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

Pre-whitened matched filter and convolutional neural network based model observer performance for mass lesion detection in non-contrast breast CT

BACKGROUND

Mathematical model observers have been shown to reasonably predict human observer performance and are useful when human observer studies are infeasible. Recently, convolutional neural networks (CNNs) have also been used as substitutes for human observers, and studies have shown their utility as an optimal observer. In this study, a CNN model observer is compared to the pre-whitened matched filter (PWMF) model observer in detecting simulated mass lesions inserted into 253 acquired breast computed tomography (bCT) images from patients imaged at our institution.

PURPOSE

To compare CNN and PWMF model observers for detecting signal-known-exactly (SKE) location-known-exactly (LKE) simulated lesions in bCT images with real anatomical backgrounds, and to use these model observers collectively to optimize parameters and understand trends in performance with breast CT.

METHODS

Spherical lesions with different diameters (1, 3, 5, 9 mm) were mathematically inserted into reconstructed patient bCT image data sets to mimic 3D mass lesions in the breast. 2D images were generated by extracting the center slice along the axial dimension or by slice averaging across adjacent slices to model thicker sections (0.4, 1.2, 2.0, 6.0, 12.4, 20.4 mm). The role of breast density was retrospectively studied using the range of breast densities intrinsic to the patient bCT data sets. In addition, mass lesions were mathematically inserted into Gaussian images matched to the mean and noise power spectrum of the bCT images to better understand the performance of the CNN in the context of a known ideal observer (the PWMF). The simulated Gaussian and bCT images were divided into training and testing data sets. Each training data set consisted of 91 600 images, and each testing data set consisted of 96 000 images. A CNN and PWMF was trained on the Gaussian training images, and a different CNN and PWMF was trained on the bCT training images. The trained model observers were tested, and receiver operating characteristic (ROC) curve analysis was used to evaluate detection performance. The area under the ROC curve (AUC) was the primary performance metric used to compare the model observers.

RESULTS

In the Gaussian background, the CNN performed essentially identically to the PWMF across lesion sizes and section thicknesses. In the bCT background, the CNN outperformed the PWMF across lesion size, breast density, and most section thicknesses. These findings suggest that there are higher-order features in bCT images that are harnessed by the CNN observer but are inaccessible to the PWMF.

CONCLUSIONS

The CNN performed equivalently to the ideal observer in Gaussian textures. In bCT background, the CNN captures more diagnostic information than the PWMF and may be a more pertinent observer when conducting optimal performance studies in breast CT images.

Model observer performance in contrast-enhanced lesions in breast CT: The influence of contrast concentration on detectability

BACKGROUND

The use of iodine-based contrast agent for better delineation of tumors in breast CT (bCT) has been shown to be compelling, similar to the tumor enhancement in contrast-enhanced breast MRI. Contrast-enhanced bCT (CE-bCT) is a relatively new tool, and a structured evaluation of different imaging parameters at play has yet to be conducted. In this investigation, data sets of acquired bCT images from 253 patients imaged at our institution were used in concert with simulated mathematically inserted spherical contrast-enhanced lesions to study the role of contrast enhancement on detectability.

PURPOSE

To quantitatively evaluate the improvement in lesion detectability due to contrast enhancement across lesion diameter, section thickness, view plane, and breast density using a pre-whitened matched filter (PWMF) model observer.

METHODS

The relationship between iodine concentration and Hounsfield units (HU) was measured using spectral modeling. The lesion enhancement from clinical CE-bCT images in 22 patients was evaluated, and the average contrast enhancement (ΔHU) was determined. Mathematically generated spherical mass lesions of varying diameters (1, 3, 5, 9, 11, 15 mm) and contrast enhancement levels (0, 0.25, 0.50, 0.75, 1) were inserted at random locations in 253 actual patient bCT datasets. Images with varying thicknesses (0.4-19.8 mm) were generated by slice averaging, and the role of view plane (coronal and axial planes) was studied. A PWMF was used to generate receiver operating characteristic (ROC) curves across parameters of lesion diameter, contrast enhancement, section thickness, view plane, and breast density. The area under the ROC curve (AUC) was used as the primary performance metric, generated from over 90,000 simulated lesions.

