A team from the Korea Institute of Science and Technology (KIST) announced that they had created a machine capable of detecting prostate cancer very quickly and with a high success rate, using only a urine analysis.
The system is based on an Artificial Intelligence algorithm, a sensor based on a special semiconductor and sensitive to the point of detecting the data necessary for AI to analyze and correlate. The urine sample analysis process takes just 20 minutes and, in the first tests, it had a success rate close to 100%.
Human oncologists achieve a success rate of just 30%. Current methods for detecting this cancer involve biopsy, invasive techniques and laboratory work. Since not everyone who is subjected to this procedure would need to do so, they are being exposed to many other risks, such as hospital infections, pain and internal bleeding.
The novelty of the KIST team was to focus only on urine analysis and set up a system sensitive enough to be able to detect the so-called ‘cancer factors’. The concentration of these factors is very low, so it does not reveal positive cases when current methods of diagnosis are used, explains The Next Web.
The institution’s press release states that four factors obtained from the developed sensor are correlated. The diagnosis using the algorithm was made in 76 urine samples.
The most recent data show that 1 in 41 men will die from this type of cancer and that it is even the second most deadly cancer among the male population. A simpler, less painful detection method can thus help save many lives.