Penalized maximum-likelihood reconstruction for improving limited-angle artifacts in a dedicated head and neck PET system

Abstract

Positron emission tomography (PET) suffers from limited spatial resolution in current head and neck cancer management. We are building a dual-panel high-resolution PET system to aid the detection of tumor involvement in small lymph nodes ( < 10 mm in diameter). The system is based on cadmium zinc telluride (CZT) detectors with cross-strip electrode readout (1 mm anode pitch and 5 mm cathode pitch). One challenge of the dual-panel system is that the limited angular coverage of the imaging volume leads to artifacts in reconstructed images, such as the elongation of lesions. In this work, we leverage a penalized maximum-likelihood (PML) reconstruction for the limited-angle PET system. The dissimilarity between the image to be reconstructed and a prior image from a low-resolution whole-body scanner is penalized. An image-based resolution model is incorporated into the regularization. Computer simulations were used to evaluate the performance of the method. Results demonstrate that the elongation of the 6-mm and 8-mm diameter hot spheres is eliminated with the regularization strength γ being 0.02 or larger. The PML reconstruction yields higher contrast recovery coefficient (CRC) of hot spheres compared to the maximum-likelihood reconstruction, as well as the low-resolution whole-body image, across all hot sphere sizes tested (3, 4, 6, and 8 mm). The method studied in this work provides a way to mitigate the limited-angle artifacts in the reconstruction from limited-angle PET data, making the high-resolution dual-panel dedicated head and neck PET system promising for head and neck cancer management.

Hybrid model of photon propagation based on the analytical and Monte Carlo methods for a dual-head PET system

“…One way is to rotate the system or engage the area detector to enlarge the angular coverage (Raylman et al 2008, Zhang et al 2015. The other way is to incorporate the point spread function (PSF) (Lee et al 2014) or prior image in reconstruction (Zhang et al 2020) to mitigate artifacts caused by the limited angle. In this study, we only use the MLEM algorithm in the reconstruction of the dual-head PET system.…”

Exploring Longitudinal MRI-Based Deep Learning Analysis in Parkinson’s Patients - A Short Survey Focus on Handedness

“…Consistency regularization, also known as consistency-aware training, is a key concept and it aims to ensure the model maintains similar predictions under various data and model perturbations [94]. The data that is not annotated can be potentially extended with a pseudo label according to the prior knowledge [25,95]. Another common strategy of training with limited data is adversarial training [96].…”

“…Our review aims to bridge this gap by offering an exploration of current imaging analysis techniques [20][21][22][23][24][25] used in PD research. We specifically emphasize a modeling perspective, encompassing deep learning (DL), one subfield of artificial intelligence (AI) techniques, made feasible by the current advancements in computing power and more availability of large datasets.…”

Deep learning-based image registration method: with application to Scanning Laser Ophthalmoscopy (SLO) longitudinal images

“…Deep learning based methods have recently shown promise for challenging registration problems and thus could be suitable for addressing the challenges in retinal image registration. [2][3][4] Lee et al, 5 describe a relevant method to learn feature classes from patches extracted from vascular structures in retinal images. The presented method relies upon basic image processing methods involving several empirically defined parameters for vascular structure identification and affine transformation estimation.…”