Deep-Learning-Based Few-Angle Cardiac SPECT Reconstruction Using Transformer

AI in SPECT Imaging: Opportunities and Challenges

SPECT is a widely used imaging modality in nuclear medicine which provides essential functional insights into cardiovascular, neurological, and oncological diseases. However, SPECT imaging suffers from limited quantitative accuracy due to low spatial resolution and high noise levels, posing significant challenges for precise diagnosis, disease monitoring, and treatment planning. Recent advances in artificial intelligence (AI), in particular deep learning-based techniques such as convolutional neural networks (CNNs), generative adversarial networks (GANs), and transformers, have led to substantial improvements in SPECT image reconstruction, enhancement, attenuation correction, segmentation, disease classification, and multimodal fusion. These AI approaches have enabled more accurate extraction of functional and anatomical information, improved quantitative analysis, and facilitated the integration of SPECT with other imaging modalities to enhance clinical decision-making. This review provides a comprehensive overview of AI-driven developments in SPECT imaging, highlighting progress in both supervised and unsupervised learning approaches, innovations in image synthesis and cross-modality learning, and the potential of self-supervised and contrastive learning strategies for improving model robustness. Additionally, we discuss key challenges, including data heterogeneity, model interpretability, and computational complexity, which continue to limit the clinical adoption of AI methods. The need for standardized evaluation metrics, large-scale multimodal datasets, and clinically validated AI models remains a crucial factor in ensuring the reliability and generalizability of AI approaches. Future research directions include the exploration of foundation models and large language models for knowledge-driven image analysis, as well as the development of more adaptive and personalized AI frameworks tailored for nuclear imaging applications.

Investigation of a deep learning-based reconstruction approach utilizing dual-view projection for myocardial perfusion SPECT imaging

Single-photon emission computed tomography (SPECT) is widely used in myocardial perfusion imaging (MPI) in clinic. However, conventional dual-head SPECT scanners require lengthy scanning times and gantry rotation, which limits the application of SPECT MPI. In this work, we proposed a deep learning-based approach to reconstruct dual-view projections, aiming to reduce acquisition time and enable non-rotational imaging for MPI based on conventional dual-head SPECT scanners. U-Net was adopted for the dual-view projection reconstruction. Initially, 2D U-Nets were used to evaluate various data organization schemes for dual-view projection as input, including paved projection, interleaved projection, and stacked projection, with and without an attenuation map. Subsequently, we developed 3D U-Nets using the optimal data organization scheme as input to further enhance reconstruction performance. The dataset consisted of a total of 116 SPECT/CT scans with 99mTc-tetrofosmin tracer acquired on a GE NM/CT 640 scanner. Reconstruction performance was assessed using quantitative metrices and absolute percentage errors, while the reconstruction images from the full-view projection were used as reference images. The 2D U-Nets provided reasonable transverse view images but exhibited slight axial discontinuity compared to the reference images, regardless of the data organization schemes. Incorporating the attenuation map reduced this axial discontinuity. Quantitatively, the 2D U-Net trained using both stacked projection and attenuation map achieved the best performance, with a normalized mean absolute error of 0.6%±0.3% and a structural similarity index measure (SSIM) of 0.93±0.04. The 3D U-Net further improved the performance with less axial discontinuity and a higher SSIM of 0.94±0.03. The localized absolute percentage errors were 1.8±16.8% and -2.0±6.3% in the left ventricular (LV) cavity and myocardium, respectively. We developed a deep learning-based image reconstruction approach for dual-view projection from a conventional SPECT scanner. The 3D U-Net, trained with the stacked projection with an attenuation map is effective for non-rotational imaging and could benefit dynamic myocardium perfusion imaging.

