A research team at Asan Medical Center (AMC) has developed Image TrialTM, which automatically manages clinical trial image data from collection to extraction.
The team expects the new system will greatly improve clinical trial efficiency when using image data such as computed tomography (CT) and magnetic resonance imaging (MRI) in clinical trials is increasing worldwide.
The system developed by the research team, led by Professor Kim Kyung-won of the Department of radiology, automatically proceeds with a series of processes of collecting and extracting clinical trial image data according to the Clinical Data Interchange Standardization Consortium (CDISC)’s standards.
CDISC is an organization that establishes a standard format for all data used in clinical trials. It is an international data standard format required by regulatory officials in the U.S., Europe, Japan, and China.
"The clinical trial image management system applies CDISC's clinical data collection standards and standard terminology from the data collection stage to standardize electronic case records," the team said. "The input data goes through an automatic verification process first. This process minimizes the error rate such as data omission and maintains the accuracy and consistency of the data."
Afterward, the system extracts the last verified data in CDISC's clinical data model format.
Except for the input of analysis results by an image analysis expert, an automated system based on the CDISC standard performs all other parts from data collection to extraction, it added.
By applying this system to actual clinical trial image data management, the team confirmed that data input and format errors in a project dramatically reduced from 176 to 0.
"We expect the development of CDISC-compliant clinical trial image management system will increase the integrity and reliability of data used in clinical trials, as well as significantly reducing the time required to purify and manage data through an automated method," Professor Kim said.
Kim noted that regulatory agencies, local and foreign pharmaceutical companies, and clinical trial consignees would be able to access image analysis data more easily and efficiently conduct data review and transmission, adding that they confirmed the effectiveness of this system in collaboration with C&R Research, a leading domestic clinical trial agency.
Journal of Biomedical Informatics has published the result of the study.