Background: Ultrasound (US) is a fundamental diagnostic tool in breast imaging. However, US remains an operator-dependent examination. Research into and the application of artificial intelligence (AI) in breast US are increasing. The aim of this rapid review was to assess the current development of US-based artificial intelligence in the field of breast cancer. Methods: Two investigators with experience in medical research performed literature searching and data extraction on PubMed. The studies included in this rapid review evaluated the role of artificial intelligence concerning BC diagnosis, prognosis, molecular subtypes of breast cancer, axillary lymph node status, and the response to neoadjuvant chemotherapy. The mean values of sensitivity, specificity, and AUC were calculated for the main study categories with a meta-analytical approach. Results: A total of 58 main studies, all published after 2017, were included. Only 9/58 studies were prospective (15.5%); 13/58 studies (22.4%) used an ML approach. The vast majority (77.6%) used DL systems. Most studies were conducted for the diagnosis or classification of BC (55.1%). At present, all the included studies showed that AI has excellent performance in breast cancer diagnosis, prognosis, and treatment strategy. Conclusions: US-based AI has great potential and research value in the field of breast cancer diagnosis, treatment, and prognosis. More prospective and multicenter studies are needed to assess the potential impact of AI in breast ultrasound.
Objective: To perform a systematic review of the methods used for background parenchymal enhancement (BPE) evaluation on breast MRI. Methods: Studies dealing with BPE assessment on breast MRI were retrieved from major medical libraries independently by four reviewers up to 6 October 2015. The keywords used for database searching are “background parenchymal enhancement”, “parenchymal enhancement”, “MRI” and “breast”. The studies were included if qualitative and/or quantitative methods for BPE assessment were described. Results: Of the 420 studies identified, a total of 52 articles were included in the systematic review. 28 studies performed only a qualitative assessment of BPE, 13 studies performed only a quantitative assessment and 11 studies performed both qualitative and quantitative assessments. A wide heterogeneity was found in the MRI sequences and in the quantitative methods used for BPE assessment. Conclusion: A wide variability exists in the quantitative evaluation of BPE on breast MRI. More studies focused on a reliable and comparable method for quantitative BPE assessment are needed. Advances in knowledge: More studies focused on a quantitative BPE assessment are needed.
Background: Radiology is an essential tool in the management of a patient. The aim of this manuscript was to build structured report (SR) Mammography based in Breast Cancer. Methods: A working team of 16 experts (group A) was composed to create a SR for Mammography Breast Cancer. A further working group of 4 experts (group B), blinded to the activities of the group A, was composed to assess the quality and clinical usefulness of the SR final draft. Modified Delphi process was used to assess level of agreement for all report sections. Cronbach’s alpha (Cα) correlation coefficient was used to assess internal consistency and to measure quality analysis according to the average inter-item correlation. Results: The final SR version was built by including n = 2 items in Personal Data, n = 4 items in Setting, n = 2 items in Comparison with previous breast examination, n = 19 items in Anamnesis and clinical context; n = 10 items in Technique; n = 1 item in Radiation dose; n = 5 items Parenchymal pattern; n = 28 items in Description of the finding; n = 12 items in Diagnostic categories and Report and n = 1 item in Conclusions. The overall mean score of the experts and the sum of score for structured report were 4.9 and 807 in the second round. The Cronbach’s alpha (Cα) correlation coefficient was 0.82 in the second round. About the quality evaluation, the overall mean score of the experts was 3.3. The Cronbach’s alpha (Cα) correlation coefficient was 0.90. Conclusions: Structured reporting improves the quality, clarity and reproducibility of reports across departments, cities, countries and internationally and will assist patient management and improve breast health care and facilitate research.
