Release date: 2016-04-29
Diagnosing cancer based on blood samples is very challenging. Usually, the doctor will make the cancer cells visible by adding chemicals to the sample, but this will also make the sample not continue to be used in other tests. Other diagnostic techniques are based on the abnormal structure of cancer cells, but this consumes more time and may incorrectly identify healthy malformed cells as cancer cells.
Researchers at the University of California, Los Angeles, have developed a new technology that combines special microscopes and artificial intelligence algorithms to identify cancer cells in their samples without loss. This technology not only reduces the time and effort to diagnose cancer, but is also a major achievement in the field of precision medicine.
The microscope used in this technique is called a photon time-extension microscope, which splits the nanosecond pulse into several rays that capture hundreds of thousands of images per second. These images are entered into a computer program that is categorized according to the 16 different physical characteristics of the cells, such as diameter, roundness, and amount of light absorbed.
Using a set of analyzed images, researchers used deep learning techniques to train computer programs to identify cancer cells. After several rounds of testing, the researchers found that the system's recognition ability increased by at least 17% compared to existing analysis tools. Researchers believe their approach will lead to more data-driven cancer diagnostic systems.
By analyzing the patient's genes, deep learning techniques have been used to help diagnose the disease. Since this technology can identify cancer cells that other methods may not recognize, it is likely to help researchers better understand the genetic mutations that cause cancer, creating new ways to treat cancer.
Source: èŒèšª five-line spectrum
Aluminum Panel Security Door,Aluminum Security Door,Aluminium Security Doors,Aluminium Burglar Doors
Zhejiang Tilanco Industry&Trade Group Co., Ltd , https://www.tilancogroup.com