Head to head comparison reveals that doctor are significantly better than symptom checker apps at accurate diagnosis
Doctors are still much better than symptom-checker apps when it comes to diagnosing illnesses.
Don't Miss: The Best Presidents' Day Sales 2017
In a head to head comparison, doctors have outperformed the apps by a significant margin showing that humans are still superior to machines despite the fact that AI has come a long way in the past few years.
To determine the diagnostic accuracy, a multi-institute team of researchers led by Harvard Medical School asked 230 doctors to diagnose health conditions, both common and uncommon, based on 45 clinical vignettes. The conditions ranged from simple to complicated ones with varying degrees of severity. The vignette included a brief description of symptoms and the patient’s medical history but no results from a physical examination or blood tests.
A total of 23 symptom checkers also went through the same situation. These are some of the most popular online apps used by people to identify the nature of their illness. By analyzing the outcome, researchers found that doctors were able to correctly diagnose more cases than apps. More specifically, doctors made a correct diagnosis more than twice as often as the symptom checkers. The difference was even more dramatic when the condition was more severe and uncommon.
“The reason we were interested in these symptom checkers is that they are used a lot. There’s more than 100 million uses of such tools.” Senior study investigator Dr. Ateev Mehrotra from at Harvard Medical School told CBS News.
A correct diagnosis is compulsory for treating a disease properly and in a timely manner. The failure to recognize a disease can lead to serious consequences in terms of patient’s health. In the test, doctors were also not 100% accurate. They made errors in about 15 percent of cases.
“Clinical diagnosis is currently as much art as it is science,” said Mehrotra. “While the computer programs were clearly inferior to physicians in terms of diagnostic accuracy, it will be critical to study future generations of computer programs that may be more accurate.”