsets studied, FRIST reconstructions corresponding to a 30 cm × 30 cm with 5 cm lateral offset, achieved the best performance. A clinical prototype using such an offset geometry has been developed and installed for clinical trials.

A novel hardware duo of beam modulation and shielding to reduce scatter acquisition and dose in cone-beam breast CT

PURPOSE

In cone-beam breast CT, scattered photons form a large portion of the acquired signal, adversely impacting image quality throughout the frequency response of the imaging system. Prior simulation studies provided proof of concept for utilization of a hardware solution to prevent scatter acquisition. Here, we report the design, implementation, and characterization of an auxiliary apparatus of fluence modulation and scatter shielding that does indeed lead to projections with a reduced level of scatter.

METHODS

An apparatus was designed for permanent installation within an existing cone-beam CT system. The apparatus is composed of two primary assemblies: a "Fluence Modulator" (FM) and a "Scatter Shield" (SS). The design of the assemblies enables them to operate in synchrony during image acquisition, converting the sourced x-rays into a moving narrow beam. During a projection, this narrow beam sweeps the entire fan angle coverage of the imaging system. As the two assemblies are contingent on one another, their joint implementation is described in the singular as apparatus FM-SS. The FM and the SS assemblies are each comprised a metal housing, a sensory system, and a robotic system. A controller unit handles their relative movements. A series of comparative studies were conducted to evaluate the performance of a cone-beam CT system in two "modes" of operation: with and without FM-SS installed, and to compare the results of physical implementation with those previously simulated. The dynamic range requirements of the utilized detector in the cone-beam CT imaging system were first characterized, independent of the mode of operation. We then characterized and compared the spatial resolution of the imaging system with, and without, FM-SS. A physical breast phantom, representative of an average size breast, was developed and imaged. Actual differences in signal level obtained with, versus without, FM-SS were then compared to the expected level gains based on previously reported simulations. Following these initial assessments, the scatter acquisition in each projection in both modes of operation was investigated. Finally, as an initial study of the impact of FM-SS on radiation dose in an average size breast, a series of Monte Carlo simulations were coupled with physical measurements of air kerma, with and without FM-SS.

RESULTS

With implementation of FM-SS, the detector's required dynamic range was reduced by a factor of 5.5. Substantial reduction in the acquisition of the scattered rays, by a factor of 5.1 was achieved. With the implementation of FM-SS, deposited dose was reduced by 27% in the studied breast.

CONCLUSIONS

The disclosed implementation of FM-SS, within a cone-beam breast CT system, results in reduction of scatter-components in acquired projections, reduction of dose deposit to the breast, and relaxation of requirements for the detector's dynamic range. Controlling or correcting for patient motion occurring during image acquisition remains an open problem to be solved prior to practical clinical usage of FM-SS cone-beam breast CT.

Flat-panel conebeam CT in the clinic: history and current state

Research into conebeam CT concepts began as soon as the first clinical single-slice CT scanner was conceived. Early implementations of conebeam CT in the 1980s focused on high-contrast applications where concurrent high resolution ( < 200    μ m ), for visualization of small contrast-filled vessels, bones, or teeth, was an imaging requirement that could not be met by the contemporaneous CT scanners. However, the use of nonlinear imagers, e.g., x-ray image intensifiers, limited the clinical utility of the earliest diagnostic conebeam CT systems. The development of consumer-electronics large-area displays provided a technical foundation that was leveraged in the 1990s to first produce large-area digital x-ray detectors for use in radiography and then compact flat panels suitable for high-resolution and high-frame-rate conebeam CT. In this review, we show the concurrent evolution of digital flat panel (DFP) technology and clinical conebeam CT. We give a brief summary of conebeam CT reconstruction, followed by a brief review of the correction approaches for DFP-specific artifacts. The historical development and current status of flat-panel conebeam CT in four clinical areas-breast, fixed C-arm, image-guided radiation therapy, and extremity/head-is presented. Advances in DFP technology over the past two decades have led to improved visualization of high-contrast, high-resolution clinical tasks, and image quality now approaches the soft-tissue contrast resolution that is the standard in clinical CT. Future technical developments in DFPs will enable an even broader range of clinical applications; research in the arena of flat-panel CT shows no signs of slowing down.

Imaging Study of Pseudo-CT Synthesized From Cone-Beam CT Based on 3D CycleGAN in Radiotherapy

Purpose

To propose a synthesis method of pseudo-CT (CT_CycleGAN) images based on an improved 3D cycle generative adversarial network (CycleGAN) to solve the limitations of cone-beam CT (CBCT), which cannot be directly applied to the correction of radiotherapy plans.

Methods

The improved U-Net with residual connection and attention gates was used as the generator, and the discriminator was a full convolutional neural network (FCN). The imaging quality of pseudo-CT images is improved by adding a 3D gradient loss function. Fivefold cross-validation was performed to validate our model. Each pseudo CT generated is compared against the real CT image (ground truth CT, CT_gt) of the same patient based on mean absolute error (MAE) and structural similarity index (SSIM). The dice similarity coefficient (DSC) coefficient was used to evaluate the segmentation results of pseudo CT and real CT. 3D CycleGAN performance was compared to 2D CycleGAN based on normalized mutual information (NMI) and peak signal-to-noise ratio (PSNR) metrics between the pseudo-CT and CT_gt images. The dosimetric accuracy of pseudo-CT images was evaluated by gamma analysis.

Results

The MAE metric values between the CT_CycleGAN and the real CT in fivefold cross-validation are 52.03 ± 4.26HU, 50.69 ± 5.25HU, 52.48 ± 4.42HU, 51.27 ± 4.56HU, and 51.65 ± 3.97HU, respectively, and the SSIM values are 0.87 ± 0.02, 0.86 ± 0.03, 0.85 ± 0.02, 0.85 ± 0.03, and 0.87 ± 0.03 respectively. The DSC values of the segmentation of bladder, cervix, rectum, and bone between CT_CycleGAN and real CT images are 91.58 ± 0.45, 88.14 ± 1.26, 87.23 ± 2.01, and 92.59 ± 0.33, respectively. Compared with 2D CycleGAN, the 3D CycleGAN based pseudo-CT image is closer to the real image, with NMI values of 0.90 ± 0.01 and PSNR values of 30.70 ± 0.78. The gamma pass rate of the dose distribution between CT_CycleGAN and CT_gt is 97.0% (2%/2 mm).

Conclusion

The pseudo-CT images obtained based on the improved 3D CycleGAN have more accurate electronic density and anatomical structure.

Correction of Cupping Artifacts in Axial Cone-Beam Computed Tomography Images by Using Image Processing Algorithms

Objectives:

One of the most important problems of cone-beam computed tomography (CBCT) imaging technique is the presence of dense objects, such as implants, amalgam fillings, and metal veneers, which result in beam-hardening artifacts. With an increase in the application of CBCT images and considering the problems in relation to cupping artifacts, some algorithms have been presented to reduce these artifacts. The aim was to present an algorithm to eliminate cupping artifacts from axial and other reconstructed CBCT images.

Materials and Methods:

We used CBCT images of NewTom VG imaging system (Verona, Italy, at Dentistry Faculty, Medical Sciences University, Tabriz, Iran) in which every image has a resolution of 366 × 320 in DICOM format. 50 images of patients with cupping artifacts were selected. Using Sobel edge detector and nonlinear gamma correction coefficient, the difference was calculated between the density of axial images in the main image and the image resulting from nonlinear gamma correction at the exact location of the radiopaque dental materials detected by Sobel. The points at which this density difference was out of a definite limit were treated as image artifacts and were eliminated from the main image by the inpainting method.

