A group of cancer research scientists in China, working with two colleagues in the United States, have discovered that artificial intelligence can be used to track metastatic cancer cells that have spread to different parts of the body.
In their study, published in the journal Nature Medicine, the team of researchers from Tianjin Medical University (TMU) in China started from a known fact: many types of cancer are not diagnosed in time, since they go unnoticed until they spread. to distant organs. Therefore, they managed to create an artificial intelligence tool that outperforms pathologists in identifying the origins of metastatic cancer cells found in the body.
Tumors look different depending on where they grow: lung cancer cells look similar, but are different from breast cancer cells. Laboratory technicians learn to recognize cancer cells in fluid samples to identify their type and help doctors treat the original tumor. However, sometimes they cannot detect all cells or identify them correctly.
In this study, experts investigated whether an artificial intelligence (AI) application could be better than laboratory technicians at this task. They trained the AI with images of metastatic cancer cells and found that it was 83% accurate in detecting a type of tumor. They compared the accuracy of the AI with that of human technicians and found that the AI had better results in several areas and with 99% that the origin is among the first three possibilities posed.
The scientists also found that those patients who had received treatment designed to combat their specific cancer were more likely to survive and also tended to live longer, suggesting that using AI to find and track metastatic cells in the body could improve the results of many types of cancer. patients.
Researchers found that patients who received personalized treatment for their specific type of cancer were more likely to survive and live longer. This suggests that using artificial intelligence to detect and track metastatic cells in the body could improve outcomes in patients with different types of cancer.
igomez@diariolas amerricas
Source: Medicalxpress.com, ludd.es, La Vanguardia,
