FBP-Net for direct reconstruction of dynamic PET images

Noninvasive Quantification of Glucose Metabolism in Mice Myocardium Using the Spline Reconstruction Technique

The spline reconstruction technique (SRT) is a fast algorithm based on a novel numerical implementation of an analytic representation of the inverse Radon transform. The purpose of this study was to compare the SRT, filtered back-projection (FBP), and the Tera-Tomo 3D algorithm for various iteration numbers, using small-animal dynamic PET data obtained from a Mediso nanoScan® PET/CT scanner. For this purpose, Patlak graphical kinetic analysis was employed to noninvasively quantify the myocardial metabolic rate of glucose (MRGlu) in seven male C57BL/6 mice (n=7). All analytic reconstructions were performed via software for tomographic image reconstruction. The analysis of all PET-reconstructed images was conducted with PMOD software (version 3.506, PMOD Technologies LLC, Fällanden, Switzerland) using the inferior vena cava as the image-derived input function. Statistical significance was determined by employing the one-way analysis of variance test. The results revealed that the differences between the values of MRGlu obtained via SRT versus FBP, and the variants of he Tera-Tomo 3D algorithm were not statistically significant (p > 0.05). Overall, the SRT appears to perform similarly to the other algorithms investigated, providing a valid alternative analytic method for preclinical dynamic PET studies.

Current status and future directions in artificial intelligence for nuclear cardiology

INTRODUCTION

Myocardial perfusion imaging (MPI) is one of the most commonly ordered cardiac imaging tests. Accurate motion correction, image registration, and reconstruction are critical for high-quality imaging, but this can be technically challenging and has traditionally relied on expert manual processing. With accurate processing, there is a rich variety of clinical, stress, functional, and anatomic data that can be integrated to guide patient management.

AREAS COVERED

PubMed and Google Scholar were reviewed for articles related to artificial intelligence in nuclear cardiology published between 2020 and 2024. We will outline the prominent roles for artificial intelligence (AI) solutions to provide motion correction, image registration, and reconstruction. We will review the role for AI in extracting anatomic data for hybrid MPI which is otherwise neglected. Lastly, we will discuss AI methods to integrate the wealth of data to improve disease diagnosis or risk stratification.

EXPERT OPINION

There is growing evidence that AI will transform the performance of MPI by automating and improving on aspects of image acquisition and reconstruction. Physicians and researchers will need to understand the potential strengths of AI in order to benefit from the full clinical utility of MPI.

Signal separation of simultaneous dual-tracer PET imaging based on global spatial information and channel attention

BACKGROUND

Simultaneous dual-tracer positron emission tomography (PET) imaging efficiently provides more complete information for disease diagnosis. The signal separation has long been a challenge of dual-tracer PET imaging. To predict the single-tracer images, we proposed a separation network based on global spatial information and channel attention, and connected it to FBP-Net to form the FBPnet-Sep model.

RESULTS

Experiments using simulated dynamic PET data were conducted to: (1) compare the proposed FBPnet-Sep model to Sep-FBPnet model and currently existing Multi-task CNN, (2) verify the effectiveness of modules incorporated in FBPnet-Sep model, (3) investigate the generalization of FBPnet-Sep model to low-dose data, and (4) investigate the application of FBPnet-Sep model to multiple tracer combinations with decay corrections. Compared to the Sep-FBPnet model and Multi-task CNN, the FBPnet-Sep model reconstructed single-tracer images with higher structural similarity, peak signal-to-noise ratio and lower mean squared error, and reconstructed time-activity curves with lower bias and variation in most regions. Excluding the Inception or channel attention module resulted in degraded image qualities. The FBPnet-Sep model showed acceptable performance when applied to low-dose data. Additionally, it could deal with multiple tracer combinations. The qualities of predicted images, as well as the accuracy of derived time-activity curves and macro-parameters were slightly improved by incorporating a decay correction module.

CONCLUSIONS

The proposed FBPnet-Sep model was considered a potential method for the reconstruction and signal separation of simultaneous dual-tracer PET imaging.

