Danville, Pa.-based Geisinger reduced the time it diagnoses intracranial hemorrhages by 96 percent through the use of a computer tool that can read CT scans to detect the rare form of internal head bleeding.
Roughly 50,000 patients each year suffer from intracranial hemorrhages, and 47 percent of them die within 30 days. This means early and accurate diagnosis is crucial to patient safety.
Realizing the challenges in detecting these types of hemorrhages, a team of physicians and researchers leveraged radiographic and other medical imaging data from the EHR to train a computer embedded with a machine learning algorithm to review CT scans and flag the most urgent images for priority review by radiologists.
"This is not about replacing doctors with machines," said Aalpen Patel, MD, chair of Geisinger System Radiology. "This is about the smart use of machine learning technology to aid medical providers in delivering better and faster care, especially in these areas where time is critical."
Other Geisinger researchers are experimenting with machine learning in different departments. For example, Brandon Fornwalt, MD, PhD, director of Geisinger's department of imaging science and innovation, is applying machine learning to patients with congenital heart disease.
"The use of intelligent computer assistance is imperative in order to sustain and improve medical care," Dr. Fornwalt said. "Geisinger is proud to be at the forefront of clinical applications of these technologies."
More articles on artificial intelligence:
Tech companies are designing their own AI-powered chips
'The mind of a mathematician with the heart of a doctor': athenahealth CTO Prakash Khot shares 4 thoughts on AI in healthcare
Analysis: 5 things to know about compensation for AI researchers