Background. The authors attempted to appraise the ability of high resolution, real‐time ultrasound to find malignant breast masses that are nonpalpable and undetectable by high quality mammography in women with radiographically dense breasts, who were referred because of palpable or mammographically detected lesions. Methods. The records of breast ultrasound examinations of 12,706 women were retrospectively reviewed. All lesions were classified according to clinical and mammographic status as palpable or nonpalpable and as visible or nonvisible, respectively. Solid masses were sampled percutaneously by fine needle aspiration biopsy (FNAB) using ultrasound guidance and either were excised surgically or followed by sequential imaging. Results. There were 1575 solid masses detected sonographically that were nonpalpable and nonvisible by mammography; percutaneous biopsies (FNABs) were performed on 279 of these. Cytologic interpretation was definitely malignant in 22, suspicious in 18 (6 confirmed cancers), and benign in 183 (no false negatives). Surgery confirmed malignancy in 44 of the 1575 solid masses (2.8%), including 16 in patients with multifocal cancers. Conclusions. Ultrasound can detect unsuspected, mammographically occult cancers in radiographically dense breasts and can alter treatment planning when a second cancer is found in a breast that otherwise was considered appropriate for conservative surgery. The authors recommend that any solid lesion detected incidentally during breast sonography either should be biopsied percutaneously under ultrasound guidance and/or closely followed with sequential scans. Cancer 1995; 76:626–30.

Solid breast masses diagnosed as fibroadenomas at FNAB may be safely followed up if volume growth rate is less than 16% per month in those younger than 50 years and less than 13% per month in those 50 years or older. Acceptable mean change in dimension for a 6-month interval is 20% for all ages.

Real-time US-guided fine-needle aspiration biopsy is a useful adjunct to clinical, mammographic, and sonographic assessment of solid breast lesions.

This work reports the use of ultrasound radio frequency (RF) time series analysis as a method for ultrasound-based classification of malignant breast lesions. The RF time series method is versatile and requires only a few seconds of raw ultrasound data with no need for additional instrumentation. Using the RF time series features, and a machine learning framework, we have generated malignancy maps, from the estimated cancer likelihood, for decision support in biopsy recommendation. These maps depict the likelihood of malignancy for regions of size 1 mm(2) within the suspicious lesions. We report an area under receiver operating characteristics curve of 0.86 (95% confidence interval [CI]: 0.84%-0.90%) using support vector machines and 0.81 (95% CI: 0.78-0.85) using Random Forests classification algorithms, on 22 subjects with leave-one-subject-out cross-validation. Changing the classification method yielded consistent results which indicates the robustness of this tissue typing method. The findings of this report suggest that ultrasound RF time series, along with the developed machine learning framework, can help in differentiating malignant from benign breast lesions, subsequently reducing the number of unnecessary biopsies after mammography screening.

No abstract