Precision Medicine News

UPMC Leverages Artificial Intelligence to Improve Breast Cancer Treatment

Researchers from UPMC Hillman Cancer Center have partnered with Realyze Intelligence to utilize artificial intelligence to improve treatment options for breast cancer patients.

A doctor in a white lab coat writing on a clipboard

Source: Getty Images

By Shania Kennedy

- UPMC cancer researchers have launched a partnership with healthcare data analytics company Realyze Intelligence, a UPMC Enterprises portfolio company, to improve treatments for early-stage breast cancer using artificial intelligence and natural language understanding (NLU), a subset of natural language processing (NLP).

NLP, at its core, enables computers to understand both written and verbal human language. NLU is more specific, using semantic and syntactic analysis of speech and text to determine the meaning of a sentence. In research, NLU is helpful because it establishes a data structure that specifies the relationships between words, which can be used for data mining and sentiment analysis.

Under the partnership, UPMC is using Realyze’s NLU platform to help gauge whether sentinel lymph node biopsy (SLNB) is appropriate in early-stage breast cancer patients younger than 70. Previous UPMC research in this area indicates that SLNB can be avoided in most patients over the age of 70 and is a low-value surgery for this demographic.

Like any procedure, SLNB is helpful for certain patients but carries the risk of significant complications for others. To identify which patient cohorts under 70 may not benefit from SLNB, the researchers are leveraging the NLU platform to read the unstructured clinical notes and structured data from patients’ EHRs, a process that is extremely time-consuming when undertaken manually by humans.

In a currently unpublished study, the researchers are examining EHR data from 602 early-stage breast cancer patients who received SLNBs from January 2015 to December 2017 at 15 UPMC hospitals in western Pennsylvania. These data were then used to create a breast cancer model focused on lymph node identification and positivity.

Initial results indicate no difference in the node positivity rate between patients older or younger than 70, especially for patients with stage I cancer, according to the press release.

The findings suggest that current guidance for SLNBs could be expanded to more patients, the researchers concluded.

“We want to make sure we are targeting the right care to the right patient to give them the best care and quality of life possible,” said Adrian Lee, PhD, lead researcher on the study, breast cancer investigator at UPMC Hillman Cancer Center and Magee-Womens Research Institute, and director of the UPMC/University of Pittsburgh Institute of Precision Medicine, in a press release.

“Sometimes the most interesting and relevant data points are in the unstructured field of a patient’s record. Having the ability to record and analyze the data from these fields is essential to understanding if SLNBs are necessary for this patient population. By using the Realyze platform rather than a cancer registry, we can quickly and efficiently extract a large amount of data in real time,” Lee continued.

UPMC Enterprises launched Realyze Intelligence last year to advance the use of AI in pinpointing and enhancing treatments for chronic diseases, adding to its efforts to combat cancer.