Models of breast lesions based on three-dimensional X-ray breast images

Abstract

This paper presents a method for creation of computational models of breast lesions with irregular shapes from patient Digital Breast Tomosynthesis (DBT) images or breast cadavers and whole-body Computed Tomography (CT) images. The approach includes six basic steps: (a) normalization of the intensity of the tomographic images; (b) image noise reduction; (c) binarization of the lesion area, (d) application of morphological operations to further decrease the level of artefacts; (e) application of a region growing technique to segment the lesion; and (f) creation of a final 3D lesion model. The algorithm is semi-automatic as the initial selection of the region of the lesion and the seeds for the region growing are done interactively. A software tool, performing all of the required steps, was developed in MATLAB. The method was tested and evaluated by analysing anonymized sets of DBT patient images diagnosed with lesions. Experienced radiologists evaluated the segmentation of the tumours in the slices and the obtained 3D lesion shapes. They concluded for a quite satisfactory delineation of the lesions. In addition, for three DBT cases, a delineation of the tumours was performed independently by the radiologists. In all cases the abnormality volumes segmented by the proposed algorithm were smaller than those outlined by the experts. The calculated Dice similarity coefficients for algorithm-radiologist and radiologist-radiologist showed similar values. Another selected tumour case was introduced into a computational breast model to recursively assess the algorithm. The relative volume difference between the ground-truth tumour volume and the one obtained by applying the algorithm on the synthetic volume from the virtual DBT study is 5% which demonstrates the satisfactory performance of the proposed segmentation algorithm. The software tool we developed was used to create models of different breast abnormalities, which were then stored in a database for use by researchers working in this field. Anthropomorphic voxel breast phantoms with realistic tissue

Dataset of patient‐derived digital breast phantoms for in silico studies in breast computed tomography, digital breast tomosynthesis, and digital mammography

“…We used breast tumoral lesions from a previously generated database 23,24 for inserting lesions inside the digital breast phantoms. That database was produced within the framework of the European Union MaXIMA project (https:// maxima-tuv.eu/) and is comprised of 138 3D digital models of breast tumors.…”

“…That database was produced within the framework of the European Union MaXIMA project (https:// maxima-tuv.eu/) and is comprised of 138 3D digital models of breast tumors. As many as 80 out of 138 digital models of breast tumors were generated by means of a mathematical algorithm based on 3D random walk approach, 23,24 while 50 models out of 138 were derived by means of segmentation of lesions in DBT diagnostic images. The remaining eight tumor models were derived from 3D images of post-mortem specimens acquired via whole body CT scans.…”

“…The public dataset made available along with this paper does not contain any simulated lesion. These were available from a previous work, 23,24 and the interested reader can include lesions within the computational phantoms as needed.…”

“…A possible approach is to devise computational models of the lesion-free breast, then allowing for insertion of malignant details like microcalcifications and tumor masses. 23,24 Hardware versions of these software phantoms have been produced in the form of complex physical phantoms manufactured by 3D printing. [25][26][27][28][29][30] Digital phantoms have been developed using either an entirely computational approach, or as patient-like phantoms derived from the analysis of patients' breasts images obtained from clinical DM, DBT or BCT exams.…”

Radiochromic film dosimetry in synchrotron radiation breast computed tomography: a phantom study

“…We have recently adopted a new approach in which a real breast anatomy is realized via a 3D printed phantom with three components mimicking the skin, the adipose and the glandular tissue, respectively (Ivanov et al, 2018;Esposito et al, 2019). Such phantoms may incorporate breast tumor models as we investigated recently (Dukov et al, 2019;Bliznakova et al, 2019). Characterization of these new phantoms for absorption-based and phase-contrast-based SR breast CT is ongoing.…”