AI Helps Diagnose Brain Tumors

A new study says that chemical analysis of blood samples combined with an artificial intelligent (AI) program, can accelerate the diagnosis of brain tumors. The research was revealed at the 2019 National Cancer Research Institute (NCRI) Cancer conference at the Scottish Event Campus, Glasgow, Scotland.

Reporting from the page Medical News, Thursday (7/11), brain tumors tend to have symptoms that are not specific. Early symptoms are usually very common like other diseases, such as headaches and memory problems. Brain scanning is the only way that can diagnose a brain tumor at this time.

Senior Clinical Lecturer and Surgery Consultant Sarah University of Edinburgh, England, Dr. Paul Brennan revealed, brain tumors have reduced the life expectancy of an average person by 20 years. This expectation is the highest of all cancers.

Dr. Brennan has worked with Dr. Matthew Baker, a chemical analyst at the University of Strathclyde, UK, and chief scientific officer at ClinSpec Diagnostics Ltd., developing tests to help doctors have a quick and efficient way to find patients who are most likely to have tumors.

This test depends on the available technique. Called infrared spectroscopy to check the chemical composition of a person's blood, combined with an AI program that can find chemical clues that indicate the possibility of brain tumors.

The researchers tried a new test on blood samples taken from 400 patients with possible signs of a brain tumor that had been referred for a brain scan at the Western General Hospital in Edinburgh, England. Of these, 40 people were later found to have brain tumors.

Using this test, researchers can correctly identify 82 percent of brain tumors. This test is also able to correctly identify 84 percent of people who don't have brain tumors.

In the case of the most common form of brain tumor, called glioma, this test is 92 percent accurate to know which tumor someone has. "These results are very promising because they show that our technique can accurately find out who is most likely to have a brain tumor and who might not.

Because this technique only requires small blood samples, if it offers the potential to test large numbers of people with suspicious symptoms and "gives the best indication of who needs an immediate brain scan. This can ultimately speed up diagnosis, reduce anxiety waiting for tests and get patients treated as quickly as possible," Baker explained. (one / opi)

Posting Komentar

0 Komentar