Currently, doctors use cancer imaging tests to answer a range of questions, such as: is it a real cancer or just a harmless lump If it’s cancer, how fast does it grow How far has it spread Will it grow back after treatment.
Studies show that artificial intelligence has the potential to improve the speed, accuracy and reliability with which doctors answer these questions.
“Artificial intelligence can automate assessments and tasks that humans can currently do but take a lot of time,”Hugo Aerts, of Harvard Medical School, said in an interview to the website of the National Cancer Institute in the United States.
And after the AI gives the result, ” the radiologist only needs to review what the AI has done, to see if he made the correct assessment?””The same speaker continues.
Perhaps the most interesting thing for scientists is the possibility of artificial intelligence going beyond what humans can currently do on their own, as it can “see” things that we humans cannot, and it can find complex patterns and relationships between very different types of data.
“Artificial intelligence is great at surpassing human performance for many tasks,”Aerts says in connection with this.
But it is still unclear how AI reaches its results, so it is currently difficult for doctors and researchers to verify whether its tools are working correctly or not.