RESULTS

An average 20% improvement (ΔAUC = 0.1) in lesion detectability due to contrast enhancement was observed across lesion diameter, section thickness, breast density, and view plane. A larger improvement was observed when stratifying patients based on breast density. For patients with VGF ≤ 40%, detection performance improved up to 20% (until AUC →1), and for patients with denser breasts (VGF > 40%), detection performance improved more drastically, ranging from 20% to 80% for 1- and 5-mm lesions. For the 1 mm lesion, detection performance raised slightly at the 1.2 mm section thickness before falling off as thickness increased. For larger lesions, detection performance was generally unaffected as section thickness increased up until it reached 5.8 mm, where performance began to decline. Detection performance was higher in the axial plane compared to the coronal plane for smaller lesions and thicker sections.

CONCLUSIONS

For emerging diagnostic tools like CE-bCT, it is important to optimize imaging protocols for lesion detection. In this study, we found that intravenous contrast can be used to detect small lesions in dense breasts. Optimal section thickness for detectability has dependencies on breast density and lesion size, therefore, display thickness should be adjusted in real-time using display software. These findings may be useful for the development of CE-bCT as well as other x-ray-based breast imaging modalities.

Deep-learning-based denoising of X-ray differential phase and dark-field images

PURPOSE

Statistical photon noise has always been a common problem in X-ray multi-contrast imaging and significantly influenced the quality of retrieved differential phase and dark-field images. We intend to develop a deep learning-based denoising algorithm to reduce the noise of retrieved X-ray differential phase and dark-field images.

METHODS

A novel deep learning based image noise suppression algorithm (named DnCNN-P) is presented. We proposed two different denoising modes: Retrieval-Denoising mode (R-D mode) and Denoising-Retrieval mode (D-R mode). While the R-D mode denoises the retrieved images, the D-R mode denoises the raw phase stepping data. The two denoising modes are evaluated under different photon counts and visibilities.

RESULTS

Experimental results show that with the algorithm DnCNN-P used, the D-R mode always exhibits a better noise reduction under diverse experimental conditions, even in the case of a low photon count and/or a low visibility. With a detected photon count of 1800 and a visibility of 0.3, compared to the differential phase images without denoising, the standard deviation is reduced by 89.1% and 16.4% in the D-R and R-D modes. Compared to the dark-field images without denoising, the standard deviation is reduced by 83.7% and 12.6% in the D-R and R-D modes, respectively.

CONCLUSIONS

The novel supervised DnCNN-P algorithm can significantly reduce the noise in retrieved X-ray differential phase and dark-field images. We believe this novel algorithm can be a promising approach to improve the quality of X-ray differential phase and dark-field images, and therefore dose efficiency in future biomedical applications.

Hyperia: A novel methodology of developing anthropomorphic breast phantoms for X-ray imaging modalities - Part I: Concept and initial findings

PURPOSE

To introduce a novel methodology for developing anthropomorphic breast phantoms for use in X-ray-based imaging modalities.

METHODS

"Hyperization" is a quasi-stippling mapping operation in which regions of varying grayscale values in a 2D image are transformed into regions of varying holes on a surface. The holes can be cut or engraved on the sheet of paper using a high-resolution laser cutter/engraver. In hyperization, the main parameters are the size and the distance between the holes. Here, we introduce the concept and chronicle the development and characterization of a proof-of-concept prototype. In this study, we hypothesized that a resulting "Hyperia" phantom would be a realistic representative of a patient's breast tissue: it would exhibit similar X-ray properties and show textural complexities. We used breast computed tomography (bCT) images of real patients as the input models. Using a previously developed segmentation method, the input CT images were segmented into different tissue classes (skin, adipose, and fibroglandular). The segmented images were then "Hyperized". A series of Monte Carlo simulations were conducted to find the optimal hyperization parameters. Different laser cutter/engraver systems and substrate materials were explored to find a viable option for developing an entire Hyperia breast phantom. The resulting phantom was imaged on a prototype breast CT system, and the resulting images were evaluated based on physical properties and similarity to the original patient data.

