AI detects 3 dangerous cancers with just one drop of dried blood

Initial tests by Chinese scientists show that with just a spot of dried blood, the tool can distinguish patients with pancreatic, stomach or colorectal cancer from healthy people within minutes.

The new tool uses machine learning, a branch of AI, to analyze metabolic byproducts, called metabolites, in blood samples. These metabolites (found in serum) act as “biomarkers” that can detect the presence of cancer in the body, according to LiveScience on April 23.

Screening for blood biomarkers has been proposed as a potential way to diagnose early-stage cancer when survival rates are higher and patients have no obvious symptoms.

Despite being some of the deadliest cancers in the world, pancreatic, stomach and colorectal cancers currently do not have a standalone blood test accurate enough to diagnose them on their own. Instead, doctors often rely on imaging or surgery to detect cancerous tissue.

The team of Chinese scientists said their new test would theoretically require less than 0.05ml of blood to diagnose the above cancers. The tool has also been shown to be able to accurately distinguish between cancer patients and healthy people.

Dr. Chaoyuan Kuang, who works at Albert Einstein College of Medicine (USA) and was not involved in the above research, said that compared to liquid blood, dried serum can be "collected, stored and transported at a lower cost and with much simpler equipment".

However, he said many more years of research are needed before the test can be widely available to the public.

Scientists estimate that by 2030, about 75% of cancer deaths will occur in low- and middle-income countries due to barriers to health care.

The new testing tool using dried blood could help improve access to testing in resource-poor settings such as remote areas, the study authors say.

The study was published in the journal Nature Sustainability.