Early diagnosis is an effective strategy for decreasing breast cancer mortality. Ultrasonography is one of the most predominant imaging modalities for breast cancer owing to its convenience and non‑invasiveness. The present study aimed to develop a model that integrates age with Breast Imaging Reporting and Data System (BI‑RADS) lexicon to improve diagnostic accuracy of ultrasonography in breast cancer. This retrospective study comprised two cohorts: A training cohort with 975 female patients from Renmin Hospital of Wuhan University (Wuhan, China) and a validation cohort with 500 female patients from Maternal and Child Health Hospital of Hubei Province (Wuhan, China). Logistic regression was used to construct a model combining BI‑RADS score with age and to determine the age‑based prevalence of breast cancer to predict a cut‑off age. The model that integrated age with BI‑RADS scores demonstrated the best performance compared with models based solely on age or BI‑RADS scores, with an area under the curve (AUC) of 0.872 (95% CI: 0.850‑0.894, P<0.001). Furthermore, among participants aged <30 years, the prevalence of breast cancer was lower than the lower limit of the reference range (2%) for BI‑RADS subcategory 4A lesions but within the reference range for BI‑RADS category 3 lesions, as indicated by linear regression analysis. Therefore, it is recommended that management for this subset of participants are categorized as BI‑RADS category 3, meaning that biopsies typically indicated could be replaced with short‑term follow‑up. In conclusion, the integrated assessment model based on age and BI‑RADS may enhance accuracy of ultrasonography in diagnosing breast lesions and young patients with BI‑RADS subcategory 4A lesions may be exempted from biopsy.

Related Articles