Results:

The resultant axial images, for producing reconstructed cross-sectional, panoramic images without cupping artifacts, were imported into NTT viewer V5.6 and utilized.

Conclusions:

With comparison, acquired images observed that the offering algorithm is practical and effective for reducing the cupping artifacts and preserving the quality of the reconstructed images. This algorithm does not need any additional equipment.

Pseudo CT Estimation using Patch-based Joint Dictionary Learning

Magnetic resonance (MR) simulators have recently gained popularity; it avoids the unnecessary radiation exposure associated with Computed Tomography (CT) when used for radiation therapy planning. We propose a method for pseudo CT estimation from MR images based on joint dictionary learning. Patient-specific anatomical features were extracted from the aligned training images and adopted as signatures for each voxel. The most relevant and informative features were identified to train the joint dictionary learning-based model. The well-trained dictionary was used to predict the pseudo CT of a new patient. This prediction technique was validated with a clinical study of 12 patients with MR and CT images of the brain. The mean absolute error (MAE), peak signal-to-noise ratio (PSNR), normalized cross correlation (NCC) indexes were used to quantify the prediction accuracy. We compared our proposed method with a state-of-the-art dictionary learning method. Overall our proposed method significantly improves the prediction accuracy over the state-of-the-art dictionary learning method. We have investigated a novel joint dictionary Iearning- based approach to predict CT images from routine MRIs and demonstrated its reliability. This CT prediction technique could be a useful tool for MRI-based radiation treatment planning or attenuation correction for quantifying PET images for PET/MR imaging.

Shading artifact correction in breast CT using an interleaved deep learning segmentation and maximum-likelihood polynomial fitting approach

PURPOSE

The purpose of this work was twofold: (a) To provide a robust and accurate method for image segmentation of dedicated breast CT (bCT) volume data sets, and (b) to improve Hounsfield unit (HU) accuracy in bCT by means of a postprocessing method that uses the segmented images to correct for the low-frequency shading artifacts in reconstructed images.

METHODS

A sequential and iterative application of image segmentation and low-order polynomial fitting to bCT volume data sets was used in the interleaved correction (IC) method. Image segmentation was performed through a deep convolutional neural network (CNN) with a modified U-Net architecture. A total of 45 621 coronal bCT images from 111 patient volume data sets were segmented (using a previously published segmentation algorithm) and used for neural network training, validation, and testing. All patient data sets were selected from scans performed on four different prototype breast CT systems. The adipose voxels for each patient volume data set, segmented using the proposed CNN, were then fit to a three-dimensional low-order polynomial. The polynomial fit was subsequently used to correct for the shading artifacts introduced by scatter and beam hardening in a method termed "flat fielding." An interleaved utilization of image segmentation and flat fielding was repeated until a convergence criterion was satisfied. Mathematical and physical phantom studies were conducted to evaluate the dependence of the proposed algorithm on breast size and the distribution of fibroglandular tissue. In addition, a subset of patient scans (not used in the CNN training, testing or validation) were used to investigate the accuracy of the IC method across different scanner designs and beam qualities.

RESULTS

The IC method resulted in an accurate classification of different tissue types with an average Dice similarity coefficient > 95%, precision > 97%, recall > 95%, and F1-score > 96% across all tissue types. The flat fielding correction of bCT images resulted in a significant reduction in either cupping or capping artifacts in both mathematical and physical phantom studies as measured by the integral nonuniformity metric with an average reduction of 71% for cupping and 30% for capping across different phantom sizes, and the Uniformity Index with an average reduction of 53% for cupping and 34% for capping.

CONCLUSION

The validation studies demonstrated that the IC method improves Hounsfield Units (HU) accuracy and effectively corrects for shading artifacts caused by scatter contamination and beam hardening. The postprocessing approach described herein is relevant to the broad scope of bCT devices and does not require any modification in hardware or existing scan protocols. The trained CNN parameters and network architecture are available for interested users.

Dosimetric study on learning-based cone-beam CT correction in adaptive radiation therapy

INTRODUCTION

Cone-beam CT (CBCT) image quality is important for its quantitative analysis in adaptive radiation therapy. However, due to severe artifacts, the CBCTs are primarily used for verifying patient setup only so far. We have developed a learning-based image quality improvement method which could provide CBCTs with image quality comparable to planning CTs (pCTs). The accuracy of dose calculations based on these CBCTs is unknown. In this study, we aim to investigate the dosimetric accuracy of our corrected CBCT (CCBCT) in brain stereotactic radiosurgery (SRS) and pelvic radiotherapy.

MATERIALS AND METHODS

We retrospectively investigated a total of 32 treatment plans from 22 patients, each of whom with both original treatment pCTs and CBCTs acquired during treatment setup. The CCBCT and original CBCT (OCBCT) were registered to the pCT for generating CCBCT-based and OCBCT-based treatment plans. The original pCT-based plans served as ground truth. Clinically-relevant dose volume histogram (DVH) metrics were extracted from the ground truth, OCBCT-based and CCBCT-based plans for comparison. Gamma analysis was also performed to compare the absorbed dose distributions between the pCT-based and OCBCT/CCBCT-based plans of each patient.

RESULTS

CCBCTs demonstrated better image contrast and more accurate HU ranges when compared side-by-side with OCBCTs. For pelvic radiotherapy plans, the mean dose error in DVH metrics for planning target volume (PTV), bladder and rectum was significantly reduced, from 1% to 0.3%, after CBCT correction. The gamma analysis showed the average pass rate increased from 94.5% before correction to 99.0% after correction. For brain SRS treatment plans, both original and corrected CBCT images were accurate enough for dose calculation, though CCBCT featured higher image quality.

CONCLUSION

CCBCTs can provide a level of dose accuracy comparable to traditional pCTs for brain and prostate radiotherapy planning and the correction method proposed here can be useful in CBCT-guided adaptive radiotherapy.

Deep dense multi-path neural network for prostate segmentation in magnetic resonance imaging

PURPOSE

We propose an approach of 3D convolutional neural network to segment the prostate in MR images.

METHODS

A 3D deep dense multi-path convolutional neural network that follows the framework of the encoder-decoder design is proposed. The encoder is built based upon densely connected layers that learn the high-level feature representation of the prostate. The decoder interprets the features and predicts the whole prostate volume by utilizing a residual layout and grouped convolution. A set of sub-volumes of MR images, centered at the prostate, is generated and fed into the proposed network for training purpose. The performance of the proposed network is compared to previously reported approaches.

RESULTS

Two independent datasets were employed to assess the proposed network. In quantitative evaluations, the proposed network achieved 95.11 and 89.01 Dice coefficients for the two datasets. The segmentation results were robust to variations in MR images. In comparison experiments, the segmentation performance of the proposed network was comparable to the previously reported approaches. In qualitative evaluations, the segmentation results by the proposed network were well matched to the ground truth provided by human experts.

CONCLUSIONS

The proposed network is capable of segmenting the prostate in an accurate and robust manner. This approach can be applied to other types of medical images.

Internal breast dosimetry in mammography: Monte Carlo validation in homogeneous and anthropomorphic breast phantoms with a clinical mammography system

PURPOSE

To validate Monte Carlo (MC)-based breast dosimetry estimations using both a homogeneous and a 3D anthropomorphic breast phantom under polyenergetic irradiation for internal breast dosimetry purposes.

