To investigate whether three-dimensional inversion imaging can improve the diagnosis of the nature of Breast Imaging Reporting and Data System (BI-RADS) category 4 breast masses to reduce unnecessary biopsies.

A total of 247 breast masses were collected from 243 patients who were diagnosed as having BI-RADS category 4 lesions by conventional ultrasound and underwent three-dimensional inversion imaging from September 2021 to June 2023 at the Second Affiliated Hospital of Anhui Medical University. Based on the distribution characteristics of dark grey and black areas within the mass in the inverted images, the diagnostic efficacy of conventional ultrasound alone and combined with three-dimensional inversion imaging for BI-RADS category 4 breast masses was compared by using a semi-quantitative 5-point inversion scoring method and plotting receiver operating characteristic (ROC) curves with pathological findings as the gold standard.

The area under the ROC curve (AUC), accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of conventional ultrasound for identifying the nature of BI-RADS category 4 breast masses were 0.841, 83.8%, 93.3%, 74.8%, 77.8%, and 92.2%, respectively; the corresponding values of the combined diagnosis were 0.887, 88.7%, 90.0%, 87.4%, 87.1%, and 90.2%. The combined diagnosis resulted in 87 BI-RADS category 4a benign masses being downgraded to category 3, and did not result in malignant masses mistakenly downgraded to category 3, which would have prevented unnecessary biopsies of 84.5% (87/103) of category 4a masses.

Conventional ultrasound combined with three-dimensional inversion imaging has good diagnostic efficacy for benign and malignant breast masses, significantly improving the diagnosic specificity and positive predictive value and thus effectively reducing unnecessary biopsies of category 4a masses.