Micro-Networks for Robust MR-Guided Low Count PET Imaging

• Published in: IEEE transactions on radiation and plasma medical sciences Vol. 5; no. 2; pp. 202 - 212
• Main Authors: Costa-Luis, Casper O. da, Reader, Andrew J.
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.03.2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Artificial neural networks; Convolutional neural network (CNN); Deep learning; deep learning (DL); Ground truth; guided reconstruction; Image classification; Image processing; Image quality; Image reconstruction; Image resolution; Image segmentation; machine learning; magnetic resonance (MR); Magnetic resonance imaging; Mathematical model; maximum-likelihood expectation maximization (MLEM); Medical imaging; Neural networks; Noise; Noise reduction; Phantoms; Positron emission; Positron emission tomography; positron emission tomography (PET); Quality control; Recovery (Medical); Regularization; Regularization methods; resolution modeling (RM); resolution recovery; Robustness; Tomography; Training; Training data; image processing; resolution recovery; positron emission tomography (PET); guided reconstruction; Convolutional neural network (CNN); image reconstruction; machine learning; magnetic resonance (MR); resolution modeling (RM); deep learning (DL); maximum-likelihood expectation maximization (MLEM)
• ISSN: 2469-7311; 2469-7303
• DOI: 10.1109/TRPMS.2020.2986414

Noise suppression is particularly important in low count positron emission tomography (PET) imaging. Post-smoothing (PS) and regularization methods which aim to reduce noise also tend to reduce resolution and introduce bias. Alternatively, anatomical information from another modality such as magnetic resonance (MR) imaging can be used to improve image quality. Convolutional neural networks (CNNs) are particularly well suited to such joint image processing, but usually require large amounts of training data and have mostly been applied outside the field of medical imaging or focus on classification and segmentation, leaving PET image quality improvement relatively understudied. This article proposes the use of a relatively low-complexity CNN (micro-net) as a post-reconstruction MR-guided image processing step to reduce noise and reconstruction artefacts while also improving resolution in low count PET scans. The CNN is designed to be fully 3-D, robust to very limited amounts of training data, and to accept multiple inputs (including competitive denoising methods). Application of the proposed CNN on simulated low (30 M) count data (trained to produce standard (300 M) count reconstructions) results in a 36% lower normalized root mean squared error (NRMSE, calculated over ten realizations against the ground truth) compared to maximum-likelihood expectation maximization (MLEM) used in clinical practice. In contrast, a decrease of only 25% in NRMSE is obtained when an optimized (using knowledge of the ground truth) PS is performed. A 26% NRMSE decrease is obtained with both RM and optimized PS. Similar improvement is also observed for low count real patient datasets. Overfitting to training data is demonstrated to occur as the network size is increased. In an extreme case, a U-net (which produces better predictions for training data) is shown to completely fail on test data due to overfitting to this case of very limited training data. Meanwhile, the resultant images from the proposed CNN (which has low training data requirements) have lower noise, reduced ringing, and partial volume effects, as well as sharper edges and improved resolution compared to conventional MLEM.