A recent study led by researchers at NYU Langone Health’s Department of Orthopedic Surgery has demonstrated the potential of using artificial intelligence to track and measure aging cells. This research combines high-resolution imaging with machine learning to analyze cells that have been damaged due to injury, aging, or disease.
The study, published in Nature Communications on July 7, focuses on senescent cells. These are cells that no longer grow and reproduce normally but play a crucial role in wound repair and aging-related diseases such as cancer and heart disease. By tracking these cells, researchers aim to gain insights into tissue regeneration and disease progression.
The team trained a computer system to analyze animal cells exposed to chemicals simulating human aging. The AI analysis identified several measurable features linked to the cell nucleus that correlated with the degree of senescence. These features included changes in shape, density, and genetic material staining.
The researchers developed a nuclear morphometric pipeline (NMP) from their analysis. This tool assigns a senescent score based on the physical characteristics of the nucleus. The NMP was tested on various cell types in mice across different ages and successfully distinguished between healthy and diseased tissues.
“Our study demonstrates that specific nuclear morphometrics can serve as a reliable tool for identifying and tracking senescent cells,” said Michael N. Wosczyna, PhD, senior investigator of the study. Dr. Wosczyna emphasized its importance for future research into tissue regeneration and progressive diseases.
The research team plans further experiments using the NMP on human tissues and aims to combine it with other biomarker tools for examining senescence’s roles in health conditions. Their ultimate goal is developing treatments that address the negative effects of cellular senescence.
Sahil Mapkar, BS, co-lead investigator from NYU Tandon School of Engineering stated: “Existing methods to identify senescent cells are difficult to use, making them less reliable than the nuclear morphometric pipeline.”
Funding for this study came from National Institutes of Health grant R01AG053438 and NYU Langone’s Department of Orthopedic Surgery.