RESUMEN
OBJECTIVE: This study assessed mentorship interest within the breast radiologist community to guide development of a mentorship program through the Society of Breast Imaging (SBI). METHODS: A 19-question survey developed by the SBI mentorship committee was distributed electronically to its members March 16, 2021, to May 7, 2021, to gauge interest in forming a society-sponsored mentorship program. Responses were analyzed, with subgroups compared using chi-square analysis. RESULTS: There was an 18% response rate (598/3277), and 65% (381/588) professed interest in an SBI-sponsored mentorship. Respondents were evenly distributed between academic (241/586, 41%) and private practice (242/586, 41%). Most were breast imaging fellowship-trained (355/593, 60%) and identified as female (420/596, 70%). For practice years, 50% (293/586) were late career (11+ years) with the remainder early-mid career (201/586, 34%) or trainees (92/586, 16%). For mentorship content areas, work/life balance was the most popular choice (275/395, 70%) followed by leadership (234/395, 59%). Most respondents were not currently mentors (279/377, 74%) or mentees (284/337, 84%). Those interested in a mentorship relationship were statistically younger (<45 years old, 234/381, 61% vs 31/207, 15%, P < 0.00001), female (289/381, 76% vs 123/207, 59%, P = 0.00003), academics (189/381, 50% vs 48/207, 23%, P < 0.00001), identified as a racial/ethnic minority (138/381, 64% vs 121/297, 15%, P < 0.00001), and fellowship-trained (262/381, 69% vs 88/207, 43%, P < 0.00001). CONCLUSION: There is demand, especially among the society's young and minority members, for an SBI-sponsored mentorship program. Work/life balance and leadership were the most popular choices for guidance.
RESUMEN
PURPOSE: To investigate whether radiomics features extracted from magnetic resonance imaging (MRI) of patients with biopsy-proven atypical ductal hyperplasia (ADH) coupled with machine learning can differentiate high-risk lesions that will upgrade to malignancy at surgery from those that will not, and to determine if qualitatively and semi-quantitatively assessed imaging features, clinical factors, and image-guided biopsy technical factors are associated with upgrade rate. METHODS: This retrospective study included 127 patients with 139 breast lesions yielding ADH at biopsy who were assessed with multiparametric MRI prior to biopsy. Two radiologists assessed all lesions independently and with a third reader in consensus according to the BI-RADS lexicon. Univariate analysis and multivariate modeling were performed to identify significant radiomic features to be included in a machine learning model to discriminate between lesions that upgraded to malignancy on surgery from those that did not. RESULTS: Of 139 lesions, 28 were upgraded to malignancy at surgery, while 111 were not upgraded. Diagnostic accuracy was 53.6%, specificity 79.2%, and sensitivity 15.3% for the model developed from pre-contrast features, and 60.7%, 86%, and 22.8% for the model developed from delta radiomics datasets. No significant associations were found between any radiologist-assessed lesion parameters and upgrade status. There was a significant correlation between the number of specimens sampled during biopsy and upgrade status (p = 0.003). CONCLUSION: Radiomics analysis coupled with machine learning did not predict upgrade status of ADH. The only significant result from this analysis is between the number of specimens sampled during biopsy procedure and upgrade status at surgery.
