Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 May;28(3):217-225.
doi: 10.5152/dir.2022.20664.

Digital breast tomosynthesis-based peritumoral radiomics approaches in the differentiation of benign and malignant breast lesions

Shuxian Niu  1 , Tao Yu  2 , Yan Cao  1 , Yue Dong  2 , Yahong Luo  2 , Xiran Jiang  1
Affiliations

Digital breast tomosynthesis-based peritumoral radiomics approaches in the differentiation of benign and malignant breast lesions

Shuxian Niu et al. Diagn Interv Radiol. 2022 May.

Abstract

PURPOSE We aimed to evaluate digital breast tomosynthesis (DBT)-based radiomics in the differentiation of benign and malignant breast lesions in women. METHODS A total of 185 patients who underwent DBT scans were enrolled between December 2017 and June 2019. The features of handcrafted and deep learning-based radiomics were extracted from the tumoral and peritumoral regions with different radial dilation distances outside the tumor. A 3-step method was used to select discriminative features and develop the radiomics signature. Discriminative clinical factors were identified by univariate logistic regression. The clinical fac- tors with P < .05 were used to build a clinical model with multivariate logistic regression. The radiomics nomogram was developed by integrating the radiomics signature and discriminative clinical factors. Discriminative performance of the radiomics signature, clinical model, nomo- gram, and breast imaging reporting and data system assessment were evaluated and compared with the receiver operating characteristic and decision curves analysis (DCA). RESULTS A total of 2 handcrafted and 2 deep features were identified as the most discriminative features from the peritumoral regions with 2 mm dilation distances and used to develop the radiomics signature. The nomogram incorporating the radiomics signature, age, and menstruation status showed the best discriminative performance with area under the curve (AUC) values of 0.980 (95% CI, 0.960 to 1.000; sensitivity =0.970, specificity =0.946) in the training cohort and 0.985 (95% CI, 0.960 to 1.000; sensitivity = 0.909, specificity = 0.966) in the validation cohort. DCA con- firmed the potential clinical usefulness of our nomogram. CONCLUSION Our results illustrate that the radiomics nomogram integrating the DBT imaging features and clinical factors (age and menstruation status) can be considered as a useful tool in aiding the clinical diagnosis of breast cancer.

PubMed Disclaimer

Conflict of interest statement

Conflict of interest disclosure The authors declared no conflicts of interest.

Figures

Figure 1.
Figure 1.
An example of a malignant breast lesion and the dilated masks in the digital breast tomosynthesis (DBT) image. The red-colored region represents the original tumoral region that was manually segmented by radiologists. The colored rings outside the tumoral region indicate the radially dilated regions. Each ring is 2 mm wide.
Figure 2.
Figure 2.
Overall workflow of this study. ROI, region of interest.
Figure 3. a-d.
Figure 3. a-d.
ROC curves of the logistic regression models based on tumoral and peritumoral regions: (a) and (b), logistic regression models using handcrafted features in the training (a) and validation (b) cohorts; (c) and (d), logistic regression models using deep features in the training (c) and validation (d) cohorts. AUC, area under the curve; ROC, receiver operating characteristic.
Figure 4.
Figure 4.
Boxplots of the 4 selected features from the DBT image.
Figure 5. a-c.
Figure 5. a-c.
Development and validation of the nomogram based on DBT data: (a), the developed nomogram model; (b) and (c), calibration curves of the model in training (b) and validation (c) cohort, respectively.
Figure 6. a, b.
Figure 6. a, b.
ROC curves of the clinical model, BI-RADS assessment, radiomics signature, and nomogram in the training (a) and validation (b) cohorts. The orange line indicates the clinical model. The green line represents BI-RADS assessment. The blue line indicates the radiomics signature. The red line represents the nomogram. BI-RADS, breast imaging reporting and data system. AUC, area under the curve.
Figure 7.
Figure 7.
Decision curve analysis for the clinical model, radiomics signature, BI-RADS assessment, and nomogram. The black line shows the hypothesis that the patients were all benign and not treated. The gray line shows the assumption that the patients were all malignant and all treated. The orange line indicates the treatment decision made by the clinical model. The green line indicates the treatment decision made by the BI-RADS assessment. The blue line indicates the treatment decision made by the radiomics signature. The red line indicates the treatment decision made by the nomogram.
Supplementary Figure S1. a, b.
Supplementary Figure S1. a, b.
Radiomics feature selection using the least absolute shrinkage and selection operator (LASSO) logistic regression. (a) Tuning parameter (λ) selection in the LASSO model used 10-fold cross-validation via minimum criteria. (b) LASSO coefficient profiles of the radiomics features.
Supplementary Figure S2.
Supplementary Figure S2.
Unsupervised cluster analysis of the selected features from peritumoral regions in the DBT (digital breast tomosynthesis) image and patients. The x-axis represents the selected features (n=4). The y-axis represents the patients (n=185). The red color represents patients with malignant tumor, while the blue color indicates the patients with benign tumor.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. McGuire A, Brown JA, Malone C, McLaughlin R, Kerin MJ. Effects of age on the detection and management of breast cancer. Cancers. 2015;7(2):908 929. 10.3390/cancers7020815) - DOI - PMC - PubMed
    1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2018. CA Cancer J Clin. 2018;68(1):7 30. 10.3322/caac.21442) - DOI - PubMed
    1. Valdora F, Houssami N, Rossi F, Calabrese M, Tagliafico AS. Rapid review: radiomics and breast cancer. Breast Cancer Res Treat. 2018;169(2):217 229. 10.1007/s10549-018-4675-4) - DOI - PubMed
    1. Michell MJ. Breast screening review—a radiologist’s perspective. Br J Radiol. 2012;85(1015):845 847. 10.1259/bjr/21332901) - DOI - PMC - PubMed
    1. Pauwels EK, Foray N, Bourguignon MH. Breast cancer induced by X-ray mammography screening? A review based on recent understanding of low-dose radiobiology. Med Princ Pract. 2016;25(2):101 109. 10.1159/000442442) - DOI - PMC - PubMed

Grants and funding

This work was supported by funding from the Youth Science and Technology Innovation Leader Support Project [RC170497], Natural Science Foundation of Liaoning Province of China [201602450], National Natural Science Foundation of China [U1708261], National Key R&D Program of Ministry of Science and Technology of China [2016YFC1303002] and Supporting Fund for Big Data in Health Care [HMB201903101], Special Fund for Research in the Public Interest of China [201402020], 2020 Key Project of Double Service for Universities in Shenyang.