The process of diagnosing cancer currently relies on chemical "contrast agents" to enhance the clarity of medical imaging techniques, including X-rays, CT (computed tomography), and MRI (magnetic resonance imaging) scans. While this method is the gold standard for many disease diagnoses, it brings challenges such as increasing patient numbers, escalating costs, and potential adverse effects. Though rare, these chemical agents can lead to allergic reactions and other immediate negative responses in some patients. In response to these concerns, a new approach using artificial intelligence (AI) is being developed to create virtual contrast-enhanced images that do not require chemical agents, aiming for a process that is safer, faster, and more cost-effective by relying solely on non-contrast images.

Researchers at Case Western Reserve University (CWRU, Cleveland, OH, USA) have received a new four-year, USD 1.125 million grant from the National Science Foundation to develop a novel imaging method that could mitigate the risks associated with chemical contrast materials while also offering cost and time benefits. This advancement is particularly timely, considering recent supply chain challenges, notably the 2022 global shortage of iodinated contrast media.

This project capitalizes on the combined engineering and medical expertise of the research team, aiming to pioneer a virtual method that provides crucial diagnostic data to clinicians and patients without the need for chemical contrast agents. The team is investigating a new category of image features to guide the creation and testing of new models tailored for MRI scans. Utilizing the advanced capabilities of AI, they are developing a technique known as an AI contrast agent which could transform diagnostic imaging.

“Virtual contrast-enhanced imaging could save time and money while continuing to provide the best care to patients,” said project leader Shuo Li, an associate professor at the university’s Case School of Engineering. 

http://www.gzjiayumed.com/en/index.asp