Home https://server7.kproxy.com/servlet/redirect.srv/sruj/smyrwpoii/p2/ Health https://server7.kproxy.com/servlet/redirect.srv/sruj/smyrwpoii/p2/ Cancer can be diagnosed accurately using a urine test with artificial intelligence

Cancer can be diagnosed accurately using a urine test with artificial intelligence



Cancer can be diagnosed accurately using a urine test with artificial intelligence

The set of sensory signals collected for each patient was then analyzed using ML to screen the patient for PCa. 76 urine samples were measured three times, thereby generating 912 biomarker signals or 228 sets of sensing signals. We used RF and NN algorithms to analyze the multimarker signals. Both algorithms provided increased accuracy, and AUROC increased in size as the number of biomarkers increased. Credit: Korea Institute of Science and Technology (KIST)

Prostate cancer is one of the most common cancers among men. Patients are determined to have prostate cancer primarily based on PSA, a cancer factor in the blood. However, since diagnostic accuracy is as low as 30%, a significant number of patients undergo additional invasive biopsy and thus suffer from resulting side effects such as bleeding and pain.


The Korea Institute of Science and Technology (KIST) announced that the collaborative research team led by Dr. Kwan Hyi Lee of the Biomaterials Research Center and Professor In Gab Jeong of the Asan Medical Center developed a technique for diagnosing prostate cancer from urine within just 20 minutes with almost 100% accuracy. The research team developed this technique by introducing a smart AI analysis method to an electrical signal-based ultra-sensitive biosensor.

As a non-invasive method, a diagnostic test using urine is convenient for patients and does not require invasive biopsy, whereby cancer is diagnosed without side effects. As the concentration of cancer factors is low in the urine, urine-based biosensors are used only for classification of risk groups rather than for accurate diagnosis so far.

Dr. Lee’s team at KIST has been working towards developing a technique for diagnosing disease from urine with an electric signal-based ultra-sensitive biosensor. A single cancer factor approach associated with a cancer diagnosis was limited to increasing the diagnostic accuracy to over 90%. To overcome this limitation, the team simultaneously used different types of cancer factors instead of just using one to improve the diagnostic accuracy innovatively.

The team developed an ultra-sensitive semiconductor sensor system that could simultaneously measure trace amounts of four selected cancers in the urine to diagnose prostate cancer. They trained AI using the correlation between the four cancer factors that were obtained from the developed sensor. The trained AI algorithm was then used to identify those with prostate cancer by analyzing complex patterns of the detected signals. The diagnosis of prostate cancer using the AI ​​assay successfully revealed 100 urine samples with almost 100 percent accuracy.

“For patients in need of surgery and / or treatments, cancer will be diagnosed with high accuracy using urine to minimize unnecessary biopsy and treatments, which can dramatically reduce medical costs and medical staff fatigue,” said Professor Jeong at Asan. Medical Center. “This research developed a smart biosensor that can only diagnose prostate cancer with almost 100 percent accuracy only through a urine test, and it can be further used for the accurate diagnosis of other cancers via a urine test,” said Dr. Lee at KIST.


Stop smoking, your bladder thanks you


More information:
Hojun Kim et al., Noninvasive Precision Screening of Prostate Cancer by Urinary Multimarker Sensor and Artificial Intelligence Analysis, ACS Nano (2020). DOI: 10.1021 / acsnano.0c06946

Provided by the National Research Council of Science & Technology

Citation: Cancer can be accurately diagnosed using an artificial intelligence urine test (2021, January 21) retrieved January 21, 2021 from https://phys.org/news/2021-01-cancer-precisely-urine-artificial-intelligence. html

This document is subject to copyright. Except for fair trade for private examination or research, no parts may be reproduced without written permission. The content is provided for informational purposes only.




Source link