AI can greatly improve the accuracy of cancer detection from MRI, X-ray and other medical imaging. Approximately 25% of cases of lung cancer diagnosis are missed when doctors check chest x-rays. New AI systems are catching more of the detectable lung cancer cases.
An iPhone-based AI system looking for tongue cancer had a detection accuracy (percentage of correct prediction) of nearly 90%. The detection inference time was around seven milliseconds. These inexpensive systems are usable in low and middle-income countries. 70% of cancer deaths are in low and middle-income countries. Cancer kills ten million people every year.
AI-powered systems have also been able to accurately detect skin cancer in 95% of images of cancerous moles and benign spots, whereas a team of 58 dermatologists was accurate 87% of the time.
Deep learning to digital breast tomosynthesis (DBT: an advanced method for cancer detection in which an X-ray arm sweeps over the breast, taking multiple images in seconds) improves cancer detection. It reduced false-positive recalls compared to screening with digital mammography (DM) alone. New AI cancer detection sensitivity increased from 77% to 85% while cutting detection time by half.
SOURCES – EETimes, British Journal of General Practice, Radiological Society of North America
Written by Brian Wang, Nextbigfuture.com