RESULTS

The simulation results indicate close similarities - both in the distribution of different tissue types and the resulting CT numbers - between the patient bCT image and the bCT of the Hyperia phantom, regardless of the breast size and density: the Pearson correlation coefficient (ρ) ranged from 0.88 in a BIRADS A breast to 0.94 in BIRADS C and D breasts (ρ of 1.00 suggests perfect structural similarity), and the volumetric mean squared error ranged from 0.0033 (in BIRADS D breast) to 0.0059 (in BIRADS A), suggesting good agreement between the resulting CT numbers. For fabricating the slices, the office paper was found to be an optimal substrate material, with the Hyperization parameters of (α, β) = (0.200 mm, 0.400 mm).

CONCLUSION

A novel phantom can be used for X-ray-based breast cancer imaging systems. The main advantage is that only one material is used for creating a contrast between different tissue types in an image.

Prior information guided auto-contouring of breast gland for deformable image registration in postoperative breast cancer radiotherapy

Background

Contouring of breast gland in planning CT is important to postoperative radiotherapy of patients after breast conserving surgery (BCS). However, the contouring task is difficult because of the poorer contrast of breast gland in planning CT. To improve its efficiency and accuracy, prior information was introduced in a 3D U-Net model to predict the contour of breast gland automatically.

Methods

The preoperative CT was first aligned to the planning CT via affine registration. The resulting transform was then applied to the contour of breast gland in preoperative CT, and the corresponding contour in planning CT was obtained. This transformed contour was a preliminary estimation of breast gland in planning CT and was used as prior information in a 3D U-Net model to obtain a more accurate contour. For evaluation, the dice similarity coefficient (DSC) and Hausdorff distance (HD) were used to assess the deep learning (DL) model's prediction accuracy.

Results

The average DSC and HD of the prediction model were 0.775±0.065 and 44.979±20.565 for breast gland without the input of prior information, while the average values were 0.830±0.038 and 17.896±5.737 with the input of prior information (0.775 vs. 0.830, P=0.0014<0.05; 44.979 vs. 17.896, P=0.002<0.05).

Conclusions

The prediction accuracy was increased significantly with the introduction of prior information, which provided valuable geometrical distribution of target for model training. This method provides an effective way to identify low-contrast targets from surrounding tissues in CT and will be useful in other image modalities.

High-resolution μ CT imaging for characterizing microcalcification detection performance in breast CT

Purpose: To demonstrate the utility of high-resolution micro-computed tomography ( μ CT ) for determining ground-truth size and shape properties of calcium grains for evaluation of detection performance in breast CT (bCT). Approach: Calcium carbonate grains ( ∼ 200    μ m ) were suspended in 1% agar solution to emulate microcalcifications ( μ Calcs ) within a fibroglandular tissue background. Ground-truth imaging was performed on a commercial μ CT scanner and was used for assessing calcium-grain size and shape, and for generating μ Calc signal profiles. Calcium grains were placed within a realistic breast-shaped phantom and imaged on a prototype bCT system at 3- and 6-mGy mean glandular dose (MGD) levels, and the non-prewhitening detectability was assessed. Additionally, the μ CT -derived signal profiles were used in conjunction with the bCT system characterization (MTF and NPS) to obtain predictions of bCT detectability. Results: Estimated detectability of the calcium grains on the bCT system ranged from 2.5 to 10.6 for 3 mGy and from 3.8 to 15.3 for 6 mGy with large fractions of the grains meeting the Rose criterion for visibility. Segmentation of μ CT images based on morphological operations produced accurate results in terms of segmentation boundaries and segmented region size. A regression model linking bCT detectability to μ Calc parameters indicated significant effects of μ Calc size and vertical position within the breast phantom. Detectability using μ CT -derived detection templates and bCT statistical properties (MTF and NPS) were in good correspondence with those measured directly from bCT ( R 2 > 0.88 ). Conclusions: Parameters derived from μ CT ground-truth data were shown to produce useful characterizations of detectability when compared to estimates derived directly from bCT. Signal profiles derived from μ CT imaging can be used in conjunction with measured or hypothesized statistical properties to evaluate the performance of a system, or system component, that may not currently be available.

