Prostate cancer is one of the most common cancers in men, diagnosed primarily by measuring a cancerous factor in the blood called “prostate-specific antigen” (PSA). Because the accuracy of this diagnostic method is close to 30%, a large number of people are sampled, and researchers are now trying to prevent this by testing urine and artificial intelligence.
The South Korean Institute of Science and Technology (KIST) recently announced in a Joint research work With the “Biological Research Center” and the “Easy Medical Center” has succeeded in developing a method for diagnosing prostate cancer by urine. This method detects cancer in just 20 minutes and has an accuracy of nearly 100%.
To develop the new detection method, the team has combined an ultra-sensitive bio-sensor based on an electrical signal with an AI-based analysis method. This method can only detect prostate cancer with a urine test and will no longer require more invasive procedures such as biopsy or biopsy.
Because cancer agents may have low concentrations in urine samples, biosensors are only used to classify high-risk groups. The approach to using a cancer agent increases its detection accuracy by more than 90%, but limits its detection.
To counter such a limitation, the researchers used different cancer agents simultaneously to increase the accuracy of the diagnosis. The team was able to develop a super-sensitive semiconductor sensor capable of simultaneously measuring the values of four cancer factors in a urine test.
The team used the link between the four factors collected by the sensor to train their desired artificial intelligence. Then, the artificial intelligence algorithm was used to analyze prostate cancer in urine samples by analyzing complex signal patterns. This method was able to successfully detect prostate cancer in 76 urine samples with an accuracy of nearly 100%.