Of the several health malignancies in men, prostate cancer is a common one – diagnosis for which is typically on the basis of prostate-specific antigen (PSA) – and unfortunately, the accuracy of which is as low as 30%. Knowing the high incidence of unreliability and inaccuracy of PSA-based testing, many patients need to undergo invasive biopsy. This often leaves patients with side effects such as bleeding and pain that have long-term implications.
To address this, recently, a team of researchers at the Korea Institute of Science and Technology developed an AI algorithm, which is combined with an ultrasensitive biosensor based on electrical signal. The advantage of the algorithm lies in its diagnosis of prostate cancer in 20 minutes with almost 100% accuracy.
Smart Biosensor to find use for diagnosis of few cancers
The smart biosensor could also find use for the diagnosis of a number of types of cancers, said one of the associates commenting on the findings. And, all this requires is urine sample of the patient.
In fact, training of the AI system involved using the correlation between the four factors for cancer obtained from the novel biosensor. Once it is ready for use, the algorithm deployed to examine complex patterns of the detected signals.
Meanwhile, researchers used the algorithm to perform tests on 76 urinary samples. This revealed the algorithm to be able to diagnose prostate cancer with near-perfect accuracy. Such an achievement could eventually improve the lives of millions of men worldwide.
And, for patients who need surgery or treatment or a combination of surgery or treatment, the algorithm will diagnosis cancer with high accuracy. It does so by using urine to minimize biopsy and treatments that are not needed, which can reduce medical expenses and fatigue of the medical staff dramatically.