METHODS

Experimental measurements were performed with a clinical digital mammography system (Mammomat Inspiration, Siemens Healthcare), using the x-ray spectrum selected by the automatic exposure control and a tube current-exposure time product of 360 mAs. A homogeneous 50% glandular breast phantom and a 3D anthropomorphic breast phantom were used to investigate the dose at different depths (range 0-4 cm with 1 cm steps) for the homogeneous case and at a depth of 2.25 cm for the anthropomorphic case. Local dose deposition was measured using thermoluminescent dosimeters (TLD), metal oxide semiconductor field-effect transistor dosimeters (MOSFET), and GafChromic™ films. A Geant4-based MC simulation was modified to match the clinical experimental setup. Thirty sensitive volumes (3.2 × 3.2 × 0.38 mm3 ) on the axial-phantom plane were included at each depth in the simulation to characterize the internal dose variation and compare it to the experimental TLD and MOSFET measurements. The experimental 2D dose maps obtained with the GafChromic™ films were compared to the simulated 2D dose distributions.

RESULTS

Due to the energy dependence of the dosimeters and due to x-ray beam hardening, dosimeters based on these three technologies have to be calibrated at each depth of the phantom. As expected, the dose was found to decrease with increasing phantom depth, with the reduction being ~93% after 4 cm for the homogeneous breast phantom. The 2D dose map showed nonuniformities in the dose distribution in the axial plane of the phantom. The mean combined standard uncertainty increased with phantom depth by up to 5.3% for TLD, 6.3% for MOSFET, and 9.6% for GafChromic™ film. In the case of a heterogeneous phantom, the dosimeters are able to detect local dose gradient variations. In particular, GafChromic™ film showed local dose variations of about 46% at the boundaries of two materials.

CONCLUSIONS

Results showed a good agreement between experimental measurements (with TLD and MOSFET) and MC data for both homogeneous and anthropomorphic breast phantoms. Larger discrepancies are found when comparing the GafChromic™ dose values to the MC results due to the inherent less precise nature of the former. MC validations in a heterogeneous background at the level of local dose deposition and in absolute terms play a fundamental role in the development of an accurate method to estimate radiation dose. The potential introduction of a breast dosimetry model involving a nonhomogeneous glandular/adipose tissue composition makes the validation of internal dose distributions estimates crucial.

PSNet: prostate segmentation on MRI based on a convolutional neural network

Automatic segmentation of the prostate on magnetic resonance images (MRI) has many applications in prostate cancer diagnosis and therapy. We proposed a deep fully convolutional neural network (CNN) to segment the prostate automatically. Our deep CNN model is trained end-to-end in a single learning stage, which uses prostate MRI and the corresponding ground truths as inputs. The learned CNN model can be used to make an inference for pixel-wise segmentation. Experiments were performed on three data sets, which contain prostate MRI of 140 patients. The proposed CNN model of prostate segmentation (PSNet) obtained a mean Dice similarity coefficient of [Formula: see text] as compared to the manually labeled ground truth. Experimental results show that the proposed model could yield satisfactory segmentation of the prostate on MRI.

A Denoising Algorithm for CT Image Using Low-rank Sparse Coding

We propose a denoising method of CT image based on low-rank sparse coding. The proposed method constructs an adaptive dictionary of image patches and estimates the sparse coding regularization parameters using the Bayesian interpretation. A low-rank approximation approach is used to simultaneously construct the dictionary and achieve sparse representation through clustering similar image patches. A variable-splitting scheme and a quadratic optimization are used to reconstruct CT image based on achieved sparse coefficients. We tested this denoising technology using phantom, brain and abdominal CT images. The experimental results showed that the proposed method delivers state-of-art denoising performance, both in terms of objective criteria and visual quality.

Improving Image Quality of Cone-Beam CT Using Alternating Regression Forest

We propose a CBCT image quality improvement method based on anatomic signature and auto-context alternating regression forest. Patient-specific anatomical features are extracted from the aligned training images and served as signatures for each voxel. The most relevant and informative features are identified to train regression forest. The well-trained regression forest is used to correct the CBCT of a new patient. This proposed algorithm was evaluated using 10 patients' data with CBCT and CT images. The mean absolute error (MAE), peak signal-to-noise ratio (PSNR) and normalized cross correlation (NCC) indexes were used to quantify the correction accuracy of the proposed algorithm. The mean MAE, PSNR and NCC between corrected CBCT and ground truth CT were 16.66HU, 37.28dB and 0.98, which demonstrated the CBCT correction accuracy of the proposed learning-based method. We have developed a learning-based method and demonstrated that this method could significantly improve CBCT image quality. The proposed method has great potential in improving CBCT image quality to a level close to planning CT, therefore, allowing its quantitative use in CBCT-guided adaptive radiotherapy.

High-resolution CT Image Retrieval Using Sparse Convolutional Neural Network

We propose a high-resolution CT image retrieval method based on sparse convolutional neural network. The proposed framework is used to train the end-to-end mapping from low-resolution to high-resolution images. The patch-wise feature of low-resolution CT is extracted and sparsely represented by a convolutional layer and a learned iterative shrinkage threshold framework, respectively. Restricted linear unit is utilized to non-linearly map the low-resolution sparse coefficients to the high-resolution ones. An adaptive high-resolution dictionary is applied to construct the informative signature which is highly connected to a high-resolution patch. Finally, we feed the signature to a convolutional layer to reconstruct the predicted high-resolution patches and average these overlapping patches to generate high-resolution CT. The loss function between reconstructed images and the corresponding ground truth high-resolution images is applied to optimize the parameters of end-to-end neural network. The well-trained map is used to generate the high-resolution CT from a new low-resolution input. This technique was tested with brain and lung CT images and the image quality was assessed using the corresponding CT images. Peak signal-to-noise ratio (PSNR), structural similarity index (SSIM) and mean absolute error (MAE) indexes were used to quantify the differences between the generated high-resolution and corresponding ground truth CT images. The experimental results showed the proposed method could enhance images resolution from low-resolution images. The proposed method has great potential in improving radiation dose calculation and delivery accuracy and decreasing CT radiation exposure of patients.

Toward image quality assessment in mammography using model observers: Detection of a calcification-like object

PURPOSE

Model observers (MOs) are of interest in the field of medical imaging to assess image quality. However, before procedures using MOs can be proposed in quality control guidelines for mammography systems, we need to know whether MOs are sensitive to changes in image quality and correlations in background structure. Therefore, as a proof of principle, in this study human and model observer (MO) performance are compared for the detection of calcification-like objects using different background structures and image quality levels of unprocessed mammography images.

METHOD

Three different phantoms, homogeneous polymethyl methacrylate, BR3D slabs with swirled patterns (CIRS, Norfolk, VA, USA), and a prototype anthropomorphic breast phantom (Institute of Medical Physics and Radiation Protection, Technische Hochschule Mittelhessen, Germany) were imaged on an Amulet Innovality (FujiFilm, Tokyo, Japan) mammographic X-ray unit. Because the complexities of the structures of these three phantoms were different and not optimized to match the characteristics of real mammographic images, image processing was not applied in this study. In addition, real mammograms were acquired on the same system. Regions of interest (ROIs) were extracted from each image. In half of the ROIs, a 0.25-mm diameter disk was inserted at four different contrast levels to represent a calcification-like object. Each ROI was then modified, so four image qualities relevant for mammography were simulated. The signal-present and signal-absent ROIs were evaluated by a non-pre-whitening model observer with eye filter (NPWE) and a channelized Hotelling observer (CHO) using dense difference of Gaussian channels. The ROIs were also evaluated by human observers in a two alternative forced choice experiment. Detectability results for the human and model observer experiments were correlated using a mixed-effect regression model. Threshold disk contrasts for human and predicted human observer performance based on the NPWE MO and CHO were estimated.

