Open AccessEfficient evaluation of low-contrast detectability of deep-CNN-based CT reconstruction using channelized Hotelling observer on the ACR accreditation phantom
Author(s) -
Mingdong Fan,
Zhongxing Zhou,
Thomas J. Vrieze,
Jia Wang,
Cynthia H. McCollough,
Lifeng Yu
Publication year - 2022
Publication title -
medical imaging 2022: physics of medical imaging
Language(s) - English
Resource type - Conference proceedings
SCImago Journal Rank - 0.192
H-Index - 176
pISSN - 0277-786X
DOI - 10.1117/12.2612414
Subject(s) - imaging phantom , channelized , computer science , artificial intelligence , observer (physics) , iterative reconstruction , computer vision , contrast (vision) , image quality , deep learning , convolutional neural network , noise reduction , pattern recognition (psychology) , image (mathematics) , nuclear medicine , medicine , telecommunications , physics , quantum mechanics
As deep-learning-based denoising and reconstruction methods are gaining more popularity in clinical CT, it is of vital importance that these new algorithms undergo rigorous and objective image quality assessment beyond traditional metrics to ensure diagnostic information is not sacrificed. Channelized Hotelling observer (CHO), which has been shown to be well correlated with human observer performance in many clinical CT tasks, has a great potential to become the method of choice for objective image quality assessment for these non-linear methods. However, practical use of CHO beyond research labs have been quite limited, mostly due to the strict requirement on a large number of repeated scans to ensure sufficient accuracy and precision in CHO computation and the lack of efficient and widely acceptable phantom-based method. In our previous work, we developed an efficient CHO model observer for accurate and precise measurement of low-contrast detectability with only 1-3 repeated scans on the most widely used ACR accreditation phantom. In this work, we applied this optimized CHO model observer to evaluating the low-contrast detectability of a deep learning-based reconstruction (DLIR) equipped on a GE Revolution scanner. The commercially available DLIR reconstruction method showed consistent increase in low-contrast detectability over the FBP and the IR method at routine dose levels, which suggests potential dose reduction to the FBP reconstruction by up to 27.5%.