28.08.2025

Researchers at Uppsala University, in collaboration with Umeå University, have developed an AI tool capable of identifying early signs of prostate cancer that pathologists may overlook. Published in Scientific Reports, the study shows that subtle tissue changes can be detected long before cancer becomes visible to the human eye, potentially improving early diagnosis and follow-up care.

Image: The image shows two researchers discussing prostate tissue images on a large screen; Copyright: Mikael Wallerstedt

Researchers at Uppsala University used AI to analyze prostate tissue biopsies, detecting subtle changes that may indicate prostate cancer before pathologists can see them.

Products and exhibitors related to this topic

Exhibitors and products related to this topic can be found in the database of MEDICA:

Study design and results

The research was conducted with Umeå University, where tissue samples were collected over several years. In total, 232 men were included, all initially assessed as healthy after biopsy.

Within 2.5 years, half of the men developed aggressive prostate cancer.
The remaining men stayed cancer-free for at least eight years.

Since the original pathology reports found no abnormalities, the team developed a new training method for the AI system. The tool analyzed biopsy images section by section, searching for patterns linked to future cancer development.

Tissue patterns and clinical relevance

“When we looked at the patterns that the AI ranked as informative, we saw changes in the tissue surrounding the glands in the prostate – changes also observed in other studies. This shows that AI analysis of routine biopsies can detect subtle signs indicating clinically significant prostate cancer before it becomes obvious to a pathologist,” Wählby explains.

The findings suggest that AI could complement pathology by indicating which men, although initially considered healthy, should receive closer follow-up.

Open data for future research

The imaging data and developed methods are openly available for further research and tool development. This transparency supports the integration of AI into future clinical diagnostics, helping medical teams improve prostate cancer screening and early detection strategies.

MEDICA-tradefair.com; Source: Uppsala Universitet

More about the UU at: www.uu.se/en