RESULTS

Global trends in threshold contrast were similar for the different background structures, but absolute contrast threshold levels differed. Contrast thresholds tended to be lower in ROIs from simple phantoms compared with ROIs from real mammographic images. The correlation between human and model observer performance was not affected by the range of image quality levels studied.

CONCLUSIONS

The correlation between human and model observer performance does not depend on image quality. This is a promising outcome for the use of model observers in image quality analysis and allows for subsequent research toward the development of MO-based quality control procedures and guidelines.

A Monte Carlo model for mean glandular dose evaluation in spot compression mammography

PURPOSE

To characterize the dependence of normalized glandular dose (DgN) on various breast model and image acquisition parameters during spot compression mammography and other partial breast irradiation conditions, and evaluate alternative previously proposed dose-related metrics for this breast imaging modality.

METHODS

Using Monte Carlo simulations with both simple homogeneous breast models and patient-specific breasts, three different dose-related metrics for spot compression mammography were compared: the standard DgN, the normalized glandular dose to only the directly irradiated portion of the breast (DgNv), and the DgN obtained by the product of the DgN for full field irradiation and the ratio of the mid-height area of the irradiated breast to the entire breast area (DgNM ). How these metrics vary with field-of-view size, spot area thickness, x-ray energy, spot area and position, breast shape and size, and system geometry was characterized for the simple breast model and a comparison of the simple model results to those with patient-specific breasts was also performed.

RESULTS

The DgN in spot compression mammography can vary considerably with breast area. However, the difference in breast thickness between the spot compressed area and the uncompressed area does not introduce a variation in DgN. As long as the spot compressed area is completely within the breast area and only the compressed breast portion is directly irradiated, its position and size does not introduce a variation in DgN for the homogeneous breast model. As expected, DgN is lower than DgNv for all partial breast irradiation areas, especially when considering spot compression areas within the clinically used range. DgNM underestimates DgN by 6.7% for a W/Rh spectrum at 28 kVp and for a 9 × 9 cm2 compression paddle.

CONCLUSION

As part of the development of a new breast dosimetry model, a task undertaken by the American Association of Physicists in Medicine and the European Federation of Organizations of Medical Physics, these results provide insight on how DgN and two alternative dose metrics behave with various image acquisition and model parameters.

X-ray scatter correction for dedicated cone beam breast CT using a forward-projection model

PURPOSE

The quality of dedicated cone-beam breast CT (CBBCT) imaging is fundamentally limited by x-ray scatter contamination due to the large irradiation volume. In this paper, we propose a scatter correction method for CBBCT using a novel forward-projection model with high correction efficacy and reliability.

METHOD

We first coarsely segment the uncorrected, first-pass, reconstructed CBBCT images into binary-object maps and assign the segmented fibroglandular and adipose tissue with the correct attenuation coefficients based on the mean x-ray energy. The modified CBBCT are treated as the prior images toward scatter correction. Primary signals are first estimated via forward projection on the modified CBBCT. To avoid errors caused by inaccurate segmentation, only sparse samples of estimated primary are selected for scatter estimation. A Fourier-Transform based algorithm, herein referred to as local filtration hereafter, is developed to efficiently estimate the global scatter distribution on the detector. The scatter-corrected images are obtained by removing the estimated scatter distribution from measured projection data.

RESULTS

We evaluate the method performance on six patients with different breast sizes and shapes representing the general population. The results show that the proposed method effectively reduces the image spatial non-uniformity from 8.27 to 1.91% for coronal views and from 6.50 to 3.00% for sagittal views. The contrast-to-deviation ratio is improved by an average factor of 1.41. Comparisons on the image details reveal that the proposed scatter correction successfully preserves fine structures of fibroglandular tissues that are lost in the segmentation process.

CONCLUSION

We propose a highly practical and efficient scatter correction algorithm for CBBCT via a forward-projection model. The method is attractive in clinical CBBCT imaging as it is readily implementable on a clinical system without modifications in current imaging protocols or system hardware.

A supervoxel-based segmentation method for prostate MR images

PURPOSE

Segmentation of the prostate on MR images has many applications in prostate cancer management. In this work, we propose a supervoxel-based segmentation method for prostate MR images.

METHODS

A supervoxel is a set of pixels that have similar intensities, locations, and textures in a 3D image volume. The prostate segmentation problem is considered as assigning a binary label to each supervoxel, which is either the prostate or background. A supervoxel-based energy function with data and smoothness terms is used to model the label. The data term estimates the likelihood of a supervoxel belonging to the prostate by using a supervoxel-based shape feature. The geometric relationship between two neighboring supervoxels is used to build the smoothness term. The 3D graph cut is used to minimize the energy function to get the labels of the supervoxels, which yields the prostate segmentation. A 3D active contour model is then used to get a smooth surface by using the output of the graph cut as an initialization. The performance of the proposed algorithm was evaluated on 30 in-house MR image data and PROMISE12 dataset.

RESULTS

The mean Dice similarity coefficients are 87.2 ± 2.3% and 88.2 ± 2.8% for our 30 in-house MR volumes and the PROMISE12 dataset, respectively. The proposed segmentation method yields a satisfactory result for prostate MR images.

CONCLUSION

The proposed supervoxel-based method can accurately segment prostate MR images and can have a variety of application in prostate cancer diagnosis and therapy.

Deep Learning based Classification for Head and Neck Cancer Detection with Hyperspectral Imaging in an Animal Model

Hyperspectral imaging (HSI) is an emerging imaging modality that can provide a noninvasive tool for cancer detection and image-guided surgery. HSI acquires high-resolution images at hundreds of spectral bands, providing big data to differentiating different types of tissue. We proposed a deep learning based method for the detection of head and neck cancer with hyperspectral images. Since the deep learning algorithm can learn the feature hierarchically, the learned features are more discriminative and concise than the handcrafted features. In this study, we adopt convolutional neural networks (CNN) to learn the deep feature of pixels for classifying each pixel into tumor or normal tissue. We evaluated our proposed classification method on the dataset containing hyperspectral images from 12 tumor-bearing mice. Experimental results show that our method achieved an average accuracy of 91.36%. The preliminary study demonstrated that our deep learning method can be applied to hyperspectral images for detecting head and neck tumors in animal models.

Automatic segmentation of the prostate on CT images using deep learning and multi-atlas fusion

Automatic segmentation of the prostate on CT images has many applications in prostate cancer diagnosis and therapy. However, prostate CT image segmentation is challenging because of the low contrast of soft tissue on CT images. In this paper, we propose an automatic segmentation method by combining a deep learning method and multi-atlas refinement. First, instead of segmenting the whole image, we extract the region of interesting (ROI) to delete irrelevant regions. Then, we use the convolutional neural networks (CNN) to learn the deep features for distinguishing the prostate pixels from the non-prostate pixels in order to obtain the preliminary segmentation results. CNN can automatically learn the deep features adapting to the data, which are different from some handcrafted features. Finally, we select some similar atlases to refine the initial segmentation results. The proposed method has been evaluated on a dataset of 92 prostate CT images. Experimental results show that our method achieved a Dice similarity coefficient of 86.80% as compared to the manual segmentation. The deep learning based method can provide a useful tool for automatic segmentation of the prostate on CT images and thus can have a variety of clinical applications.

