Radiologists aren't going to be replaced by computers tomorrow, but Denver-area radiologist and Forbes contributor Paul Hsieh acknowledges the rise of deep learning AI technology has opened the door to medical work previously thought to be out of reach of automation.
"If you asked me 10 years ago, I would have said, 'No way!' But if you ask me today, my answer would be more hesitant, 'Not yet—but perhaps someday soon.'" he wrote. "For example, deep learning algorithms have been able to diagnose the presence or absence of tuberculosis (TB) in chest x-ray images with astonishing accuracy."
The promise of deep learning extends beyond radiology, too. Researchers from Google trained AI to detect breast cancer on microscopic slides of lymph node tissue, identifying cancer much more accurately than human pathologists.
However, the most interesting use of deep learning technology is uncovering associations yet undiscovered by human scientists. For example, a Stanford University study pitted the predictive value of American College of Cardiology (ACC)/American Heart Association (AHA) risk factors for cardiovascular disease against the predictive value of several artificial intelligence algorithms.
All four AI methods performed significantly better than the ACC/AHA guidelines,” wrote Science contributor Matthew Hutson. “In the test sample of about 83,000 records, that amounts to 355 additional patients whose lives could have been saved.”
Read the full article at Forbes by following the link below: