AsianScientist (Aug. 19, 2021) – Known today as a helpful home convenience and an important industry tool for industry, artificial intelligence (AI) is also aiding the fight against cancer. Publishing their findings in ACS Nano, scientists in Korea have developed a technique that uses AI to accurately diagnose prostate cancer from urine samples within only 20 minutes.
Globally, prostate cancer is the second most commonly diagnosed cancer, with 14 percent of the population affected in Asia. Typically, for an accurate diagnosis, prospective patients must undergo uncomfortable, invasive biopsies that often come with bleeding and pain.
While urine tests can also be undertaken, the cancer factors present in such samples tend to be low, making them useful for identifying risk groups rather than precise diagnosis.
To reduce the need for invasive biopsies, a collaborative team from the Korea Institute of Science and Technology (KIST) and Asan Medical Center led by Dr. Lee Kwan Hyi developed a non-invasive method capable of diagnosing prostate cancer with just a urine sample.
The technique uses an ultrasensitive biosensor that uses electrical signals to simultaneously measure trace amounts of four selected cancer factors in urine. To analyze the complex patterns of the detected signals and confirm prostate cancer diagnosis, the biosensor relies on AI.
Incredibly, the AI-powered biosensor successfully diagnosed prostate cancer from 76 urinary samples with almost 100 percent accuracy, opening the doors for a more accurate and less invasive option for detecting prostate cancer— and maybe even other cancers.
“For patients who need surgery or treatments, cancer can be diagnosed with high accuracy by utilizing urine to minimize unnecessary biopsy and treatments, which can dramatically reduce medical costs and medical staff’s fatigue,” said study co-author Professor Jeong In Gab from Asan Medical Center.
Source: National Research Council of Science & Technology; Photo: Shutterstock.
Disclaimer: This article does not necessarily reflect the views of AsianScientist or its staff.