Computer-aided diagnosis of masses in breast computed tomography imaging: deep learning model with combined handcrafted and convolutional radiomic features

Purpose: A computer-aided diagnosis (CADx) system for breast masses is proposed, which incorporates both handcrafted and convolutional radiomic features embedded into a single deep learning model. Approach: The model combines handcrafted and convolutional radiomic signatures into a multi-view architecture, which retrieves three-dimensional (3D) image information by simultaneously processing multiple two-dimensional mass patches extracted along different planes through the 3D mass volume. Each patch is processed by a stream composed of two concatenated parallel branches: a multi-layer perceptron fed with automatically extracted handcrafted radiomic features, and a convolutional neural network, for which discriminant features are learned from the input patches. All streams are then concatenated together into a final architecture, where all network weights are shared and the learning occurs simultaneously for each stream and branch. The CADx system was developed and tested for diagnosis of breast masses ( N = 284 ) using image datasets acquired with independent dedicated breast computed tomography systems from two different institutions. The diagnostic classification performance of the CADx system was compared against other machine and deep learning architectures adopting handcrafted and convolutional approaches, and three board-certified breast radiologists. Results: On a test set of 82 masses (45 benign, 37 malignant), the proposed CADx system performed better than all other model architectures evaluated, with an increase in the area under the receiver operating characteristics curve (AUC) of 0.05 ± 0.02 , and achieving a final AUC of 0.947, outperforming the three radiologists ( AUC = 0.814 - 0.902 ). Conclusions: In conclusion, the system demonstrated its potential usefulness in breast cancer diagnosis by improving mass malignancy assessment.

Multi-marker quantitative radiomics for mass characterization in dedicated breast CT imaging

Purpose

To develop and evaluate the diagnostic performance of an algorithm for multi‐marker radiomic‐based classification of breast masses in dedicated breast computed tomography (bCT) images.

Methods

Over 1000 radiomic descriptors aimed at quantifying mass and border heterogeneity, morphology, and margin sharpness were developed and implemented. These included well‐established texture and shape feature descriptors, which were supplemented with additional approaches for contour irregularity quantification, spicule and lobe detection, characterization of degree of infiltration, and differences in peritumoral compartments. All descriptors were extracted from a training set of 202 bCT masses (133 benign and 69 malignant), and their individual diagnostic performance was investigated in terms of area under the receiver operating characteristics (ROC) curve (AUC) of single‐feature‐based linear discriminant analysis (LDA) classifiers. Subsequently, the most relevant descriptors were selected through a multiple‐step feature selection process (including stability analysis, statistical significance, evaluation of feature interaction, and dimensionality reduction), and used to develop a final LDA radiomic model for classification of benign and malignant masses, which was then tested on an independent test set of 82 cases (45 benign and 37 malignant).

Results

The majority of the individual radiomic descriptors showed, on the training set, an AUC value deriving from a linear decision boundary higher than 0.65, with the lower limit of the associated 95% confidence interval (C.I.) not overlapping with random chance (AUC = 0.5). The final LDA radiomic model resulted in a test set AUC of 0.90 (95% C.I. 0.80–0.96).

Conclusions

The proposed multi‐marker radiomic approach achieved high diagnostic accuracy in bCT mass classification, using a radiomic signature based on different feature types. While future studies with larger datasets are needed to further validate these results, quantitative radiomics applied to bCT shows potential to improve the breast cancer diagnosis pipeline.

