On single still images drawn from all 15 views, the model achieved an average overall accuracy of 91.7 percent compared to an average of 79.4 percent for board-certified echocardiograpers classifying a subset of the same test images.
Computers Are More Accurate Than Echocardiographers in Interpreting Echocardiograms
Imaging is a critical part of medical diagnosis; Interpreting medical images typically requires extensive training and practice and is a complex and time-intensive process
Deep learning, specifically using convolutional neural networks (CNNs), is a cutting-edge machine learning technique that has proven “unreasonably successful at learning patterns in images and has shown great promise helping experts with image-based diagnosis.
Researchers trained a computer to assess the most common echocardiogram (echo) views using more than 230,000 echo images. They then tested both the computer and human technicians on new samples.
On single still images drawn from all 15 views, the model achieved an average overall accuracy of 91.7 percent compared to an average of 79.4 percent for board-certified echocardiograpers classifying a subset of the same test images
August 6, 2017
Dr. Parker’s Commentary
Deep Learning/Artificial Intelligence is one of the most exciting trends in 21st century medicine.
This has the potential tor revolutionize the field. In something that calls for such complex and numerous tasks of interpretation, it is not surprising that a computer with enough processing power is superior to a mere human.
“Anything that could give rise to smarter-than-human intelligence – in the form of Artificial Intelligence, brain-computer interfaces, or neuroscience-based human intelligence enhancement – wins hands down beyond contest as doing the most to change the world.
Nothing else is even in the same league.”