Abstract
BackgroundTumors in radiologically dense breast were overlooked on mammograms more often than tumors in low-density breasts. A fast reproducible and automated method of assessing percentage mammographic density (PMD) would be desirable to support decisions whether ultrasonography should be provided for women in addition to mammography in diagnostic mammography units. PMD assessment has still not been included in clinical routine work, as there are issues of interobserver variability and the procedure is quite time consuming. This study investigated whether fully automatically generated texture features of mammograms can replace time-consuming semi-automatic PMD assessment to predict a patient’s risk of having an invasive breast tumor that is visible on ultrasound but masked on mammography (mammography failure).MethodsThis observational study included 1334 women with invasive breast cancer treated at a hospital-based diagnostic mammography unit. Ultrasound was available for the entire cohort as part of routine diagnosis. Computer-based threshold PMD assessments (“observed PMD”) were carried out and 363 texture features were obtained from each mammogram. Several variable selection and regression techniques (univariate selection, lasso, boosting, random forest) were applied to predict PMD from the texture features. The predicted PMD values were each used as new predictor for masking in logistic regression models together with clinical predictors. These four logistic regression models with predicted PMD were compared among themselves and with a logistic regression model with observed PMD. The most accurate masking prediction was determined by cross-validation.ResultsAbout 120 of the 363 texture features were selected for predicting PMD. Density predictions with boosting were the best substitute for observed PMD to predict masking. Overall, the corresponding logistic regression model performed better (cross-validated AUC, 0.747) than one without mammographic density (0.734), but less well than the one with the observed PMD (0.753). However, in patients with an assigned mammography failure risk >10%, covering about half of all masked tumors, the boosting-based model performed at least as accurately as the original PMD model.ConclusionAutomatically generated texture features can replace semi-automatically determined PMD in a prediction model for mammography failure, such that more than 50% of masked tumors could be discovered.
Highlights
Tumors in radiologically dense breast were overlooked on mammograms more often than tumors in low-density breasts
Patients were selected in the following hierarchical order from a total of 3974 breast cancers registered in the breast center’s database: invasive breast cancer; no contralateral breast cancer; mammography at primary diagnosis performed at the university breast center; physical availability of mammograms for the affected and contralateral sides; availability of a structured Breast Imaging Reporting and Data System (BI-RADS) or analogous assessment of the mammogram and ultrasound scan
The study shows that prediction of masking on diagnostic mammograms can be improved if mammographic density estimations using texture features are added to a prediction rule based on age, body mass index (BMI), prior surgery, menopausal and hormone replacement therapy (HRT) status, and imaging technique
Summary
Tumors in radiologically dense breast were overlooked on mammograms more often than tumors in low-density breasts. A fast reproducible and automated method of assessing percentage mammographic density (PMD) would be desirable to support decisions whether ultrasonography should be provided for women in addition to mammography in diagnostic mammography units. This study investi‐ gated whether fully automatically generated texture features of mammograms can replace time-consuming semiautomatic PMD assessment to predict a patient’s risk of having an invasive breast tumor that is visible on ultrasound but masked on mammography (mammography failure). The effort to improve breast cancer detection faces several challenges One of these is how to integrate different diagnostic methods into a single diagnostic process. Some diagnostic units use ultrasound for every patient, but others do so only for certain indications, such as dense breasts, or if the patient requests it [1]. The reasons for the unsystematic way in which ultrasound is used lie in the associated costs and the lack of prediction models capable of identifying those patients in whom an additional method would increase sensitivity without necessarily decreasing specificity
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
