AI can help patients with high-risk breast lesions avoid unnecessary surgery
Researchers have developed new AI models that can predict when atypical ductal hyperplasia (ADH) breast lesions will be cancerous, publishing their findings in JCO Clinical Cancer Informatics.
A team led by Saeed Hassanpour, PhD, of Dartmouth College in Hanover, New Hampshire, tested six machine learning models for predicting when ADH would upgrade to cancer after a core needle biopsy. The goal was to avoid a full surgical excision in cases where patients could instead undergo active surveillance and hormonal therapy.
The authors tested the models using 128 lesions from 124 patients who underwent a surgical excision from 2011 to 2017. They noted that the gradient-boosting trees model, with an area under the curve (AUC) of 68 percent and accuracy of 78 percent, and the random forest model, with an AUC of 67 percent and accuracy of 77 percent, had the strongest performances.
In addition, the team noted, their research revealed that age at biopsy, esion size, the number of biopsies, needle gauge and the patient’s family history of breast cancer were key features in determining if the ADH would upgrade to cancer.
“Our results suggest there are robust clinical differences between women at low versus high risk for ADH upgrade to cancer based on core needle biopsy data that allowed our machine learning model to reliably predict malignancy upgrades in our dataset,” Hassanpour said in a prepared statement from Dartmouth College. “This study also identified important clinical variables involved in ADH upgrade risk.”
The team said that its research can spot 98 percent of all malignant ADH breast lesions and will save 16 percent of women who would have undergone the surgical excision from going through the surgery.
“Our model can potentially help patients and clinicians choose an alternative management approach in low-risk cases,” Hassanpour said in the same statement. “In the era of personalized medicine, such models can be desirable for patients who value a shared decision-making approach with the ability to choose between surgical excision for certainty versus surveillance to avoid cost, stress, and potential side effects in women at low risk for upgrade of ADH to cancer.”
Michael has more than 18 years of experience as a professional writer and editor. He has written at length about cardiology, radiology, artificial intelligence and other key healthcare topics.