National Institute of Biomedical Imaging and Bioengineering 5R01EB036082-02

Development and clinical validation of domain-aware anthropomorphic model observers for task-based image quality assessment

This NIH-funded project develops deep-learning anthropomorphic model observers to assess medical image quality, aiming for domain-aware models that generalize across imaging algorithms and datasets, focusing on liver CT imaging.

Description

The project involves developing deep-learning anthropomorphic model observers (AMO) for medical imaging quality assessment. The project aims to create domain-aware AMO that can generalize across different imaging algorithms and datasets. The project will develop extensive annotated datasets from simulated, phantom, and clinical CT imaging of the liver. The project is funded by NIH PA-20-185, with an award amount of $582,908. The project duration is from July 1, 2025, to March 31, 2029.

View original record