A Rapid AI-based Screening Test for Prostate Cancer Biomarkers in Urine

A team of Korean investigators developed a highly accurate rapid technique for diagnosing prostate cancer based on AI (artificial intelligence) monitoring of urinary biomarkers.

Screening for prostate cancer generally relies on the serum prostate-specific antigen (PSA) test, which unfortunately generates a high rate of false positive results (up to 80%). This generates a large number of unnecessary biopsies with resultant unpleasant side effects, such as bleeding and pain. Considering how often the PSA test is used, there is a critical unmet need for a precision screening method for diagnosis of prostate cancer.

Toward this end, investigators at the Korea Institute of Science and Technology (Seoul) and their collaborators developed a technique for diagnosing the disease from urine by utilizing an electrical-signal-based ultrasensitive biosensor. The semiconductor sensor system was capable of simultaneously measuring trace amounts of four selected cancer factors in urine. The correlation of clinical state with the sensing signals from urinary biomarkers was analyzed by two common machine learning algorithms. As the number of biomarkers was increased, both algorithms provided a monotonic increase in screening performance.

When applied to urine samples from patients, the 20-minute AI analysis successfully detected 76 urinary samples with greater than 99% accuracy.

Senior author Dr. Kwan Hyi Lee, a principal research scientist at the Korea Institute of Science and Technology, said, "This research developed a smart biosensor that can rapidly diagnose prostate cancer with almost 100% accuracy only through a urine test, and it can be further utilized in the precise diagnoses of other cancers using a urine test."

The AI-based assay for diagnosis of prostate cancer was described in the December 9, 2020, online edition of the journal ACS Nano.

Related Links:
Korea Institute of Science and Technology