Advanced artificial intelligence (AI) tools and techniques allow researchers to use the vast amounts of data now available from electronic health records and wearables to help solve complex health problems and improve population health.
To further integrate the power of AI into health care, the McKelvey School of Engineering at Washington University in St. Louis has launched the AI for Health Institute to design data-driven tools to characterize complex diseases, support clinical decisions and drive precision health. The institute was introduced at the AI & Digital Health Summit, which took place Oct. 18-19 at Washington University.
Chenyang Lu, the Fullgraf Professor of Computer Science and Engineering at the McKelvey School of Engineering and director of the institute, said the AI for Health Institute intends to establish Washington University as a leader in AI for health in a very competitive field.
“This is the new frontier of health care,” said Lu, who also holds appointments as a professor of anesthesiology and of medicine in the School of Medicine. “Given the complexity of the problems and the messiness of the data, basic AI tools are insufficient to solve a lot of these problems. That’s where cutting-edge AI comes in.”
Lu said AI tools and techniques are being developed at an explosive rate, but there has historically been a barrier between the health and engineering and AI communities. In contrast, the new institute will collaborate with the Institute for Informatics, Data Science & Biostatistics (I2DB) at the School of Medicine.
“The new AI for Health Institute represents an exciting opportunity to more formally connect the School of Engineering with what I2DB is doing at the School of Medicine, facilitating cross-cutting collaborations around AI, particularly at this pivotal moment where we have a tremendous opportunity to impact the greater good and the overall health of our communities,” said Philip R. O. Payne, the Janet and Bernard Becker Professor, associate dean for health information and data science, chief data scientist at the School of Medicine and professor of computer science and engineering at McKelvey Engineering.
“Our plan is to grow collaborative teams across engineering and health, create a competitive edge in recruiting, build an organization and infrastructure for large research initiatives and translate AI to health care,” Lu said.
Establishing the institute lays the groundwork for interdisciplinary research initiatives that will leverage four initial cores: equity, fairness and privacy in AI; wearables for health care; imaging AI; and natural language processing. At its beginnings, the institute’s research focuses primarily on neurosurgery, perioperative care, mental health care, digital pathology, telemedicine and critical care, reproductive health care, and infectious diseases. As the institute grows, more cores and focus areas will be developed.
“The AI for Health Institute will provide outstanding resources to accelerate improvements in health by harnessing big data, machine learning technology and artificial intelligence for use in health care delivery and health services research,” said Victoria J. Fraser, MD, the Adolphus Busch Professor of Medicine and chair of the Department of Medicine at the School of Medicine.
In addition to a 12-member steering committee, the institute has 64 faculty members, with 37 from the School of Medicine, 23 from McKelvey Engineering and four from other schools.
Lu and other faculty members who are part of the institute have been doing groundbreaking work in these areas over the past few years with funding from the National Institutes of Health (NIH), the Agency for Healthcare Research and Quality, BJC HealthCare and the Fullgraf Foundation. Lu and collaborators, many of whom are faculty members of the AI for Health Institute and I2DB, have used Fitbit wearables to detect mental health disorders in the community, monitor potential complications after pancreatic cancer surgery and other surgeries. They also used AI predictions to support care during surgery, such as using electronic health record data to help predict and identify patients at risk for complications from surgery. In addition, they were able to look at electronic health record logs to predict physician burnout.
Originally published by the McKelvey School of Engineering