DEMIST: A Deep-Learning-Based Detection-Task-Specific Denoising Approach for Myocardial Perfusion SPECT

There is an important need for methods to process myocardial perfusion imaging (MPI) single-photon emission computed tomography (SPECT) images acquired at lower-radiation dose and/or acquisition time such that the processed images improve observer performance on the clinical task of detecting perfusion defects compared to low-dose images. To address this need, we build upon concepts from model-observer theory and our understanding of the human visual system to propose a detection task-specific deep-learning-based approach for denoising MPI SPECT images (DEMIST). The approach, while performing denoising, is designed to preserve features that influence observer performance on detection tasks. We objectively evaluated DEMIST on the task of detecting perfusion defects using a retrospective study with anonymized clinical data in patients who underwent MPI studies across two scanners (N = 338). The evaluation was performed at low-dose levels of 6.25%, 12.5%, and 25% and using an anthropomorphic channelized Hotelling observer. Performance was quantified using area under the receiver operating characteristics curve (AUC). Images denoised with DEMIST yielded significantly higher AUC compared to corresponding low-dose images and images denoised with a commonly used task-agnostic deep learning-based denoising method. Similar results were observed with stratified analysis based on patient sex and defect type. Additionally, DEMIST improved visual fidelity of the low-dose images as quantified using root mean squared error and structural similarity index metric. A mathematical analysis revealed that DEMIST preserved features that assist in detection tasks while improving the noise properties, resulting in improved observer performance. The results provide strong evidence for further clinical evaluation of DEMIST to denoise low-count images in MPI SPECT.

Deep learning-based PET image denoising and reconstruction: a review

This review focuses on positron emission tomography (PET) imaging algorithms and traces the evolution of PET image reconstruction methods. First, we provide an overview of conventional PET image reconstruction methods from filtered backprojection through to recent iterative PET image reconstruction algorithms, and then review deep learning methods for PET data up to the latest innovations within three main categories. The first category involves post-processing methods for PET image denoising. The second category comprises direct image reconstruction methods that learn mappings from sinograms to the reconstructed images in an end-to-end manner. The third category comprises iterative reconstruction methods that combine conventional iterative image reconstruction with neural-network enhancement. We discuss future perspectives on PET imaging and deep learning technology.

Semi-supervised low-dose SPECT restoration using sinogram inner-structure aware graph neural network

Objective.To mitigate the potential radiation risk, low-dose single photon emission computed tomography (SPECT) is of increasing interest. Numerous deep learning-based methods have been developed to perform low-dose imaging while maintaining image quality. However, most existing methods seldom explore the unique inner-structure inherent within sinograms. In addition, traditional supervised learning methods require large-scale labeled data, where the normal-dose data serves as annotation and is intractable to acquire in low-dose imaging. In this study, we aim to develop a novel sinogram inner-structure-aware semi-supervised framework for the task of low-dose SPECT sinogram restoration.Approach.The proposed framework retains the strengths of UNet, meanwhile introducing a sinogram-structure-based non-local neighbors graph neural network (SSN-GNN) module and a window-based K-nearest neighbors GNN (W-KNN-GNN) module to effectively exploit the inherent inner-structure within SPECT sinograms. Moreover, the proposed framework employs the mean teacher semi-supervised learning approach to leverage the information available in abundant unlabeled low-dose sinograms.Main results.The datasets exploited in this study were acquired from the (Extended Cardiac-Torso) XCAT anthropomorphic digital phantoms, which provide realistic images for imaging research of various modalities. Quantitative as well as qualitative results demonstrate that the proposed framework achieves superior performance compared to several state-of-the-art reconstruction methods. To further validate the effectiveness of the proposed framework, ablation and robustness experiments were also performed. The experimental results show that each component of the proposed framework effectively improves the model performance, and the framework exhibits superior robustness with respect to various noise levels. Besides, the proposed semi-supervised paradigm showcases the efficacy of incorporating supplementary unlabeled low-dose sinograms.Significance.The proposed framework improves the quality of low-dose SPECT reconstructed images by utilizing sinogram inner-structure and incorporating supplementary unlabeled data, which provides an important tool for dose reduction without sacrificing the image quality.