A Patch-based CBCT Scatter Artifact Correction Using Prior CT

We have developed a novel patch-based cone beam CT (CBCT) artifact correction method based on prior CT images. First, we used the image registration to align the planning CT with the CBCT to reduce the geometry difference between the two images. Then, we brought the planning CT-based prior information into the Bayesian deconvolution framework to perform the CBCT scatter artifact correction based on patch-wise nonlocal mean strategy. We evaluated the proposed correction method using a Catphan phantom with multiple inserts based on contrast-to-noise ratios (CNR) and signal-to-noise ratios (SNR), and the image spatial non-uniformity (ISN). All values of CNR SNR and ISN in the corrected CBCT image were much closer to those in the planning CT images. The results demonstrated that the proposed CT-guided correction method could significantly reduce scatter artifacts and improve the image quality. This method has great potential to correct CBCT images allowing its use in adaptive radiotherapy.

Learning with distribution of optimized features for recognizing common CT imaging signs of lung diseases

Common CT imaging signs of lung diseases (CISLs) are defined as the imaging signs that frequently appear in lung CT images from patients. CISLs play important roles in the diagnosis of lung diseases. This paper proposes a novel learning method, namely learning with distribution of optimized feature (DOF), to effectively recognize the characteristics of CISLs. We improve the classification performance by learning the optimized features under different distributions. Specifically, we adopt the minimum spanning tree algorithm to capture the relationship between features and discriminant ability of features for selecting the most important features. To overcome the problem of various distributions in one CISL, we propose a hierarchical learning method. First, we use an unsupervised learning method to cluster samples into groups based on their distribution. Second, in each group, we use a supervised learning method to train a model based on their categories of CISLs. Finally, we obtain multiple classification decisions from multiple trained models and use majority voting to achieve the final decision. The proposed approach has been implemented on a set of 511 samples captured from human lung CT images and achieves a classification accuracy of 91.96%. The proposed DOF method is effective and can provide a useful tool for computer-aided diagnosis of lung diseases on CT images.

Population of 224 realistic human subject-based computational breast phantoms

PURPOSE

To create a database of highly realistic and anatomically variable 3D virtual breast phantoms based on dedicated breast computed tomography (bCT) data.

METHODS

A tissue classification and segmentation algorithm was used to create realistic and detailed 3D computational breast phantoms based on 230 + dedicated bCT datasets from normal human subjects. The breast volume was identified using a coarse three-class fuzzy C-means segmentation algorithm which accounted for and removed motion blur at the breast periphery. Noise in the bCT data was reduced through application of a postreconstruction 3D bilateral filter. A 3D adipose nonuniformity (bias field) correction was then applied followed by glandular segmentation using a 3D bias-corrected fuzzy C-means algorithm. Multiple tissue classes were defined including skin, adipose, and several fractional glandular densities. Following segmentation, a skin mask was produced which preserved the interdigitated skin, adipose, and glandular boundaries of the skin interior. Finally, surface modeling was used to produce digital phantoms with methods complementary to the XCAT suite of digital human phantoms.

RESULTS

After rejecting some datasets due to artifacts, 224 virtual breast phantoms were created which emulate the complex breast parenchyma of actual human subjects. The volume breast density (with skin) ranged from 5.5% to 66.3% with a mean value of 25.3% ± 13.2%. Breast volumes ranged from 25.0 to 2099.6 ml with a mean value of 716.3 ± 386.5 ml. Three breast phantoms were selected for imaging with digital compression (using finite element modeling) and simple ray-tracing, and the results show promise in their potential to produce realistic simulated mammograms.

CONCLUSIONS

This work provides a new population of 224 breast phantoms based on in vivo bCT data for imaging research. Compared to previous studies based on only a few prototype cases, this dataset provides a rich source of new cases spanning a wide range of breast types, volumes, densities, and parenchymal patterns.

Dosimetry in x-ray-based breast imaging

The estimation of the mean glandular dose to the breast (MGD) for x-ray based imaging modalities forms an essential part of quality control and is needed for risk estimation and for system design and optimisation. This review considers the development of methods for estimating the MGD for mammography, digital breast tomosynthesis (DBT) and dedicated breast CT (DBCT). Almost all of the methodology used employs Monte Carlo calculated conversion factors to relate the measurable quantity, generally the incident air kerma, to the MGD. After a review of the size and composition of the female breast, the various mathematical models used are discussed, with particular emphasis on models for mammography. These range from simple geometrical shapes, to the more recent complex models based on patient DBCT examinations. The possibility of patient-specific dose estimates is considered as well as special diagnostic views and the effect of breast implants. Calculations using the complex models show that the MGD for mammography is overestimated by about 30% when the simple models are used. The design and uses of breast-simulating test phantoms for measuring incident air kerma are outlined and comparisons made between patient and phantom-based dose estimates. The most widely used national and international dosimetry protocols for mammography are based on different simple geometrical models of the breast, and harmonisation of these protocols using more complex breast models is desirable.

Library based x-ray scatter correction for dedicated cone beam breast CT

PURPOSE

The image quality of dedicated cone beam breast CT (CBBCT) is limited by substantial scatter contamination, resulting in cupping artifacts and contrast-loss in reconstructed images. Such effects obscure the visibility of soft-tissue lesions and calcifications, which hinders breast cancer detection and diagnosis. In this work, we propose a library-based software approach to suppress scatter on CBBCT images with high efficiency, accuracy, and reliability.

METHODS

The authors precompute a scatter library on simplified breast models with different sizes using the geant4-based Monte Carlo (MC) toolkit. The breast is approximated as a semiellipsoid with homogeneous glandular/adipose tissue mixture. For scatter correction on real clinical data, the authors estimate the breast size from a first-pass breast CT reconstruction and then select the corresponding scatter distribution from the library. The selected scatter distribution from simplified breast models is spatially translated to match the projection data from the clinical scan and is subtracted from the measured projection for effective scatter correction. The method performance was evaluated using 15 sets of patient data, with a wide range of breast sizes representing about 95% of general population. Spatial nonuniformity (SNU) and contrast to signal deviation ratio (CDR) were used as metrics for evaluation.

RESULTS

Since the time-consuming MC simulation for library generation is precomputed, the authors' method efficiently corrects for scatter with minimal processing time. Furthermore, the authors find that a scatter library on a simple breast model with only one input parameter, i.e., the breast diameter, sufficiently guarantees improvements in SNU and CDR. For the 15 clinical datasets, the authors' method reduces the average SNU from 7.14% to 2.47% in coronal views and from 10.14% to 3.02% in sagittal views. On average, the CDR is improved by a factor of 1.49 in coronal views and 2.12 in sagittal views.

CONCLUSIONS

The library-based scatter correction does not require increase in radiation dose or hardware modifications, and it improves over the existing methods on implementation simplicity and computational efficiency. As demonstrated through patient studies, the authors' approach is effective and stable, and is therefore clinically attractive for CBBCT imaging.

