Low- and Zero-dose Contrast-enhanced MRI Using Deep Learning

Project Summary MotivationGadolinium based contrast agentsGBCAsare used in approximately a third of all MRI scansThe unique relaxation parameters of GBCAs create indispensable image contrast for a wide range of clinical applicationssuch as angiography and tumor detectionHoweverthe usage of GBCAs has been linked to the development of nephrogenic systemic fibrosisNSFNSF can be painfulcause severe disabilityand even deathThe risk of developing NSF prevents millions of patients with advanced chronic kidney diseaseCKDfrom receiving contrast enhanced MRI examsThe recent identification of gadolinium deposition within the brain and body has raised additional safety concerns about the usage of GBCAsStudies have demonstrated increased signal intensity on the unenhanced Tweighted MR images that is correlated with previous GBCA exposureand this gadolinium retention is independent of renal functionWhile initial reports focused on linear GBCAsmore recent reports show that gadolinium deposition occurs with macrocyclic GBCAs as wellalbeit at lower levelsFDA has recently issued warnings about gadolinium retention following contrast enhanced MRIand required GBCA manufacturers to conduct human and animal studies to further assess the safety of these contrast agentsThis project addresses these concerns by developing low dose and zero dose contrast enhanced MRI using artificial intelligenceAIand deep learningDLApproachThis fast track project has two phases and three aimsAimPhase Iis to develop a DL method that can synthesize full dose contrast enhanced MR images using pre contrast images and contrast enhanced images acquired with onlyof standard GBCA doseA software infrastructure will be constructed to seamlessly integrate the DL software between MR scanners and PACSAimPhase IIis to develop a DL method that can synthesize full dose contrast enhanced MR images using GBCA free acquisitions with different image contrastIn AimPhase IIwe will clinically validate and evaluate both low dose and zero dose DL methodsincluding on patients with mildto moderate CKDNon inferiority tests and diagnostic performance of the synthesized full dose images compared to the true full dose images will be performedSignificanceThis work will lead to safer contrast enhanced MRIThe low dose and zero dose contrast enhanced MRI method will benefit not only millions of patients with advanced CKDwho cannot currently undergo contrast enhanced MRIbut many more patients with normal kidney functionwho are at the risk of gadolinium retention after contrast enhanced MRI Project Narrative Gadolinium based contrast agentsGBCAsare widely used in MRI exams to create indispensable image contrast for monitoring treatment and investigating pathology and functionHoweverthe usage of GBCAs has been linked to the development of nephrogenic systemic fibrosispreventing patients with advanced chronic kidney disease from receiving contrast enhanced MRI examsas well as potential gadolinium deposition in the body and brain for patients with normal kidney functionThis project aims to address these problems by developing and validating low dose and zero dose contrast enhanced MRI using deep learning!