City of Hope and UC Berkeley AI Platform Calculates Breast Cancer Risk by Squeezing Individual Cells

City of Hope and UC Berkeley researchers develop MechanoAge, an AI platform that uses low-cost sensors to assess breast cancer risk via cell stiffness.

By: AXL Media

Published: Apr 24, 2026, 6:23 AM EDT

Source: Information for this report was sourced from EurekAlert!

Quantifying Cancer Risk Through Cellular Stress

A pioneering collaboration between City of Hope and the University of California, Berkeley has resulted in the creation of a microfluidic device designed to evaluate breast cancer risk at a cellular level. This first of its kind platform works by physically squeezing individual breast epithelial cells through narrow channels to observe how they deform and recover under mechanical stress. Mark LaBarge, a professor at City of Hope, noted that while genetic testing is available for a small percentage of women, the vast majority lack a known predisposition. This tool provides tangible, cell based evidence that fills a critical gap for women who currently rely on vague population estimates to understand their personal risk.

The Mechanical Age Versus Chronological Age

The research team uncovered a fundamental biological insight: breast cells possess a mechanical age that can differ significantly from a person’s actual chronological age. By measuring how cells respond to being compressed, the MechanoAge platform can identify signs of accelerated aging that correlate with higher cancer risk. According to Lydia Sohn, a professor of mechanical engineering at UC Berkeley, this is the first time that the mechanical properties of biological cells have been quantified in a way similar to how engineers study the aging of materials like concrete or metal. The findings suggest that the older a cell’s mechanical age, the more susceptible an individual is to developing breast cancer.

Machine Learning and Node Pore Sensing Technology

The platform utilizes a process known as mechano-node-pore sensing, which measures electrical current across a liquid filled channel. As cells pass through the device, they disrupt the current, providing data on their size and shape, while narrow segments of the channel force the cells to squeeze and then bounce back. Researchers integrated this data with a machine learning algorithm specifically trained to detect physical differences between cells. The AI successfully identified that cells from older women, or those at high risk, were consistently stiffer and took a longer time to return to their original shape after being subjected to pressure.