Model can predict risk of upgrade of high-risk breast lesions to cancer using traditional, text features
TUESDAY, Oct. 17, 2017 (HealthDay News) — A machine learning model can predict the risk of upgrade of high-risk breast lesions (HRLs) to cancer, according to a study published online Oct. 17 in Radiology.
Manisha Bahl, M.D., M.P.H., from Massachusetts General Hospital in Boston, and colleagues identified consecutive patients with biopsy-proven HRLs who underwent surgery or at least two years of imaging follow-up. To identify HRLs at low risk for upgrade to cancer, a random forest machine learning model was developed. The model included traditional features such as age and HRL histologic results and also text features from the biopsy pathologic report.
A total of 1,006 HRLs were identified, and they had a cancer upgrade rate of 11.4 percent. The researchers developed the model with 671 HRLs and tested it with an independent set of 335 HRLs. Age and HRL histologic results were among the most important traditional features; “severely atypical” was an important text feature from the pathologic reports. If those categorized with the model to be at low risk for upgrade underwent surveillance and the remainder were excised instead of surgical excision of all HRLs, 97.4 percent of malignancies would have been diagnosed at surgery and 30.6 percent of surgeries of benign lesions could have been avoided.
“Use of this model could decrease unnecessary surgery by nearly one-third and could help guide clinical decision making with regard to surveillance versus surgical excision of HRLs,” the authors write.
Two authors disclosed having a patent in process with Massachusetts General Hospital and the Massachusetts Institute of Technology, and one author disclosed ties to GE Healthcare.
Abstract/Full Text (subscription or payment may be required)
Copyright © 2017 HealthDay. All rights reserved.
