Artificial intelligence has various applications these days. A group of researchers from MIT University created a tool called Sybil It has been shown that it can identify the first invisible signs of lung cancer long before and significantly increase the chances of survival of patients.
Doctors now use low-dose CT scans (LDCT) to screen people with suspected lung cancer. Now, researchers have developed a system with artificial intelligence that can review LDCT scans without the need for the help of a radiologist, and the likelihood of lung cancer in The next six years Diagnose the patient.
The Sybil system uses 3D CT scans and has been able to identify lung cancer patients with high accuracy in tests it has undergone in several clinics. Researchers to train this system of hundreds CT scan of cancerous tumors used to teach the symptoms of this disease to their artificial intelligence.
Sybil AI can also predict the future
Eventually, Sybil could even detect cancer in people when there were no obvious symptoms. “We found that even when we couldn’t see the cancer, this AI was still able to predict which lungs would eventually become cancerous,” said Jeremy Volvand, one of the senior authors of the paper. that we saw [Sybil] “It can predict the most likely side of the lung for disease formation, it was very interesting for us.”
So far, this system has only been tested on people with a history of smoking, but the researchers want to use it in the future Non-smokers And those who decades ago have quit smoking Also test. The number of infections in these people is increasing, while they are not usually in the group of people at risk, so they are not prioritized in screenings. Therefore, quick identification of the disease in them can be very important.
The results of this research have been published in the Journal of Clinical Oncology.