Search

With the goal of developing an accurate and fast image reconstruction algorithm for ultrasound computed tomography, we combine elements of model- and data-driven approaches and propose a learned method which addresses the disadvantages of both approaches. We design a deep neural network which accounts for a nonlinear forward operator and primal-dual algorithm by its inherent network architecture. The network is trained end-to-end, with ultrasound pressure field data as input to get directly an optimized reconstruction of speed of sound and attenuation images. The training and test data are based on a set of Optical and Acoustic Breast Phantom Database, where we use the image as ground truth and simulate pressure field data according to our forward model. Extensive experiments show that our method achieves significant improvements over state-of-the-art reconstruction methods in this field. Experiments show that the proposed algorithm improves the measures structural similarity measure (SSIM) from 0.74 to 0.95 and root mean squared error (RMSE) from 0.13 to 0.09 on average concerning the speed of sound reconstruction, while it improves the SSIM from 0.60 to 0.94 and RMSE from 0.24 to 0.10 on average in attenuation reconstruction.

A promising candidate for improved imaging of breast cancer is ultrasound computer tomography (USCT). Current experimental USCT systems are still focused in elevation dimension resulting in a large slice thickness, limited depth of field, loss of out-of-plane reflections, and a large number of movement steps to acquire a stack of images. 3D USCT emitting and receiving spherical wave fronts overcomes these limitations. We built an optimized 3D USCT, realizing for the first time the full benefits of a 3D system. The point spread function could be shown to be nearly isotropic in 3D, to have very low spatial variability and fit the predicted values. The contrast of the phantom images is very satisfactory in spite of imaging with a sparse aperture. The resolution and imaged details of the reflectivity reconstruction are comparable to a 3 T MRI volume. Important for the obtained resolution are the simultaneously obtained results of the transmission tomography. The KIT 3D USCT was then tested in a pilot study on ten patients. The primary goals of the pilot study were to test the USCT device, the data acquisition protocols, the image reconstruction methods and the image fusion techniques in a clinical environment. The study was conducted successfully; the data acquisition could be carried out for all patients with an average imaging time of six minutes per breast. The reconstructions provide promising images. Overlaid volumes of the modalities show qualitative and quantitative information at a glance. This paper gives a summary of the involved techniques, methods, and first results.

Ultrasound Computer Tomography (USCT) is a promising breast imaging modality under development. Comparison to a standard method like mammography is essential for further development. Due to significant differences in image dimensionality and compression state of the breast, correlating USCT images and X-ray mammograms is challenging. In this paper we present a 2D/3D registration method to improve the spatial correspondence and allow direct comparison of the images. It is based on biomechanical modeling of the breast and simulation of the mammographic compression. We investigate the effect of including patient-specific material parameters estimated automatically from USCT images. The method was systematically evaluated using numerical phantoms and in-vivo data. The average registration accuracy using the automated registration was 11.9mm. Based on the registered images a method for analysis of the diagnostic value of the USCT images was developed and initially applied to analyze sound speed and attenuation images based on X-ray mammograms as ground truth. Combining sound speed and attenuation allows differentiating lesions from surrounding tissue. Overlaying this information on mammograms, combines quantitative and morphological information for multimodal diagnosis.

Due to their different physical origin, X-ray mammography and Magnetic Resonance Imaging (MRI) provide complementary diagnostic information. However, the correlation of their images is challenging due to differences in dimensionality, patient positioning and compression state of the breast. Our automated registration takes over part of the correlation task. The registration method is based on a biomechanical finite element model, which is used to simulate mammographic compression. The deformed MRI volume can be compared directly with the corresponding mammogram. The registration accuracy is determined by a number of patient-specific parameters. We optimize these parameters--e.g. breast rotation--using image similarity measures. The method was evaluated on 79 datasets from clinical routine. The mean target registration error was 13.2mm in a fully automated setting. On basis of our results, we conclude that a completely automated registration of volume images with 2D mammograms is feasible. The registration accuracy is within the clinically relevant range and thus beneficial for multimodal diagnosis.

Rationale and objectives To evaluate the semi-automatic image registration accuracy of X-ray-mammography (XR-M) with high-resolution high-field (3.0 T) MR-mammography (MR-M) in an initial pilot study. Material and methods MR-M was acquired on a high-field clinical scanner at 3.0 T (T1-weighted 3D VIBE ± Gd). XR-M was obtained with state-of-the-art full-field digital systems. Seven patients with clearly delineable mass lesions >10 mm both in XR-M and MR-M were enrolled (exclusion criteria: previous breast surgery; surgical intervention between XR-M and MR-M). XR-M and MR-M were matched using a dedicated image-registration algorithm allowing semi-automatic non-linear deformation of MR-M based on finite-element modeling. To identify registration errors (RE) a virtual craniocaudal 2D mammogram was calculated by the software from MR-M (with and w/o Gadodiamide/Gd) and matched with corresponding XR-M. To quantify REs the geometric center of the lesions in the virtual vs. conventional mammogram were subtracted. The robustness of registration was quantified by registration of X-MRs to both MR-Ms with and w/o Gadodiamide. Results Image registration was performed successfully for all patients. Overall RE was 8.2 mm (1 min after Gd; confidence interval/CI: 2.0–14.4 mm, standard deviation/SD: 6.7 mm) vs. 8.9 mm (no Gd; CI: 4.0–13.9 mm, SD: 5.4 mm). The mean difference between pre- vs. post-contrast was 0.7 mm (SD: 1.9 mm). Conclusion Image registration of high-field 3.0 T MR-mammography with X-ray-mammography is feasible. For this study applying a high-resolution protocol at 3.0 T, the registration was robust and the overall registration error was sufficient for clinical application.

aDepartment of Neuroradiology, University of Erlangen-Nurnberg, Schwabachanlage 6, D-91054, Germany bDepartment of Radiology, Division of Molecular and Gender Imaging, Medical University Vienna, Wahringer Gurtel 18–20, A-1090, Vienna, Austria c Karlsruhe Institute of Technology (KIT), Institute for Data Processing and Electronics, Postfach 3640, D-76021 Karlsruhe, Germany d Institute of Diagnostic and Interventional Radiology I, Friedrich-Schiller-University Jena, Erlanger Allee 101, D-07740 Jena, Germany