Assessment of Nonmass Lesions Detected with Screening Breast US Based on Mammographic Findings.

Radiology

From the Department of Radiology, Seoul National University Hospital, Seoul, South Korea (S.M.H., W.K.M., H.Y.); Department of Radiology, Seoul National University College of Medicine, Seoul, South Korea (S.M.H., W.K.M.); Institute of Radiation Medicine, Seoul National University Medical Research Ce

Published: November 2024


Category Ranking

98%

Total Visits

921

Avg Visit Duration

2 minutes

Citations

20

Article Abstract

Background Breast nonmass lesions (NMLs) are observed at screening and diagnostic US. However, knowledge is limited on imaging features of NMLs at screening US. Purpose To identify features of NMLs at screening US that are suspicious for malignancy based on mammographic findings. Materials and Methods This retrospective, multicenter study included asymptomatic women who underwent screening US between January 2012 and December 2019. Eligible women had NMLs at US, concurrent screening mammography, and record of a final diagnosis. Logistic regression analyses were used to identify factors associated with malignancy. The diagnostic performance of each sonographic feature according to mammographic findings was calculated. A reader study was performed to assess interreader agreement for sonographic features. Results Among 993 NMLs in 993 patients (mean age, 50 years ± 9 [SD]), 914 (92.0%) were benign and 79 (8.0%) were malignant. Mean size was larger for malignant NMLs than for benign NMLs (2.6 cm ± 1.1 vs 1.9 cm ± 0.8; < .001). In multivariable analysis, associated calcifications (odds ratio [OR], 21.6 [95% CI: 8.0, 58.2]; < .001), posterior shadowing (OR, 6.9 [95% CI: 2.6, 18.4]; < .001), segmental distribution (OR, 6.2 [95% CI: 2.7, 14.4]; < .001), mixed echogenicity (OR, 5.0 [95% CI: 1.8, 14.0]; < .001), and size (OR, 1.5 [95% CI: 1.1, 2.1]; = .01) at US were associated with malignancy. Associated calcifications, posterior shadowing, segmental distribution, and mixed echogenicity showed positive predictive values (PPVs) of 44%, 22%, 22.9%, and 16.6%, respectively. Having a negative mammogram was associated with a lower malignancy rate (2.8% vs 28.8%) and lower PPVs for sonographic features (0.7%-10.4% vs 24%-55%) than having a positive mammogram. Interreader agreement for sonographic features was good to excellent (Fleiss κ 95% CI lower bound range, 0.63-0.81). Conclusion Calcifications, posterior shadowing, segmental distribution, and mixed echogenicity associated with NMLs can be considered suspicious features for malignancy at screening US. As malignancy rates and PPVs differ according to mammographic abnormalities, combined assessment is mandatory. © RSNA, 2024 See also the editorial by Grimm in this issue.

Download full-text PDF

Source
http://dx.doi.org/10.1148/radiol.240043DOI Listing

Publication Analysis

Top Keywords

mammographic findings
12
sonographic features
12
posterior shadowing
12
segmental distribution
12
mixed echogenicity
12
nonmass lesions
8
based mammographic
8
nmls
8
features nmls
8
nmls screening
8

Similar Publications

Purpose: To investigate the effectiveness of an integrated deep-learning (DL) algorithm, the Mixture of Radiological Findings Specific Experts (MoRFSE), in breast cancer classification by imitating the diagnostic decision-making process of radiologists.

Methods: A total of 2,764 mammographic images (1,462 breast cancer, 248 benign lesions, and 1,054 normal breast tissue) from the TOMPEI-CMMD were used. The MoRFSE comprises three DL models: a gate network for categorization (gNet) and two classification expert networks (cExp and mExp) specialized in capturing the distinct characteristics of calcifications and masses, respectively.

View Article and Find Full Text PDF

Breast density is a recognized risk factor for breast cancer and can affect the sensitivity of mammography. Consequently, magnetic resonance imaging (MRI) is recommended as a screening modality for women with increased breast density. However, mammography remains the primary method for assessing a woman's breast density classification.

View Article and Find Full Text PDF

Improving Benign and Malignant Classifications in Mammography with ROI-Stratified Deep Learning.

Bioengineering (Basel)

August 2025

Graduate School of Information Science, University of Hyogo, 7-1-28 Minatojima Minamimachi, Chuo-ku, Kobe-shi 650-0047, Hyogo, Japan.

Deep learning has achieved widespread adoption for medical image diagnosis, with extensive research dedicated to mammographic image analysis for breast cancer screening. This study investigates the hypothesis that incorporating region-of-interest (ROI) mask information for individual mammographic images during deep learning can improve the accuracy of benign/malignant diagnoses. Swin Transformer and ConvNeXtV2 deep learning models were used to evaluate their performance on the public VinDr and CDD-CESM datasets.

View Article and Find Full Text PDF

Adolescence represents a key opportunity for breast cancer prevention, as the rapid proliferation of breast tissue during puberty creates a critical window of vulnerability for the development of cancerous cells. With increasing research on adolescent dietary factors and breast cancer risk, we conducted a systematic review and meta-analysis to summarize the associations between adolescent diet and risk of breast cancer in adulthood, as well as benign breast disease (BBD) and high mammographic breast density, which are markers for breast cancer. We searched Web of Science, Ovid MEDLINE, Cochrane CENTRAL and Embase for epidemiological studies assessing dietary intakes in adolescent girls (aged 10-18 years), published through 16 October 2024, with no language or time restrictions.

View Article and Find Full Text PDF

Background: Mammography is the cornerstone of breast cancer screening. Its diagnostic performance, however, is influenced by population demographics such as age and breast density.

Purpose: The aim of this study was to establish contemporary performance benchmarks for mammography screening in Bahrain's primary health-care centres (PHCs) and to identify areas for quality improvement.

View Article and Find Full Text PDF