By Aine Cryts
In 1975, engineers and physicians at University of California, San Francisco (USCF) helped develop a technology that’s known today as MRI. (Eight years later, the technique for producing clear, dramatic images using MRI was featured at RSNA ’83.)
Fast-forward 44 years and UCSF announced in October the launch of a new research center to accelerate the use of artificial intelligence (AI) in radiology. The Center for Intelligent Imaging, or ci2, will use advanced computational techniques and industry collaborations to impact patient diagnoses and care. The research center’s investigators will partner with Santa Clara, Calif.-based technology firm NVIDIA Corp. to create infrastructure and tools focused on enabling the translation of AI into clinical practice.
Sharmila Majumdar, PhD, who leads operations at the research center, tells AXIS Imaging News that ci2’s work “will enable radiologists to play a pivotal role in precision- and individually tailored medicine.” Areas of research that are of interest to radiologists include AI image analysis, and data analytics.
“The volume of radiology data is growing exponentially, and given the complexity and the vast amounts of information contained in 3D images, radiologists are often not exploiting the full potential of imaging,” adds Majumdar, who’s also professor and vice chair of the department of radiology and biomedical imaging at UCSF.
At ci2, Majumdar will lead a National Institutes of Health-funded study to evaluate chronic back pain. The study will use AI-fueled algorithms, data analysis, quantitative sensory assessments, brain imaging, and biomechanical evaluation of the spine.
The research center’s partnership with NVIDIA is the result of conversations between the two organizations over the past two years. “The collaboration not only includes the deployment of [NVIDIA’s] state-of-the-art hardware and software, but also interacting with the company’s computer architects, systems engineers, and data scientists on-site, working together to make medical imaging a core part of patient care, and to improve healthcare worldwide,” Majumdar tells AXIS.
London-based machine learning company Kheiron Medical Technologies, Ltd. is the first start-up to partner with ci2 on AI imaging research and development. Kheiron will work with the new research center’s breast imaging group to validate that the company’s breast cancer screening software can be safely and feasibly used in ethnically diverse populations.
Aine Cryts is a contributing writer for AXIS Imaging News.