An energy minimization method for the correction of cupping artifacts in cone-beam CT

The purpose of this study was to reduce cupping artifacts and improve quantitative accuracy of the images in cone-beam CT (CBCT). An energy minimization method (EMM) is proposed to reduce cupping artifacts in reconstructed image of the CBCT. The cupping artifacts are iteratively optimized by using efficient matrix computations, which are verified to be numerically stable by matrix analysis. Moreover, the energy in our formulation is convex in each of its variables, which brings the robustness of the proposed energy minimization algorithm. The cupping artifacts are estimated as a result of minimizing this energy. The results indicate that proposed algorithm is effective for reducing the cupping artifacts and preserving the quality of the reconstructed image. The proposed method focuses on the reconstructed image without requiring any additional physical equipment; it is easily implemented and provides cupping correction using a single scan acquisition. The experimental results demonstrate that this method can successfully reduce the magnitude of cupping artifacts. The correction algorithm reported here may improve the uniformity of the reconstructed images, thus assisting the development of perfect volume visualization and threshold-based visualization techniques for reconstructed images.

A level set method for cupping artifact correction in cone-beam CT

PURPOSE

To reduce cupping artifacts and improve the contrast-to-noise ratio in cone-beam computed tomography (CBCT).

METHODS

A level set method is proposed to reduce cupping artifacts in the reconstructed image of CBCT. The authors derive a local intensity clustering property of the CBCT image and define a local clustering criterion function of the image intensities in a neighborhood of each point. This criterion function defines an energy in terms of the level set functions, which represent a segmentation result and the cupping artifacts. The cupping artifacts are estimated as a result of minimizing this energy.

RESULTS

The cupping artifacts in CBCT are reduced by an average of 90%. The results indicate that the level set-based algorithm is practical and effective for reducing the cupping artifacts and preserving the quality of the reconstructed image.

CONCLUSIONS

The proposed method focuses on the reconstructed image without requiring any additional physical equipment, is easily implemented, and provides cupping correction through a single-scan acquisition. The experimental results demonstrate that the proposed method successfully reduces the cupping artifacts.

A minimum spanning forest based classification method for dedicated breast CT images

PURPOSE

To develop and test an automated algorithm to classify different types of tissue in dedicated breast CT images.

METHODS

Images of a single breast of five different patients were acquired with a dedicated breast CT clinical prototype. The breast CT images were processed by a multiscale bilateral filter to reduce noise while keeping edge information and were corrected to overcome cupping artifacts. As skin and glandular tissue have similar CT values on breast CT images, morphologic processing is used to identify the skin based on its position information. A support vector machine (SVM) is trained and the resulting model used to create a pixelwise classification map of fat and glandular tissue. By combining the results of the skin mask with the SVM results, the breast tissue is classified as skin, fat, and glandular tissue. This map is then used to identify markers for a minimum spanning forest that is grown to segment the image using spatial and intensity information. To evaluate the authors' classification method, they use DICE overlap ratios to compare the results of the automated classification to those obtained by manual segmentation on five patient images.

RESULTS

Comparison between the automatic and the manual segmentation shows that the minimum spanning forest based classification method was able to successfully classify dedicated breast CT image with average DICE ratios of 96.9%, 89.8%, and 89.5% for fat, glandular, and skin tissue, respectively.

CONCLUSIONS

A 2D minimum spanning forest based classification method was proposed and evaluated for classifying the fat, skin, and glandular tissue in dedicated breast CT images. The classification method can be used for dense breast tissue quantification, radiation dose assessment, and other applications in breast imaging.

A general method for cupping artifact correction of cone-beam breast computed tomography images

PURPOSE

Cone-beam breast computed tomography (CBBCT), a promising breast cancer diagnostic technique, has been under investigation for the past decade. However, owing to scattered radiation and beam hardening, CT numbers are not uniform on CBBCT images. This is known as cupping artifact, and it presents an obstacle for threshold-based volume segmentation. In this study, we proposed a general post-reconstruction method for cupping artifact correction.

METHODS

There were four steps in the proposed method. First, three types of local region histogram peaks were calculated: adipose peaks with low CT numbers, glandular peaks with high CT numbers, and unidentified peaks. Second, a linear discriminant analysis classifier, which was trained by identified adipose and glandular peaks, was employed to identify the unidentified peaks as adipose or glandular peaks. Third, adipose background signal profile was fitted according to the adipose peaks using the least squares method. Finally, the adipose background signal profile was subtracted from original image to obtain cupping corrected image

RESULTS

In experimental study, standard deviation of adipose tissue CT numbers was obviously reduced and the CT numbers were more uniform after cupping correction by proposed method; in simulation study, root-mean-square errors were significantly reduced for both symmetric and asymmetric cupping artifacts, indicating that the proposed method was effective to both artifacts.

CONCLUSIONS

A general method without a circularly symmetric assumption was proposed to correct cupping artifacts in CBBCT images for breast. It may be properly applied to images of real patient breasts with natural pendent geometry.

Breast density evaluation using spectral mammography, radiologist reader assessment, and segmentation techniques: a retrospective study based on left and right breast comparison

RATIONALE AND OBJECTIVES

The purpose of this study was to compare the precision of mammographic breast density measurement using radiologist reader assessment, histogram threshold segmentation, fuzzy C-mean segmentation, and spectral material decomposition.

MATERIALS AND METHODS

Spectral mammography images from a total of 92 consecutive asymptomatic women (aged 50-69 years) who presented for annual screening mammography were retrospectively analyzed for this study. Breast density was estimated using 10 radiologist reader assessment, standard histogram thresholding, fuzzy C-mean algorithm, and spectral material decomposition. The breast density correlation between left and right breasts was used to assess the precision of these techniques to measure breast composition relative to dual-energy material decomposition.

RESULTS

In comparison to the other techniques, the results of breast density measurements using dual-energy material decomposition showed the highest correlation. The relative standard error of estimate for breast density measurements from left and right breasts using radiologist reader assessment, standard histogram thresholding, fuzzy C-mean algorithm, and dual-energy material decomposition was calculated to be 1.95, 2.87, 2.07, and 1.00, respectively.

CONCLUSIONS

The results indicate that the precision of dual-energy material decomposition was approximately factor of two higher than the other techniques with regard to better correlation of breast density measurements from right and left breasts.

3D in vivo imaging of rat hearts by high frequency ultrasound and its application in myofiber orientation wrapping

Cardiac ultrasound plays an important role in the imaging of hearts in basic cardiovascular research and clinical examinations. 3D ultrasound imaging can provide the geometry or motion information of the heart. Especially, the wrapping of cardiac fiber orientations to the ultrasound volume could supply useful information on the stress distributions and electric action spreading. However, how to acquire 3D ultrasound volumes of the heart of small animals in vivo for cardiac fiber wrapping is still a challenging problem. In this study, we provide an approach to acquire 3D ultrasound volumes of the rat hearts in vivo. The comparison between both in vivo and ex vivo geometries indicated 90.1% Dice similarity. In this preliminary study, the evaluations of the cardiac fiber orientation wrapping errors were 24.7° for the acute angle error and were 22.4° for the inclination angle error. This 3D ultrasound imaging and fiber orientation estimation technique have potential applications in cardiac imaging.

Dedicated Breast CT: Feasibility for Monitoring Neoadjuvant Chemotherapy Treatment

Objectives:

In this prospective pilot study, the feasibility of non-contrast dedicated breast computed tomography (bCT) to determine primary tumor volume and monitor its changes during neoadjuvant chemotherapy (NAC) treatment was investigated.

Materials and Methods:

Eleven women who underwent NAC were imaged with a clinical prototype dedicated bCT system at three time points – pre-, mid-, and post-treatment. The study radiologist marked the boundary of the primary tumor from which the tumor volume was quantified. An automated algorithm was developed to quantify the primary tumor volume for comparison with radiologist's segmentation. The correlation between pre-treatment tumor volumes from bCT and MRI, and the correlation and concordance in tumor size between post-treatment bCT and pathology were determined.