Artificial Intelligence in Nuclear Cardiology: An Update and Future Trends

Myocardial perfusion imaging (MPI), using either single photon emission computed tomography (SPECT) or positron emission tomography (PET), is one of the most commonly ordered cardiac imaging tests, with prominent clinical roles for disease diagnosis and risk prediction. Artificial intelligence (AI) could potentially play a role in many steps along the typical MPI workflow, from image acquisition through to clinical reporting and risk estimation. AI can be utilized to improve image quality, reducing radiation exposure and image acquisition times. Once images are acquired, AI can help optimize motion correction and image registration during image reconstruction or provide direct image attenuation correction. Utilizing these image sets, AI can segment a number of anatomic features from associated computed tomographic imaging or even generate synthetic attenuation imaging. Lastly, AI may play an important role in disease diagnosis or risk prediction by combining the large number of potentially important clinical, stress, and imaging-related variables. This review will focus on the most recent developments in the field, providing clinicians and researchers with a timely update on the field. Additionally, it will discuss future trends including applications of AI during multiple points of the typical MPI workflow to maximize clinical utility and methods to maximize the information that can be obtained from hybrid imaging.

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.

Artificial Intelligence in Nuclear Cardiology

Artificial intelligence (AI) encompasses a variety of computer algorithms that have a wide range of potential clinical applications in nuclear cardiology. This article will introduce core terminology and concepts for AI including classifications of AI as well as training and testing regimens. We will then highlight the potential role for AI to improve image registration and image quality. Next, we will discuss methods for AI-driven image attenuation correction. Finally, we will review advancements in machine learning and deep-learning applications for disease diagnosis and risk stratification, including efforts to improve clinical translation of this valuable technology with explainable AI models.

Unsupervised PET logan parametric image estimation using conditional deep image prior

Recently, deep learning-based denoising methods have been gradually used for PET images denoising and have shown great achievements. Among these methods, one interesting framework is conditional deep image prior (CDIP) which is an unsupervised method that does not need prior training or a large number of training pairs. In this work, we combined CDIP with Logan parametric image estimation to generate high-quality parametric images. In our method, the kinetic model is the Logan reference tissue model that can avoid arterial sampling. The neural network was utilized to represent the images of Logan slope and intercept. The patient's computed tomography (CT) image or magnetic resonance (MR) image was used as the network input to provide anatomical information. The optimization function was constructed and solved by the alternating direction method of multipliers (ADMM) algorithm. Both simulation and clinical patient datasets demonstrated that the proposed method could generate parametric images with more detailed structures. Quantification results showed that the proposed method results had higher contrast-to-noise (CNR) improvement ratios (PET/CT datasets: 62.25%±29.93%; striatum of brain PET datasets : 129.51%±32.13%, thalamus of brain PET datasets: 128.24%±31.18%) than Gaussian filtered results (PET/CT datasets: 23.33%±18.63%; striatum of brain PET datasets: 74.71%±8.71%, thalamus of brain PET datasets: 73.02%±9.34%) and nonlocal mean (NLM) denoised results (PET/CT datasets: 37.55%±26.56%; striatum of brain PET datasets: 100.89%±16.13%, thalamus of brain PET datasets: 103.59%±16.37%).

Computed Tomography Imaging under Artificial Intelligence Reconstruction Algorithm Used in Recovery of Sports Injury of the Knee Anterior Cruciate Ligament

This study aimed to analyze the influence of artificial intelligence (AI) reconstruction algorithm on computed tomography (CT) images and the application of CT image analysis in the recovery of knee anterior cruciate ligament (ACL) sports injuries. A total of 90 patients with knee trauma were selected for enhanced CT scanning and randomly divided into three groups. Group A used the filtered back projection (FBP) reconstruction algorithm, and the tube voltage was set to 120 kV during CT scanning. Group B used the iDose4 reconstruction algorithm, and the tube voltage was set to 120 kV during CT scanning. In group C, the iDose4 reconstruction algorithm was used, and the tube voltage was set to 100 kV during CT scanning. The noise, signal-to-noise ratio (SNR), carrier-to-noise ratio (CNR), CT dose index volume (CTDI), dose length product (DLP), and effective radiation dose (ED) of the three groups of CT images were compared. The results showed that the noise of groups B and C was smaller than that of group A (P < 0.05), and the SNR and CNR of groups B and C were higher than those of group A. The images of patients in group A with the FBP reconstruction algorithm were noisy, and the boundaries were not clear. The noise of the images obtained by the iDose4 reconstruction algorithm in groups B and C was improved, and the image resolution was also higher. The agreement between arthroscopy and CT scan results was 96%. Therefore, the iterative reconstruction algorithm of iDose4 can improve the image quality. It was of important value in the diagnosis of knee ACL sports injury.

