Researchers at the Stanford University School of Medicine have found that computers trained to detect lung cancer tissue may actually be more accurate than any human doctor. The study’s findings were published in Nature Communications.
“Pathology as it is practiced now is very subjective. Two highly skilled pathologists assessing the same slide will agree only about 60 percent of the time. This approach replaces this subjectivity with sophisticated, quantitative measurements that we feel are likely to improve patient outcomes,” said Michael Snyder, PhD, professor and chair of genetics. 
In order to train the computer to pick up lung cancer, they used 2,186 images from the Cancer Genome Atlas. Individuals whose pictures were used had been diagnosed with adenocarcinoma or squamous cell carcinoma.
Previously, doctors examined tumor tissue mounted on glass slides with a light microscope to determine the severity of the cancer. In layman’s terms, the more abnormal the tissue looks, the more severe the cancer.
Researchers say that this new technology can be incredibly helpful in detecting cancer at any stage. Computers can detect 10,000 traits that are specific to cancer, whereas the human eye can only detect a few hundred. This can mean more precise and accurate diagnoses that won’t be left up to human error.
While scientists at Stanford have only trained the computer to diagnose lung cancer, they say that the technology will work for any type once it is programmed.
“We began the study without any preconceived ideas, and we let the software determine which characteristics are important. In hindsight, everything makes sense. And the computers can assess even tiny differences across thousands of samples many times more accurately and rapidly than a human,” Michael Snyder added in a press release.
The work is part of Stanford University’s focus on precision health. The goal is to be able to not only prevent diseases, but to also be able to catch them early and treat them effectively. A highly accurate diagnosis is one of the first steps to that process. 
 News Medical