Artificial intelligence predicts the progress and spread of cancer

Scientists have used artificial intelligence to predict how cancer will progress and spread, which could help doctors make a treatment that is most effective for each patient.

The research group of the London Cancer Research Institute (ICR) and the University of Edinburgh have developed a new technique called Revolver (repeat cancer development) , selecting DNA mutations in cancer and using pine believe to predict future genetic changes.

They said that the ever-changing nature of tumors is one of the biggest challenges to treatment - with cancer often developing into resistance.

But if doctors can predict how the tumor will develop, they can intervene sooner to stop the tumor before it has a chance to develop or become resistant, increasing the chances of survival. multiply.

The team also found an association between repeating tumor mutations and survival rates.

This suggests that repeated DNA mutations can be used as a prognostic indicator, helping shape future treatment.

We can use this AI tool to intervene at an earlier stage of the tumor.

For example, the researchers found that patients with breast tumors had a series of errors in the genetic material that encodes p53 tumor-suppressing proteins , followed by mutations in chromosome 8, with extra life less than other similar genetic changes.

The team has developed a new machine learning technology to convey knowledge about tumors in similar patients.

This method has identified samples in the order of genetic mutations that appear in tumors that are repeated both in and between the tumors of the patient, taking a sample of mutations of the tumor to predict other patterns. .

Researchers who used 768 acres of tumors from 178 patients were reported in previous studies of lung cancer, breast and kidney cancer, and analyzed data in each type of cancer to detect. exactly and compare changes in each tumor.

By identifying repeating patterns and combining this with current knowledge about biology and the evolution of cancer, scientists can predict future tumor growth trajectories.

If tumors with certain patterns are found to be resistant to a specific treatment, this new method may be used to predict whether patients will be resistant to treatment in the future.

The study was published in Nature Methods.

Dr. Andrea Sottoriva, head of the research team, said: 'We have developed a great artificial intelligence tool . With this tool, we hope to eliminate one of the biggest challenges of Cancer is an unpredictable development ".

"By providing an insight into the future, we can use this AI tool to intervene at an earlier stage, predicting the next development of cancer."

According to ICR CEO, Professor Paul Workman, the evolution of cancer is the biggest challenge we face while seeking more effective treatments for patients."If we can predict how the tumor will grow, the treatment can be changed before adaptation and resistance develop," he said.