Reduction of scatter in breast CT yields improved microcalcification visibility

The inadequate visibility of microcalcifications-small calcium deposits that cue radiologists to early stages of cancer-is a major limitation in current designs of dedicated breast computed tomography (bCT). This limitation has previously been attributed to the constituent components, spatial resolution, and utilized dose. Scattered radiation has been considered an occurrence with low-frequency impacts that can be compensated for in post-processing. We hypothesized, however, that the acquisition of scattered radiation has a far more detrimental impact on clinically relevant features than has previously been understood. Critically, acquisition of scatter leads to the reduced visibility of microcalcifications. This hypothesis was investigated and supported via mathematical derivations and simulation studies. We conducted a series of comparative studies in which four bCT systems were simulated under iso-dose and iso-resolution conditions, characterizing the dependencies of microcalcification contrast on accumulated scatter. Included among the simulated systems is a novel bCT design-narrow beam bCT (NB-bCT)-that captures nearly zero scatter. We find that current bCT systems suffer from significant levels of scatter. As validated in theory, depending on the system and size of microcalcifications, between 25% and over 70% of contrast resolution is lost due to scatter. The results in NB-bCT, however, provide evidence that by removing scatter build-up in projections, the contrast of microcalcifications in a bCT image is preserved, regardless of their size or location in the breast.

Deep learning object detection of maxillary cyst-like lesions on panoramic radiographs: preliminary study

OBJECTIVES

This study aimed to examine the performance of deep learning object detection technology for detecting and identifying maxillary cyst-like lesions on panoramic radiography.

METHODS

Altogether, 412 patients with maxillary cyst-like lesions (including several benign tumors) were enrolled. All panoramic radiographs were arbitrarily assigned to the training, testing 1, and testing 2 datasets of the study. The deep learning process of the training images and labels was performed for 1000 epochs using the DetectNet neural network. The testing 1 and testing 2 images were applied to the created learning model, and the detection performance was evaluated. For lesions that could be detected, the classification performance (sensitivity) for identifying radicular cysts or other lesions were examined.

RESULTS

The recall, precision, and F-1 score for detecting maxillary cysts were 74.6%/77.1%, 89.8%/90.0%, and 81.5%/83.1% for the testing 1/testing 2 datasets, respectively. The recall was higher in the anterior regions and for radicular cysts. The sensitivity was higher for identifying radicular cysts than for other lesions.

CONCLUSIONS

Using deep learning object detection technology, maxillary cyst-like lesions could be detected in approximately 75-77%.

Effects of kV, filtration, dose, and object size on soft tissue and iodine contrast in dedicated breast CT

PURPOSE

Clinical use of dedicated breast computed tomography (bCT) requires relatively short scan times necessitating systems with high frame rates. This in turn impacts the x-ray tube operating range. We characterize the effects of tube voltage, beam filtration, dose, and object size on contrast and noise properties related to soft tissue and iodine contrast agents as a way to optimize imaging protocols for soft tissue and iodine contrast at high frame rates.

METHODS

This study design uses the signal-difference-to-noise ratio (SDNR), noise-equivalent quanta (NEQ), and detectability (d´) as measures of imaging performance for a prototype breast CT scanner that utilizes a pulsed x-ray tube (with a 4 ms pulse width) at 43.5 fps acquisition rate. We assess a range of kV, filtration, breast phantom size, and mean glandular dose (MGD). Performance measures are estimated from images of adipose-equivalent breast phantoms machined to have a representative size and shape of small, medium, and large breasts. Water (glandular tissue equivalent) and iodine contrast (5 mg/ml) were used to fill two cylindrical wells in the phantoms.