Ductal carcinoma in situ (DCIS) is a common neoplasm that may be associated with focal invasive breast cancer lesions. The aim of our study was to evaluate the role of preoperative magnetic resonance imaging (MRI) in determining occult invasive presence and disease extent in patients with preoperative diagnosis of pure DCIS. We analyzed 125 patients with postoperative pure DCIS (n = 91) and DCIS plus invasive component (n = 34). Diagnostic mammography (MRX) showed a size underestimation rate of 30.4% while MRI showed an overestimation rate of 28.6%. Comparing the mean absolute error between preoperative MRI and MRX evaluations and final disease extent, MRI showed an improved accuracy of 51.2%. In our analysis preoperative breast MRI showed a better accuracy in predicting postoperative pathologic extent of disease, adding strength to the growing evidences that preoperative MRI can lead to a more appropriate management of DCIS patients.
Objective: Atypical ductal hyperplasia (ADH) is a proliferative lesion associated with a variable increased risk of breast malignancy, but the management of the patients is still not completely defined, with mandatory surgical excision in most cases. To report the results of the conservative management with mammographic checks of patients with ADH diagnosed by vacuum assisted breast biopsy (VAB), without residual calcifications. Methods: The authors accessed the institutional database of radiological, surgical and pathological anatomy. Inclusion criteria were: ADH diagnosed by VAB on a single group of microcalcifications, without residual post-procedure; follow-up at least of 12 months. Exclusion criteria were the presence of personal history of breast cancer or other high-risk lesions; association with other synchronous lesions, both more and less advanced proliferative lesions. Results: The 65 included patients were all females, with age range of 40–79 years (mean 54 years). The maximum diameter range of the groups of microcalcifications was 4–11 mm (mean 6.2 mm), all classified as BI-RADS 4b (Breast Imaging Reporting and Data System 4b) and defined as fine pleomorphic in 29 cases (45%) or amorphous in 36 cases (55%). The range of follow-up length was 12–156 months (mean 67 months). Only one patients developed new microcalcifications, in the same breast, 48 months after and 15 mm from the first VAB, interpreted as low-grade ductal carcinoma in situ (DCIS) at surgical excision. Conclusion: These results could justify the conservative management, in a selected group of patients, being the malignancy rate lower than 2%, considered in the literature as the “probably benign” definition. Advances in knowledge: Increasing the length of follow-up of selected patients conservatively managed can improve the management of ADH cases.
To evaluate quantitative measurements of background parenchymal enhancement (BPE) on breast MRI and compare them with observer-based scores. BPE of 48 patients (mean age: 48 years; age range: 36–66 years) referred to 3.0-T breast MRI between 2012 and 2014 was evaluated independently and blindly to each other by two radiologists. BPE was estimated qualitatively with the standard Breast Imaging Reporting and Data System (BI-RADS) scale and quantitatively with a semi-automatic and an automatic software interface. To assess intrareader agreement, MRIs were re-read after a 4-month interval by the same two readers. The Pearson correlation coefficient (r) and the Bland–Altman method were used to compare the methods used to estimate BPE. p-value <0.05 was considered significant. The mean value of BPE with the semi-automatic software evaluated by each reader was 14% (range: 2–79%) for Reader 1 and 16% (range: 1–61%) for Reader 2 (p > 0.05). Mean values of BPE percentages for the automatic software were 17.5 ± 13.1 (p > 0.05 vs semi-automatic). The automatic software was unable to produce BPE values for 2 of 48 (4%) patients. With BI-RADS, interreader and intrareader values were κ = 0.70 [95% confidence interval (CI) 0.49–0.91] and κ = 0.69 (95% CI 0.46–0.93), respectively. With semi-automated software, interreader and intrareader values were κ = 0.81 (95% CI 0.59–0.99) and κ = 0.85 (95% CI 0.43–0.99), respectively. BI-RADS scores correlated with the automatic (r = 0.55, p < 0.001) and semi-automatic scores (r = 0.60, p < 0.001). Automatic scores correlated with the semi-automatic scores (r = 0.77, p < 0.001). The mean percentage difference between automatic and semi-automatic scores was 3.5% (95% CI 1.5–5.2). BPE quantitative evaluation is feasible with both semi-automatic and automatic software and correlates with radiologists' estimation. Computerized BPE quantitative evaluation is feasible with both semi-automatic and automatic software. Computerized BPE quantitative scores correlate with radiologists' estimation.