https://www.fda.gov/MedicalDevices/ScienceandResearch/ResearchPrograms/ucm477418.htm
VICTRE in 3
-
What is VICTRE? VICTRE is a research program aiming at demonstrating that computational modeling can play an increasingly predominant role in the regulatory assessment of imaging products.
-
Rationale: Expensive and lengthy clinical trials delay regulatory evaluation. This burden can stiffle innovation affecting patient access to novel, high-quality imaging technologies.
-
Strategy: We propose an in silico replication of an existing clinical trial with demonstration of savings and benefits for stakeholders.
Summary
Powerful and open-source radiation imaging system simulation and image analysis tools are now becoming available to industry, academia, and government researchers, allowing for a greater understanding of the effect of system design and modifications on the performance of new imaging technologies. These tools have the potential to facilitate less-burdensome regulatory evaluation and rapid deployment of meritorious imaging devices while demonstrating significant pitfalls in defective designs. In this program, we aim to demonstrate the benefits of computational modeling for entire imaging chains in comparison to traditional methods that rely on burdensome animal and human testing. Work in this area includes the generation of accurate, realistic, and computationally efficient radiation imaging system models for advanced detector and novel modality designs; and patient models that capture the variability and multi-level complexity of human anatomy and physiology in normal and diseased states. In addition, we are developing observer models that track or predict the performance of the image interpreter for clinically meaningful tasks, including early detection of abnormalities and quantitative imaging biomarkers. Finally, we are developing computational tools and experimental prototypes that have the potential to significantly advance x-ray imaging technology.
Current areas of focus include the development of Monte Carlo simulation codes than can simulate radiographic images and detector response with high accuracy, the design of high-resolution, virtual anatomical models, the development of patient-specific dynamic x-ray filters, the investigation of new imaging modalities based on analysis of the information encoded in scattered x rays and the research of computational observers for clinically relavant diagnostic tasks.
Our group has a history of contributing open-source modeling tools to the imaging community. Our current goal is to demonstrate that virtual clinical trials for the regulatory evaluation of emerging medical imaging technologies can, in some cases, supplant or significantly minimize the need for burdensome studies that rely on the radiation exposure of patients and image interpretation by humans.