Scientists at the University of California, Los Angeles, have developed a computer program that can detect cancer from a patient’s blood sample. Scientists call the program Cancer Locator or CancerLocator, which works by measuring the DNA of a tumor that circulates in the bloodstream.
The program works by detecting specific molecular strands of DNA that move freely in the patient’s bloodstream. This program compares the number and type of arrays with the types of cancers it has in the database. The DNA of cancer cells is generally known to end up in the bloodstream and this program is very appreciative because it can find cancer at very early stages.
“The higher the concentration of tumor DNA in the bloodstream, the easier and more accurate the program will produce diagnostic results,” says UCLA professor Jasmin Zou. “For this reason, tumors in organs through which more blood circulates, such as the liver, are easier to find, while in breasts through which less blood flows, they are harder to detect.”
In the study, a new computer program and a Vector machine were tested with blood samples from 29 patients with liver cancer, 12 with lung cancer and 5 with breast cancer. Testing was performed 10 times on each sample to confirm the results. The vector machine method had an error coefficient of 0.604 while with the new program the error was much lower 0.265
The scientists also performed tests on healthy people and compared them with tests of people who are sick, as well as with vector machine tests. Again, an error of 0.265 was proved.
The results of this research were published in the journal Genome Biology
Information obtained from: Kang, S., Li, Q., Chen, Q., Zhou, Y., Park, S., & Lee, G. et al. (2017). CancerLocator: non-invasive cancer diagnosis and tissue-of-origin prediction using methylation profiles of cell-free DNA. Genome Biology, 18 (1). doi: 10.1186 / s13059-017-1191-5