GE Healthcare and the University of California, San Francisco (UCSF) Center for Digital Health Innovation (CDHI) are teaming up to develop a library of deep-learning algorithms.
Initially, the collaboration is focusing on areas of high-volume and high-impact medical imaging to create algorithms that distinguish between what is considered a normal result and what requires follow-up, according to UCSF and GE. For example, one algorithm currently being developed will focus on pneumothorax, teaching machines how to distinguish between normal and abnormal imaging studies so that clinicians can prioritize and move quickly to treat patients with this life-threatening condition.
Other planned algorithms will offer the potential for tasks such as predicting patient trajectories, automating the triage of routine care, improving process efficiency, and enabling the development of more personalized therapies, UCSF and GE said. The goal is to help providers improve diagnostic accuracy and patient outcomes, as well as enhance clinical workflows and productivity, they said.
"Next-generation data science techniques have already transformed the industrial and consumer world," UCSF CDHI Director Dr. Michael Blum said in a statement. "With this collaboration, these technologies will be applied to our clinical data and images to provide clinicians with actionable information in near real-time. Together, we will develop tools and algorithms that will allow clinicians and researchers to identify problems and ask questions that are only achievable with vast computing power and datasets."
These algorithms can be deployed worldwide via the GE Health Cloud and "smart" GE imaging machines to share the research with clinicians around the world, UCSF and GE said. As the partnership progresses, they will integrate data from a variety of imaging technologies and also incorporate clinical datasets from the electronic health record and other sources to enrich the development of algorithms and improve sensitivity.
