On Monday, Google Artificial Intelligence (AI) researchers and medical researchers published a study showing that they have been successful in training a deep learning algorithm that has the potential to detect lung cancer and that too with a 94.4% success rate.
The study was published Monday in the journal Nature Medicine.
According to the researchers, the findings indicated that apart from a high accuracy rate, the deep learning algorithm literally outclassed the radiologists under certain circumstances.
The team conducted two studies. In one study, a prior scan was available, while in another, it was not.
The algorithm was trained on computed tomography (CT) scans of people who have lung cancer, without it, and with nodules turned cancerous.
The first study in which the prior scan was available found that the deep learning algorithm had a higher identification rate when compared with six radiologists. The second study in which the prior scan was unavailable found that the machine and humans were even.
The project manager at Google Dr. Danile Tse told the New York Times (NYT), “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.”
However, the findings are still far from being proved as accurate enough to use the algorithm across medical institutions for cancer screening. The study does offer some promising findings.
The director of the Scripps Research Translational Institute in California Dr. Eric Topol said, “Lung CT for smokers, it’s so bad that it’s hard to make it worse.”
Many tech companies, including Google, have already started seeking algorithms as a tool for detection of a variety of medical conditions. At the same time, researchers are working on this type of technology to acknowledge the risks of using such a system. Dr. Tse told the NYT, “We are collaborating with institutions around the world to get a sense of how the technology can be implemented into clinical practice in a productive way. We don’t want to get ahead of ourselves.”
