UPMC Deploys AI-Powered Prostate Cancer Diagnostic

The University of Pittsburgh Medical Center (UPMC) and Ibex Medical Analytics announced a partnership focused on leveraging Ibex’s artificial intelligence (AI) solution, Galen Prostate, to support pathologists in diagnosing prostate cancer.

The American Cancer Society (ACS) reports that prostate cancer is the most common cancer in US males other than skin cancer, with about 268,490 new cases and 34,500 deaths from the disease in 2022. An estimated one in eight males will be diagnosed with prostate cancer, but early detection may help improve outcomes in certain populations.

ACS recommends screening for prostate cancer using prostate-specific antigen (PSA) blood tests and digital rectal exams (DRE). An abnormal result from either test may necessitate a prostate biopsy.

UPMC will leverage Galen Prostate as part of this process, digitizing prostate biopsies and analyzing them with the platform ahead of a pathologist’s review, according to the press release. The platform is designed to provide diagnostic insights and clinical decision support in an effort to improve the quality and turnaround time of cancer diagnoses.

"Galen Prostate serves as a 'digital assistant' for our pathologists, helping automate routine processes and providing augmented insights for each biopsy," said Rajiv Dhir, MD, medical director of anatomic pathology services at UPMC, in the press release. "We look forward to expanding the deployment of this technology across our network as we adopt digital workflows that enable our pathology departments to increase efficiency, optimize workflow and ensure the highest standard of care and long-term outcomes for cancer patients."

This partnership comes after UPMC published a 2020 Ibex-funded study in The Lancet Digital Health evaluating Ibex’s latest software release, Galen 3.0, for prostate cancer detection. The study found that the tool demonstrated high accuracy in detecting prostate cancer and assessing clinically relevant features.

In the past few years, AI tools have increasingly been leveraged to help improve prostate cancer detection and care.

In 2021, University of Michigan researchers found that a commercially available genomic test has the potential to identify prostate cancer patients who are at high risk of aggressive prostate cancer progression.

The researchers noted that a positive cancer diagnosis following a prostate biopsy leaves patients and care teams with two options: proceed with radiation or therapy, or take a 'watch-and-wait' approach, also known as active surveillance. Choosing an option relies on identifying which patients may be at risk of aggressive cancer progression, which can be difficult.

Using the genomic test, the researchers found that patients with higher risk scores who were placed on active surveillance had a shorter time before active treatment was needed, while those with high scores who underwent surgery to remove the prostate gland experienced shorter times until their prostate-specific antigen levels increased again or when they started anti-androgen therapy.