Results:

Tumor volumes from automated and radiologist's segmentations were correlated (Pearson's r = 0.935, P < 0.001) and were not different over all time points [P = 0.808, repeated measures analysis of variance (ANOVA)]. Pre-treatment tumor volumes from MRI and bCT were correlated (r = 0.905, P < 0.001). Tumor size from post-treatment bCT was correlated with pathology (r = 0.987, P = 0.002) for invasive ductal carcinoma larger than 5 mm and the maximum difference in tumor size was 0.57 cm. The presence of biopsy clip (3 mm) limited the ability to accurately measure tumors smaller than 5 mm. All study participants were pathologically assessed to be responders, with three subjects experiencing complete pathologic response for invasive cancer and the reminder experiencing partial response. Compared to pre-treatment tumor volume, there was a statistically significant (P = 0.0003, paired t-test) reduction in tumor volume at mid-treatment observed with bCT, with an average tumor volume reduction of 47%.

Conclusions:

This pilot study suggests that dedicated non-contrast bCT has the potential to serve as an expedient imaging tool for monitoring tumor volume changes during NAC. Larger studies are needed in future.

Breast tomosynthesis with monochromatic beams: a feasibility study using Monte Carlo simulations

The aim of this study is to investigate the impact on image quality of using monochromatic beams for lower dose breast tomosynthesis (BT). For this purpose, modeling and simulation of BT and mammography imaging processes have been performed using two x-ray beams: one at 28 kVp and a monochromatic one at 19 keV at different entrance surface air kerma ranging between 0.16 and 5.5 mGy. Two 4 cm thick computational breast models, in a compressed state, were used: one simple homogeneous and one heterogeneous based on CT breast images, with compositions of 50% glandular-50% adipose and 40% glandular-60% adipose tissues by weight, respectively. Modeled lesions, representing masses and calcifications, were inserted within these breast phantoms. X-ray transport in the breast models was simulated with previously developed and validated Monte Carlo application. Results showed that, for the same incident photon fluence, the use of the monochromatic beam in BT resulted in higher image quality compared to the one using polychromatic acquisition, especially in terms of contrast. For the homogenous phantom, the improvement ranged between 15% and 22% for calcifications and masses, respectively, while for the heterogeneous one this improvement was in the order of 33% for the masses and 17% for the calcifications. For different exposures, comparable image quality in terms of signal-difference-to-noise ratio and higher contrast for all features was obtained when using a monochromatic 19 keV beam at a lower mean glandular dose, compared to the polychromatic one. Monochromatic images also provide better detail and, in combination with BT, can lead to substantial improvement in visualization of features, and particularly better edge detection of low-contrast masses.

Power spectrum analysis of the x-ray scatter signal in mammography and breast tomosynthesis projections

PURPOSE

To analyze the frequency domain characteristics of the signal in mammography images and breast tomosynthesis projections with patient tissue texture due to detected scattered x-rays.

METHODS

Acquisitions of x-ray projection images of 19 different patient breasts were simulated using previously acquired volumetric patient images. Acquisition of these images was performed with a dedicated breast CT prototype system, and the images were classified into voxels representing skin, adipose, and glandular tissue with a previously validated automated algorithm. The classified three dimensional images then underwent simulated mechanical compression representing that which is performed during acquisition of mammography and breast tomosynthesis images. The acquisition of projection images of each patient breast was simulated using Monte Carlo methods with each simulation resulting in two images: one of the primary (non-scattered) signal and one of the scatter signal. To analyze the scatter signal for both mammography and breast tomosynthesis, two projections images of each patient breast were simulated, one with the x-ray source positioned at 0° (mammography and central tomosynthesis projection) and at 30° (wide tomosynthesis projection). The noise power spectra (NPS) for both the scatter signal alone and the total signal (primary + scatter) for all images were obtained and the combined results of all patients analyzed. The total NPS was fit to the expected power-law relationship NPS(f) = k/f β and the results were compared with those previously published on the power spectrum characteristics of mammographic texture. The scatter signal alone was analyzed qualitatively and a power-law fit was also performed.

RESULTS

The mammography and tomosynthesis projections of three patient breasts were too small to analyze, so a total of 16 patient breasts were analyzed. The values of β for the total signal of the 0° projections agreed well with previously published results. As expected, the scatter power spectrum reflected a fast drop-off with increasing spatial frequency, with a reduction of four orders of magnitude by 0.1 lp/mm. The β values for the scatter signal were 6.14 and 6.39 for the 0° and 30° projections, respectively.

CONCLUSIONS

Although the low-frequency characteristics of scatter in mammography and breast tomosynthesis were known, a quantitative analysis of the frequency domain characteristics of this signal was needed in order to optimize previously proposed software-based x-ray scatter reduction algorithms for these imaging modalities.

Breast Tissue Classification in Digital Tomosynthesis Images Based on Global Gradient Minimization and Texture Features

Digital breast tomosynthesis (DBT) is a pseudo-three-dimensional x-ray imaging modality proposed to decrease the effect of tissue superposition present in mammography, potentially resulting in an increase in clinical performance for the detection and diagnosis of breast cancer. Tissue classification in DBT images can be useful in risk assessment, computer-aided detection and radiation dosimetry, among other aspects. However, classifying breast tissue in DBT is a challenging problem because DBT images include complicated structures, image noise, and out-of-plane artifacts due to limited angular tomographic sampling. In this project, we propose an automatic method to classify fatty and glandular tissue in DBT images. First, the DBT images are pre-processed to enhance the tissue structures and to decrease image noise and artifacts. Second, a global smooth filter based on L0 gradient minimization is applied to eliminate detailed structures and enhance large-scale ones. Third, the similar structure regions are extracted and labeled by fuzzy C-means (FCM) classification. At the same time, the texture features are also calculated. Finally, each region is classified into different tissue types based on both intensity and texture features. The proposed method is validated using five patient DBT images using manual segmentation as the gold standard. The Dice scores and the confusion matrix are utilized to evaluate the classified results. The evaluation results demonstrated the feasibility of the proposed method for classifying breast glandular and fat tissue on DBT images.

A Minimum Spanning Forest Based Hyperspectral Image Classification Method for Cancerous Tissue Detection

Hyperspectral imaging is a developing modality for cancer detection. The rich information associated with hyperspectral images allow for the examination between cancerous and healthy tissue. This study focuses on a new method that incorporates support vector machines into a minimum spanning forest algorithm for differentiating cancerous tissue from normal tissue. Spectral information was gathered to test the algorithm. Animal experiments were performed and hyperspectral images were acquired from tumor-bearing mice. In vivo imaging experimental results demonstrate the applicability of the proposed classification method for cancer tissue classification on hyperspectral images.

