Artificial Intelligence Advances Early Detection of Blood Mutations Linked to Cancer and Heart Disease

Researchers at Mayo Clinic have developed an innovative artificial intelligence (AI) tool capable of identifying early blood mutations associated with an increased risk of cancer, particularly leukemia, and cardiovascular diseases. These blood mutations, often present in a slow-growing cluster of mutated blood cells, occur in approximately 20% of older adults and can significantly elevate disease risk without obvious symptoms.

This condition, known as clonal hematopoiesis of indeterminate potential (CHIP), originates in the bone marrow where blood stem cells normally produce cells vital for oxygen transport, immune defense, and organ function. When these stem cells acquire mutations in genes linked to blood cancers, they can multiply abnormally, forming clusters of mutated cells that expand over time. Although CHIP is usually asymptomatic, it has been linked to a higher incidence of death, notably from heart disease, and increases the likelihood of developing leukemia more than tenfold.

To better understand and detect this hidden risk, Mayo Clinic researchers created UNISOM (Unified Somatic Calling and Machine Learning), an advanced tool that pinpoints CHIP-related mutations within standard genetic datasets. Developed under the leadership of Ph.D. scholars Shulan Tian and Eric Klee, UNISOM enables clinicians to identify mutations that traditional methods often miss, especially those present in fewer than 5% of blood cells, by analyzing whole-exome and whole-genome sequencing data.

Detecting these early genetic changes enhances the potential for proactive monitoring and personalized disease prevention strategies. Dr. Klee emphasizes that early molecular detection signifies a major step forward in personalized medicine, translating genomic discoveries into practical clinical tools. Dr. Tian highlights that the tool has already demonstrated high accuracy, identifying nearly 80% of CHIP mutations, and aims to be expanded to larger, more diverse datasets to support wider clinical use.

Understanding and identifying CHIP mutations early may lead to improved management of patients at risk, ultimately reducing the burden of blood cancers and cardiovascular diseases among aging populations.

Source: https://medicalxpress.com/news/2025-08-ai-tool-early-blood-mutations.html