Pathologic findings in reduction mammoplasty procedures identified by natural language processing of breast pathology reports: A surrogate for the population incidence of cancer and high risk lesions.

2018 
e13569Background: Breast reduction surgery removes a random sample of breast tissue in otherwise asymptomatic women and thus provides a method to evaluate the background incidence of breast pathology in this population. Our goal was to identify the rate of atypical breast lesions and of cancers women of various ages in the largest mammoplasty cohort reported to date. Methods: With institutional Review Board (IRB) approval, pathologic reports from patients undergoing bilateral mammoplasty procedures in 5 institutions were analyzed using Natural Language Processing, verified by human review. Patients with a previous history of cancer were excluded. Results: A total of 5121 patients were included. The median age was 41 years (range 13-85), while the median age was higher in patients with any incidental finding when compared to patients with normal pathologic report (51 vs 39, p = 0.0001). Pathological findings were detected in 7.1% of procedures (table 1). Benign high-risk lesions (LCIS and atypical hyperpla...
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    2
    Citations
    NaN
    KQI
    []