Quantitative analysis for breast density estimation in low dose chest CT scans

A computational method was developed for the measurement of breast density using chest computed tomography (CT) images and the correlation between that and mammographic density. Sixty-nine asymptomatic Asian women (138 breasts) were studied. With the marked lung area and pectoralis muscle line in a template slice, demons algorithm was applied to the consecutive CT slices for automatically generating the defined breast area. The breast area was then analyzed using fuzzy c-mean clustering to separate fibroglandular tissue from fat tissues. The fibroglandular clusters obtained from all CT slices were summed then divided by the summation of the total breast area to calculate the percent density for CT. The results were compared with the density estimated from mammographic images. For CT breast density, the coefficient of variations of intraoperator and interoperator measurement were 3.00 % (0.59 %-8.52 %) and 3.09 % (0.20 %-6.98 %), respectively. Breast density measured from CT (22 ± 0.6 %) was lower than that of mammography (34 ± 1.9 %) with Pearson correlation coefficient of r=0.88. The results suggested that breast density measured from chest CT images correlated well with that from mammography. Reproducible 3D information on breast density can be obtained with the proposed CT-based quantification methods.

Multiscale segmentation of the skull in MR images for MRI-based attenuation correction of combined MR/PET

BACKGROUND AND OBJECTIVE

Combined magnetic resonance/positron emission tomography (MR/PET) is a relatively new, hybrid imaging modality. MR-based attenuation correction often requires segmentation of the bone on MR images. In this study, we present an automatic segmentation method for the skull on MR images for attenuation correction in brain MR/PET applications.

MATERIALS AND METHODS

Our method transforms T1-weighted MR images to the Radon domain and then detects the features of the skull image. In the Radon domain we use a bilateral filter to construct a multiscale image series. For the repeated convolution we increase the spatial smoothing in each scale and make the width of the spatial and range Gaussian function doubled in each scale. Two filters with different kernels along the vertical direction are applied along the scales from the coarse to fine levels. The results from a coarse scale give a mask for the next fine scale and supervise the segmentation in the next fine scale. The use of the multiscale bilateral filtering scheme is to improve the robustness of the method for noise MR images. After combining the two filtered sinograms, the reciprocal binary sinogram of the skull is obtained for the reconstruction of the skull image.

RESULTS

This method has been tested with brain phantom data, simulated brain data, and real MRI data. For real MRI data the Dice overlap ratios are 92.2%±1.9% between our segmentation and manual segmentation.

CONCLUSIONS

The multiscale segmentation method is robust and accurate and can be used for MRI-based attenuation correction in combined MR/PET.

Personalized estimates of radiation dose from dedicated breast CT in a diagnostic population and comparison with diagnostic mammography

This study retrospectively analyzed the mean glandular dose (MGD) to 133 breasts from 132 subjects, all women, who participated in a clinical trial evaluating dedicated breast CT in a diagnostic population. The clinical trial was conducted in adherence to a protocol approved by institutional review boards and the study participants provided written informed consent. Individual estimates of MGD to each breast from dedicated breast CT was obtained by combining x-ray beam characteristics with estimates of breast dimensions and fibroglandular fraction from volumetric breast CT images, and using normalized glandular dose coefficients. For each study participant and for the breast corresponding to that imaged with breast CT, an estimate of the MGD from diagnostic mammography (including supplemental views) was obtained from the DICOM image headers for comparison. This estimate uses normalized glandular dose coefficients corresponding to a breast with 50% fibroglandular weight fraction. The median fibroglandular weight fraction for the study cohort determined from volumetric breast CT images was 15%. Hence, the MGD from diagnostic mammography was corrected to be representative of the study cohort. Individualized estimates of MGD from breast CT ranged from 5.7 to 27.8 mGy. Corresponding to the breasts imaged with breast CT, the MGD from diagnostic mammography ranged from 2.6 to 31.6 mGy. The mean (± inter-breast SD) and the median MGD (mGy) from dedicated breast CT exam were 13.9 ± 4.6 and 12.6, respectively. For the corresponding breasts, the mean (± inter-breast SD) and the median MGD (mGy) from diagnostic mammography were 12.4 ± 6.3 and 11.1, respectively. Statistical analysis indicated that at the 0.05 level, the distributions of MGD from dedicated breast CT and diagnostic mammography were significantly different (Wilcoxon signed ranks test, p = 0.007). While the interquartile range and the range (maximum-minimum) of MGD from dedicated breast CT was lower than diagnostic mammography, the median MGD from dedicated breast CT was approximately 13.5% higher than that from diagnostic mammography. The MGD for breast CT is based on a 1.45 mm skin layer and that for diagnostic mammography is based on a 4 mm skin layer; thus, favoring a lower estimate for MGD from diagnostic mammography. The median MGD from dedicated breast CT corresponds to the median MGD from four to five diagnostic mammography views. In comparison, for the same 133 breasts, the mean and the median number of views per breast during diagnostic mammography were 4.53 and 4, respectively. Paired analysis showed that there was approximately equal likelihood of receiving lower MGD from either breast CT or diagnostic mammography. Future work will investigate methods to reduce and optimize radiation dose from dedicated breast CT.

Extracting Cardiac Myofiber Orientations from High Frequency Ultrasound Images

Cardiac myofiber plays an important role in stress mechanism during heart beating periods. The orientation of myofibers decides the effects of the stress distribution and the whole heart deformation. It is important to image and quantitatively extract these orientations for understanding the cardiac physiological and pathological mechanism and for diagnosis of chronic diseases. Ultrasound has been wildly used in cardiac diagnosis because of its ability of performing dynamic and noninvasive imaging and because of its low cost. An extraction method is proposed to automatically detect the cardiac myofiber orientations from high frequency ultrasound images. First, heart walls containing myofibers are imaged by B-mode high frequency (>20 MHz) ultrasound imaging. Second, myofiber orientations are extracted from ultrasound images using the proposed method that combines a nonlinear anisotropic diffusion filter, Canny edge detector, Hough transform, and K-means clustering. This method is validated by the results of ultrasound data from phantoms and pig hearts.

Automatic Segmentation of Right Ventricle on Ultrasound Images Using Sparse Matrix Transform and Level Set

An automatic framework is proposed to segment right ventricle on ultrasound images. This method can automatically segment both epicardial and endocardial boundaries from a continuous echocardiography series by combining sparse matrix transform (SMT), a training model, and a localized region based level set. First, the sparse matrix transform extracts main motion regions of myocardium as eigenimages by analyzing statistical information of these images. Second, a training model of right ventricle is registered to the extracted eigenimages in order to automatically detect the main location of the right ventricle and the corresponding transform relationship between the training model and the SMT-extracted results in the series. Third, the training model is then adjusted as an adapted initialization for the segmentation of each image in the series. Finally, based on the adapted initializations, a localized region based level set algorithm is applied to segment both epicardial and endocardial boundaries of the right ventricle from the whole series. Experimental results from real subject data validated the performance of the proposed framework in segmenting right ventricle from echocardiography. The mean Dice scores for both epicardial and endocardial boundaries are 89.1%±2.3% and 83.6±7.3%, respectively. The automatic segmentation method based on sparse matrix transform and level set can provide a useful tool for quantitative cardiac imaging.

A review of breast tomosynthesis. Part II. Image reconstruction, processing and analysis, and advanced applications

Many important post-acquisition aspects of breast tomosynthesis imaging can impact its clinical performance. Chief among them is the reconstruction algorithm that generates the representation of the three-dimensional breast volume from the acquired projections. But even after reconstruction, additional processes, such as artifact reduction algorithms, computer aided detection and diagnosis, among others, can also impact the performance of breast tomosynthesis in the clinical realm. In this two part paper, a review of breast tomosynthesis research is performed, with an emphasis on its medical physics aspects. In the companion paper, the first part of this review, the research performed relevant to the image acquisition process is examined. This second part will review the research on the post-acquisition aspects, including reconstruction, image processing, and analysis, as well as the advanced applications being investigated for breast tomosynthesis.