New machine learning tool identifies breast lesions likely to become cancerous

por Lauren Dubinsky, Senior Reporter | October 19, 2017
Health IT Rad Oncology Women's Health
May reduce unnecessary surgeries
A machine learning tool developed by Massachusetts General Hospital can identify breast lesions that are likely to become cancerous, according to a new study published in Radiology.

"Currently, there are no imaging or other features that reliably allow us to distinguish between high-risk lesions that warrant surgery from those that can be safely observed," Dr. Manisha Bahl, study author, told HCB News. "[That has] led to unnecessary surgery for high-risk lesions that are not associated with cancer."

Bahl and her team trained the tool using data on a group of patients with biopsy-proved high-risk lesions, who either underwent surgery or at least two years of imaging follow-up. Among the 1,006 high-risk lesions identified, 115 became cancerous.
DOTmed text ad

Reveal Mobi Pro now available for sale in the US

Reveal Mobi Pro integrates the Reveal 35C detector with SpectralDR technology into a modern mobile X-ray solution. Mobi Pro allows for simultaneous acquisition of conventional & dual-energy images with a single exposure. Contact us for a demo at no cost.

After training the tool with two-thirds of the high-risk lesions, the team tested it on the remaining 335 lesions. It accurately predicted 37 of the 38 lesions that became cancerous and could have helped avoid nearly one-third of the unnecessary surgeries.

"Machine learning allows us to incorporate the full spectrum of diverse and complex data that we have available, such as patient risk factors and imaging features, in order to predict which high-risk lesions are likely to be upgraded to cancer, and ultimately, to help our patients make more informed decisions about surgery versus surveillance," said Bahl.

The team is working to incorporate this tool into their daily clinical practice at MGH and hope to use it to guide decision-making in the near future. Their next step is to integrate actual mammogram images and histopathology slides into the machine learning tool.

You Must Be Logged In To Post A Comment