Is Digital Breast Tomosynthesis the Better Mammogram for Local Breast Cancer Staging?

Moy, Linda

doi:10.1148/radiol.2019190590

Cited by 2 publications

(1 citation statement)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As a pseudo-3D mammography technology, digital breast tomosynthesis (DBT) enables cross-sectional imaging and achieves better characterization and localization of lesions than 2D mammography (Moy 2019). DBT is associated with higher specificity and a greater proportion of breast cancers detected than 2D mammography alone (Conant et al 2019).…”

Section: Introductionmentioning

confidence: 99%

A collection input based support tensor machine for lesion malignancy classification in digital breast tomosynthesis

Yang¹,

Zhou

et al. 2019

Phys. Med. Biol.

View full text Add to dashboard Cite

Digital breast tomosynthesis (DBT) with improved lesion conspicuity and characterization has been adopted in screening practice. DBT-based diagnosis strongly depends on physicians’ experience, so an automatic lesion malignancy classification model using DBT could improve the consistency of diagnosis among different physicians. Tensor-based approaches that use the original imaging data as input have shown promising results for many classification tasks. However, DBT data are pseudo-3D volumetric imaging as the slice spacing of DBT is much coarser than that of the in-plane resolution. Thus, directly constructing DBT as the third-order tensor in a conventional tensor-based classifier with introducing additional information to the original DBT data along the slice-spacing dimension will lead to inconsistency across all three dimensions. To avoid such inconsistency, we introduce a collection input based support tensor machine (CISTM)-based classifier that uses the tensor collection as input for classifying lesion malignancy in DBT. In CISTM, instead of introducing the third dimension directly into the geometry construction, the third-dimension structural relationship is related by weight parameters in the decision function, which is dynamically and automatically constructed during the classifier training process and is more consistent with the pseudo-3D nature of DBT. We tested our method on a DBT dataset of 926 images among which 262 were malignant and 664 were benign. We compared our method with the latest tensor-based method, KSTM (kernelled support tensor machine), which does not consider the unique non-uniform resolution property of DBT. Experimental results illustrate that the CISTM-based classifier is effective for classifying breast lesion malignancy in DBT and that it outperforms the KSTM-based classifier.

show abstract

Section: Introductionmentioning

confidence: 99%

A collection input based support tensor machine for lesion malignancy classification in digital breast tomosynthesis

Yang¹,

Zhou

et al. 2019

Phys. Med. Biol.

View full text Add to dashboard Cite

show abstract

A risk prediction stratification for non-mass breast lesions, combining clinical characteristics and imaging features on ultrasound, mammography, and MRI

Xie,

Zhang

2024

Front. Oncol.

View full text Add to dashboard Cite

ObjectivesGiven the inevitable trend of domestic imaging center mergers and the current lack of comprehensive imaging evaluation guidelines for non-mass breast lesions, we have developed a novel BI-RADS risk prediction and stratification system for non-mass breast lesions that integrates clinical characteristics with imaging features from ultrasound, mammography, and MRI, with the aim of assisting clinicians in interpreting imaging reports.MethodsThis study enrolled 350 patients with non-mass breast lesions (NMLs), randomly assigning them to a training set of 245 cases (70%) and a test set of 105 cases (30%). Radiologists conducted comprehensive evaluations of the lesions using ultrasound, mammography, and MRI. Independent predictors were identified using LASSO logistic regression, and a predictive risk model was constructed using a nomogram generated with R software, with subsequent validation in both sets.ResultsLASSO logistic regression identified a set of independent predictors, encompassing age, clinical palpation hardness, distribution and morphology of calcifications, peripheral blood supply as depicted by color Doppler imaging, maximum lesion diameter, patterns of internal enhancement, distribution of non-mass lesions, time–intensity curve (TIC), and apparent diffusion coefficient (ADC) values. The predictive model achieved area under the curve (AUC) values of 0.873 for the training group and 0.877 for the testing group. The model’s positive predictive values were as follows: BI-RADS 2 = 0%, BI-RADS 3 = 0%, BI-RADS 4A = 6.25%, BI-RADS 4B = 26.13%, BI-RADS 4C = 80.84%, and BI-RADS 5 = 97.33%.ConclusionThe creation of a risk-predictive BI-RADS stratification, specifically designed for non-mass breast lesions and integrating clinical and imaging data from multiple modalities, significantly enhances the precision of diagnostic categorization for these lesions.

show abstract

Is Digital Breast Tomosynthesis the Better Mammogram for Local Breast Cancer Staging?

Cited by 2 publications

References 10 publications

A collection input based support tensor machine for lesion malignancy classification in digital breast tomosynthesis

A collection input based support tensor machine for lesion malignancy classification in digital breast tomosynthesis

A risk prediction stratification for non-mass breast lesions, combining clinical characteristics and imaging features on ultrasound, mammography, and MRI

Contact Info

Product

Resources

About