Breast compression simulation using ICP-based B-spline deformation for correspondence analysis in mammography and MRI datasets

Abstract

Mammography is the most commonly used imaging modality in breast cancer screening and diagnosis. The analysis of 2D mammographic images can be difficult due to the projective nature of the imaging technique and poor contrast between tumorous and healthy fibro-glandular tissue. Contrast-enhanced magnetic resonance imaging (MRI) can overcome these disadvantages by providing a 3D dataset of the breast. The detection of corresponding image structures is challenging due to large breast deformations during the image acquisition. We present a method for analyzing 2D/3D intra-individual correspondences between mammography and MRI datasets. Therefore, an ICP-based B-spline registration is used to approximate the breast deformation differences. The resulting deformed MR image is projected onto the 2D plane to enable a comparison with the 2D mammogram. A first evaluation based on six mammograms revealed an average accuracy of 4.87 mm. In contrast to previous FEM-based approaches, we propose a fast and easy to implement 3D/3D-registration, for simulating the mammographic breast compression. PURPOSEMammography is the breast imaging modality that is used most commonly in cancer screening and diagnosis. It provides a two-dimensional projection image of the compressed breast (between two plates), characterized by a fine resolution ([< 0.1× < 0.1]mm 2 ). Due to the projective nature of this imaging technique and a poor contrast between tumorous and healthy fibro-glandular tissue, the analysis of mammographic images can be challenging. Contrast-enhanced magnetic resonance imaging (MRI) can be used additionally 1 to overcome this disadvantages by providing a three-dimensional image of the female breast with functional information, which is commonly used for tumour staging (resolution ≈ [1 × 1 × 2-3]mm 3 ). In contrast to mammography, MRI is acquired without compression of the breast.To analyze the combined information provided by the two imaging modalities, the examining radiologist has to localize corresponding structures in the mammogram and the MR image. This task is challenging due to (1) different dimensionality and spatial resolution, (2) different contrasts between tissue types and (3) different deformation of the breast during acquisition. The huge breast deformation differences make a localization of corresponding structures especially challenging.In this paper, we present a method to enable an analysis of 2D/3D intra-individual correspondences between mammography and MRI. For this purpose, an ICP-based B-spline registration has been developed within the context of this work to approximate the breast deformation. To apply a surface based registration, the MR surface information is extracted from the 3D MR image and the mammographic surface information is approximated by reconstructing the mammographic volume. The resulting deformed MR image is projected onto the 2D plane to enable a comparison with the mammogram.Previously presented approaches for correspondence analysis between MRI and mammography have, for example, used ...

Finite-element modeling of compression and gravity on a population of breast phantoms for multimodality imaging simulation

“…16 Had corresponding mammograms been available, an alternative method would have been to use an iterative surface based registration algorithm similar to Krüger et al to provide a mapping between the bCT data and a compressed shell developed from the mammogram. 40 It should also be noted that the desired thickness specified by the user may intentionally differ from the compressed thickness that may actually occur in a corresponding mammogram. This would allow a researcher to either include or exclude variability of breast thickness in their study design.…”

A review of biomechanically informed breast image registration

“…When these modalites are not co-registered at the point of acquisition, image registration has been used to fuse them. Examples include fusion of the functional information present in MRI with the x-ray attenuation representation of mammography, Behrenbruch et al (2003) and Krueger et al (2013), registration of MRI and ultrasound to guide biopsy, Causer et al (2008), and registration of MRI and PET-CT to aid breast tumour characterisation, Dmitriev et al (2013).…”

A compact breast shape acquisition system for improving diffuse optical tomography image reconstructions

“…When a 3-D imaging modality is not available, two-dimensional (2-D) mammography [19] has also been used to provide the shape information. In such case, a simple way to recover a 3-D breast surface is to extrude the 2-D breast contour along the compression axis [29, 30], or sweep the 2-D breast contour along the contour line extracted from an orthogonal view [31]. These methods either rely on assumptions about the 3-D location of certain features (e.g.…”