RESULTS

Air kerma levels required for obtaining an MGD of 6 mGy ranged from 7.1 to 9.1 mGy and are reported across all kV, filtration, and breast phantom sizes. However, at 50 kV, the thick filters (0.3 mm of Cu or Gd) exceeded the maximum available mA of the x-ray generator, and hence, these conditions were excluded from subsequent analysis. There was a strong positive association between measurements of SDNR and d' (R²  > 0.97) within the range of parameters investigated in this work. A significant decrease in soft tissue SDNR was observed for increasing phantom size and increasing kV with a maximum SDNR at 50 kV with 0.2 mm Cu or 0.2 mm Gd filtration. For iodine contrast SDNR, a significant decrease was observed with increasing phantom size, but a decrease in SDNR for increasing kV was only observed for 70 kV (50 and 60 kV were not significantly different). Thicker Gd filtration (0.3 mm Gd) resulted in a significant increase in iodine SDNR and decrease in soft tissue SDNR but requires significantly more tube current to deliver the same MGD.

CONCLUSIONS

The choice of 60 kV with 0.2 mm Gd filtration provides a good trade-off for maximizing both soft tissue and iodine contrast. This scanning technique takes advantage of the ~50 keV Gd k-edge to produce contrast and can be achieved within operating range of the x-ray generator used in this work. Imaging at 60 kV allows for a greater range in dose delivered to the large breast sizes when uniform image quality is desired across all breast sizes. While imaging performance metrics (i.e., detectability index and SDNR) were shown to be strongly correlated, the methodologies presented in this work for the estimation of NEQ (and subsequently d') provides a meaningful description of the spatial resolution and noise characteristics of this prototype bCT system across a range of beam quality, dose, and object sizes.

A fluence modulation and scatter shielding apparatus for dedicated breast CT: Theory of operation

PURPOSE

To introduce an auxiliary apparatus of fluence modulation and scatter shielding for dedicated breast computed tomography (bCT) and the corresponding patient-specific method of image acquisition.

METHODS

The apparatus is composed of two assemblies, referred to herein as the "Dynamic Fluence Gate" (FG) and "Scatter Shield" (SS), that work in synchrony to form a narrow beam sweeping the entire fan angle coverage of the imaging system during a projection. The apparatus, as a whole, is referred to as FG-SS. FG and SS are pre-patient and post-patient units, respectively. Each is composed of a rotating drum, on top of which are installed two sheets of high x-ray attenuating material, a sensory system, and the constituent robotics. The sheets of each unit are positioned such that an opening - a window Fluence Modulation and Scatter Shielding is formed between them. The rotations of the drums and positioning of the sheets are synchronized and adjusted such that a line of sight is created between the source, FG window, the breast, and the SS window. With line of sight achieved, the narrow beam transitions from the source to the detector. The fluence of the narrow beam during a projection depends on the size, shape, and positioning of the breast. The FG-SS method of imaging is discussed mathematically and demonstrated using computer simulations. A series of Monte Carlo simulations are conducted to evaluate the performance of the system as relates to its impact on the imager's dynamic range, dose distribution to the breast, noise inhomogeneity in reconstructed images, and scatter buildup in projections within small, medium, and large breasts composed of homogeneous medium breast tissue.

RESULTS

Implementation of FG-SS results in near scatter-free projections, reduction in both dose and the required dynamic range of the imager, and equalization of the quantum noise distribution in the reconstructed image. Using the disclosed design, the dynamic range was reduced by factors ranging from 1.6 to 5.5 across the range of breast sizes studied. A reduction in the acquisition of the scattered rays, varying between the factors of 6.1 (in the small breast) and 9.8 (in that large breast) was achieved and consequently, shading artifacts were suppressed. Noise in the CT image was equalized by reducing the overall spatial variation from 29% to <5% in small breast and from 45% to 14% in the large breast. An overall reduction in deposited dose to the breast was achieved - between 26% and 39% depending on the breast size.

CONCLUSIONS

Utilization of the FG-SS apparatus and technique was demonstrated via simulations to result in: significant reductions in dose to the breast, reductions in scatter uptake in projections, reduced required dynamic range of the imager, and homogenizing of quantum noise throughout the reconstructed image.