A.I. Took Test To Detect Lung Cancer And Smashed It
Artificial Intelligence (A.I.) may be able to help doctors detect lung cancer as the tech took a test to spot the disease, passing with flying colours.
The study was conducted by researchers from Google and several medical centres, and tested whether A.I. was more successful at diagnosing cancer than the human eye of radiologists.
Researchers applied artificial intelligence to CT scans, used to screen people for lung cancer. CT scans can spot definite cancers, as well as anomalies which may develop into cancer, however it can miss tumors, or mistake benign spots for malignancies.
The A.I. was trained by researchers who provided CT scans from patients whose diagnoses were known, and then they began to test the technology’s own diagnostic skills.
According to the New York Times, Dr Daniel Tse, a project manager at Google, said:
The whole experimentation process is like a student in school. We’re using a large data set for training, giving it lessons and pop quizzes so it can begin to learn for itself what is cancer, and what will or will not be cancer in the future.
We gave it a final exam on data it’s never seen after we spent a lot of time training, and the result we saw on final exam — it got an A.
A.I. was 94 per cent accurate when tested against 6,716 cases with known diagnoses. When up against six expert radiologists, with no prior scan available, the A.I. beat the doctors in diagnosing. When there was an earlier scan available, however, A.I. and doctors were neck and neck.
The system would hopefully work to help doctors with diagnosis, though it must be studied and tested rigorously before being used in clinics.
If you have a story you want to tell, send it to [email protected]