Artificial intelligence calculates the spread of cancerous tumors

AI could enable more effective treatment for cancer

Cancer is a deadly disease affecting more and more people around the world. Researchers have now predicted, with the help of artificial intelligence, how the disease progresses and spreads further. This could help doctors in the future to develop the most effective treatment for each patient.

Scientists at the Institute of Cancer Research London (ICR) and the University of Edinburgh have found in their current research that artificial intelligence can predict how cancer will evolve and spread. The physicians published the results of their study in the English language journal "Nature Methods".

AI can predict future genetic changes

The newly developed technique is called by the experts revolver (Repeated Evolution of Cancer). They identify patterns in the DNA mutation within cancers. The information can then be used to predict future genetic changes.

The ever-changing structure of tumors is one of the biggest challenges in the treatment of cancer, explain the experts. However, if physicians were able to predict how a tumor will develop, they could intervene earlier to medically stop the cancer as it progresses, before it can develop or develop resistance. This would significantly increase the chance of survival for patients.

Patterns of mutation of tumors can be predicted

The research team also identified a link between certain sequences of repeated tumor mutations and survival probability. This suggests that repeating patterns of DNA mutations could be used as an indicator of prognosis to shape the future treatment of the disease. The physicians developed a new machine learning method that transfers knowledge about tumors to similar patients. This method identifies patterns in the order in which genetic mutations occur in tumors, using the pattern of mutation of one tumor to predict the pattern of another tumor.

Samples from 178 patients were examined

The researchers used 768 specimens from tumors from 178 patients already used in previous studies for lung, breast, kidney and colon cancer. They analyzed the data within each cancer to accurately identify and compare changes in the tumors. By identifying repetitive patterns and combining them with current knowledge about cancer biology and cancer development, scientists could predict the future development of tumors. If tumors with certain patterns develop resistance to treatment, this new method could be used to predict whether certain patients will be affected in the future.

AI can make predictions about the development of tumors

A powerful tool for artificial intelligence has been developed, which can make predictions about the future developmental stages of tumors based on specific patterns, says study author Dr. Andrea Sottoriva from the Institute of Cancer Research London. This also applies to mutations that have hitherto remained hidden in complex data sets.

Cancer often develops unpredictably

Hopefully this new tool will undermine one of the great dangers of cancer - the fact that it usually develops unpredictably, the scientists explain. The tool could be used in the future to intervene at an earlier stage and to predict the next step of the cancer.

Predicting the development of tumors allows for more effective treatment

If it can be predicted how a tumor will develop, treatment could be changed before adaptation and drug resistance occur, so physicians would always be one step ahead of the cancer. This new approach, which uses an AI, could allow the treatment to be personalized in more detail, earlier than before, and tailored to the characteristics of each tumor. In addition, predictions could be made about how a tumor will develop in the future, the experts emphasize. (As)