Researchers out of UC San Francisco developed the algorithm using deep learning, a type of AI modeled on how the human brain processes information. To train the algorithm to identify early signs of Alzheimer’s disease, they supplied it with images from the positron emission tomography, or PET, scans of nearly 1,000 patients’ brains.
The deep learning algorithm achieved 82 percent specificity at 100 percent sensitivity at predicting patients’ final clinical diagnosis by radiologic readers. On average, the algorithm reached its conclusion 75.8 months — or more than six years — prior to a radiologic reader’s diagnosis.
“A deep learning algorithm can be used as an early prediction tool for Alzheimer disease, especially in conjunction with other biochemical and imaging tests, thereby providing an opportunity for early therapeutic intervention,” the study authors concluded.
More articles on artificial intelligence:
Philips launches AI incubator with 19 startups
Mayo Clinic partners with AI firm to screen patients for heart conditions
Scripps partners with Nvidia to apply AI to genomic, digital sensor data
At the Becker's 11th Annual IT + Revenue Cycle Conference: The Future of AI & Digital Health, taking place September 14–17 in Chicago, healthcare executives and digital leaders from across the country will come together to explore how AI, interoperability, cybersecurity, and revenue cycle innovation are transforming care delivery, strengthening financial performance, and driving the next era of digital health. Apply for complimentary registration now.