![\"Frans](\"https://cdn.uc.assets.prezly.com/a90b37bb-f26c-499d-9d61-a5c0d25442a9/-/crop/3693x3693/923,0/-/preview/-/scale_crop/128x128/center/-/format/auto/\"){kind=avatar}
\n Frans Steenhoudt\n Perscontact wetenschap en onderzoek\n \n
\nA study performed on 3504 individual breast microcalcifications (MCs) scanned using high-resolution 3D micro-CT reveals that the surrounding tissue microenvironment provides stronger diagnostic information than the calcified core alone for distinguishing benign from malignant breast lesions.
Where does the diagnostic information associated with breast MCs lie: in the calcium deposits themselves, or in the tissue immediately surrounding them? Researchers from Vrije Universiteit Brussel (VUB)-imec and UZ Brussel have addressed this question using high-resolution 3D micro-computed tomography (micro-CT) images.
The biological importance of this surrounding tissue environment is already known from histology and other research. The advance here is to isolate and compare its contribution to classification performance with that of the calcified core in micrometer-scale 3D X-ray images - for the first time, to the researchers’ knowledge.
Breast MCs are well-established radiological findings in mammography and digital breast tomosynthesis. Their morphology and distribution play an important role in assessing suspicious breast lesions and determining whether further diagnostic investigation is warranted.
Traditionally, image-based assessment has focused primarily on the MCs themselves, often examining their characteristics within clusters. What has remained unclear is how much diagnostic information can be obtained from their immediate tissue surroundings in high-resolution 3D X-ray images, independently of the calcified core. Conventional breast imaging does not provide the spatial resolution required to investigate these structures at the micrometer scale. Ex vivo micro-CT imaging offers an opportunity to study them in 3D.
Investigating 3504 individual microcalcifications in 3D
To address this question, the researchers analysed 94 paraffin-embedded breast biopsy blocks using high-resolution micro-CT, with a spatial resolution of approximately 8 μm.
They segmented 3504 individual MCs and extracted quantitative image measurements, known as radiomic features, to train machine-learning classifiers to predict whether the corresponding breast lesions were histopathologically benign or malignant.
Unlike previous approaches based on clusters of MCs at mammographic resolution, the study investigated individual MCs at the micrometer scale. The researchers conducted three complementary analyses to distinguish the contribution of the calcified core from that of the immediately surrounding tissue, while controlling for preprocessing and segmentation effects.
These included varying the image preprocessing window while keeping the segmentation mask constant, systematically expanding and contracting the MC masks, and extracting features exclusively from concentric shells inside and outside the calcified core. The last analysis was particularly important: by excluding the calcified core and analysing only the surrounding tissue, the researchers could directly test how much diagnostic information remained without the calcium deposit itself.
The surrounding tissue carries the strongest diagnostic signal
The results revealed a substantial improvement in classification performance when information from the MCs’ surrounding tissue was incorporated.
Increasing the preprocessing window around an unchanged calcification mask improved the area under the receiver operating characteristic curve (AUC) from 0.688 to 0.811.
Similarly, moderate dilation of the segmentation masks to include adjacent tissue increased AUC from approximately 0.689 to 0.813.
Most notably, classifiers trained exclusively on features extracted from concentric tissue shells maintained an AUC of approximately 0.81, comparable to models incorporating both the calcified core and the surrounding microenvironment.
These findings indicate that the MC-surrounding tissue is a dominant source of discriminative information in this experimental setting.
The results also highlight an important methodological consideration: the size of the preprocessing window can substantially affect radiomic feature extraction and downstream classification performance, even when the segmentation mask remains unchanged.
Rethinking imaging biomarkers in breast cancer
The findings suggest that quantitative imaging approaches focusing exclusively on the mineralised deposits may overlook valuable diagnostic information. Looking slightly beyond the calcified core, into its immediate surrounding tissue, can provide additional information about the underlying pathology.
By identifying the immediate tissue microenvironment as a particularly informative region, the study provides a rationale for developing radiomic biomarkers that explicitly characterise the interface between MCs and adjacent breast tissue.
The study is part of fundamental research using ex vivo micro-CT imaging of breast biopsy specimens. This approach enables detailed investigation of individual MCs and their immediate tissue surroundings at a spatial resolution that is not achievable with conventional clinical breast imaging. The findings therefore do not imply that the same structures can currently be resolved, or that equivalent classification performance can be achieved, using clinical breast imaging. Instead, they provide new insights into the spatial origin of diagnostic imaging information and may guide future research into quantitative breast imaging.
Further research will be needed to independently validate these findings and investigate which tissue structures and cell types contribute to the diagnostic signal through detailed histological analysis. This could help explain the biological basis of the imaging findings and inform the development of clinically applicable imaging biomarkers.
A multidisciplinary collaboration
The study brings together expertise in medical image analysis, artificial intelligence, radiology and pathology from VUB's Department of Electronics and Informatics (ETRO)-imec, VUB's Laboratory of Experimental Pathology, and the Departments of Pathology and Radiology at UZ Brussel.
Scientific publication: Micro-CT Radiomics Reveals the Microcalcification-Surrounding Tissue Interface as the Diagnostic Epicenter of Breast Microcalcifications (link)
Authors: Redona Brahimetaj, Jan Cornelis, Guillaume Courtoy, Wim Waelput, Inneke Willekens, Johan De Mey, Ramses Forsyth and Bart Jansen.
Redona Brahimetaj, Vrije Universiteit Brussel-imec: +32 46 55 97 448
\u200bredona.brahimetaj@vub.be
", "mainvisual": { "thumbnail": "https://cdn.uc.assets.prezly.com/e658c324-260d-475d-bb6e-a6bae70f0307/-/scale_crop/250x250/center/-/format/auto/", "large": "https://cdn.uc.assets.prezly.com/e658c324-260d-475d-bb6e-a6bae70f0307/-/preview/500x500/-/format/auto/", "original": "https://cdn.uc.assets.prezly.com/e658c324-260d-475d-bb6e-a6bae70f0307/" }, "header": { "large": "https://cdn.uc.assets.prezly.com/c55b53f8-bb91-476f-a730-702d2a22046b/-/format/auto/", "release": "https://cdn.uc.assets.prezly.com/c55b53f8-bb91-476f-a730-702d2a22046b/-/format/auto/" }, "contacts": [ { "name": "Frans Steenhoudt", "company": null, "description": "Perscontact wetenschap en onderzoek", "email": "frans.steenhoudt@vub.be", "website": "https://www.vub.be", "address": null, "telephone": null, "mobile": "+32 4 75 68 64 69", "twitter": null, "facebook": null, "linkedin": null, "instagram": null } ], "author": { "first_name": "Sam", "last_name": "Jaspers" }, "format_version": 6 }