Deep learning-based image reconstruction and post-processing methods in positron emission tomography for low-dose imaging and resolution enhancement

Image processing plays a crucial role in maximising diagnostic quality of positron emission tomography (PET) images. Recently, deep learning methods developed across many fields have shown tremendous potential when applied to medical image enhancement, resulting in a rich and rapidly advancing literature surrounding this subject. This review encapsulates methods for integrating deep learning into PET image reconstruction and post-processing for low-dose imaging and resolution enhancement. A brief introduction to conventional image processing techniques in PET is firstly presented. We then review methods which integrate deep learning into the image reconstruction framework as either deep learning-based regularisation or as a fully data-driven mapping from measured signal to images. Deep learning-based post-processing methods for low-dose imaging, temporal resolution enhancement and spatial resolution enhancement are also reviewed. Finally, the challenges associated with applying deep learning to enhance PET images in the clinical setting are discussed and future research directions to address these challenges are presented.

A back‐projection‐and‐filtering‐like (BPF‐like) reconstruction method with the deep learning filtration from listmode data in TOF‐PET

PURPOSE

The time-of-flight (TOF) information improves signal-to-noise ratio (SNR) for positron emission tomography (PET) imaging. Existing analytical algorithms for TOF PET usually follow a filtered back-projection process on reconstructing images from the sinogram data. This work aims to develop a back-projection-and-filtering-like (BPF-like) algorithm that reconstructs the TOF PET image directly from listmode data rapidly.

METHODS

We extended the 2D conventional non-TOF PET projection model to a TOF case, where projection data are represented as line integrals weighted by the one-dimensional TOF kernel along the projection direction. After deriving the central slice theorem and the TOF back-projection of listmode data, we designed a deep learning network with a modified U-net architecture to perform the spatial filtration (reconstruction filter). The proposed BP-Net method was validated via Monte Carlo simulations of TOF PET listmode data with three different time resolutions for two types of activity phantoms. The network was only trained on the simulated full-dose XCAT dataset and then evaluated on XCAT and Jaszczak data with different time resolutions and dose levels.

RESULTS

Reconstructed images show that when compared with the conventional BPF algorithm and the MLEM algorithm proposed for TOF PET, the proposed BP-Net method obtains better image quality in terms of peak signal-to-noise ratio, relative mean square error, and structure similarity index; besides, the reconstruction speed of the BP-Net is 1.75 times faster than BPF and 29.05 times faster than MLEM using 15 iterations. The results also indicate that the performance of the BP-Net degrades with worse time resolutions and lower tracer doses, but degrades less than BPF or MLEM reconstructions.

CONCLUSION

In this work, we developed an analytical-like reconstruction in the form of BPF with the reconstruction filtering operation performed via a deep network. The method runs even faster than the conventional BPF algorithm and provides accurate reconstructions from listmode data in TOF-PET, free of rebinning data to a sinogram.

A review on AI in PET imaging

Artificial intelligence (AI) has been applied to various medical imaging tasks, such as computer-aided diagnosis. Specifically, deep learning techniques such as convolutional neural network (CNN) and generative adversarial network (GAN) have been extensively used for medical image generation. Image generation with deep learning has been investigated in studies using positron emission tomography (PET). This article reviews studies that applied deep learning techniques for image generation on PET. We categorized the studies for PET image generation with deep learning into three themes as follows: (1) recovering full PET data from noisy data by denoising with deep learning, (2) PET image reconstruction and attenuation correction with deep learning and (3) PET image translation and synthesis with deep learning. We introduce recent studies based on these three categories. Finally, we mention the limitations of applying deep learning techniques to PET image generation and future prospects for PET image generation.

Fast γ Photon Imaging for Inner Surface Defects Detecting

Only a few effective methods can detect internal defects and monitor the internal state of complex structural parts. On the basis of the principle of PET (positron emission computed tomography), a new measurement method, using γ photon to detect defects of an inner surface, is proposed. This method has the characteristics of strong penetration, anti-corrosion and anti-interference. With the aim of improving detection accuracy and imaging speed, this study also proposes image reconstruction algorithms, combining the classic FBP (filtered back projection) with MLEM (maximum likelihood expectation Maximization) algorithm. The proposed scheme can reduce the number of iterations required, when imaging, to achieve the same image quality. According to the operational demands of FPGAs (field-programmable gate array), a BPML (back projection maximum likelihood) algorithm is adapted to the structural characteristics of an FPGA, which makes it feasible to test the proposed algorithms therein. Furthermore, edge detection and defect recognition are conducted after reconstructing the inner image. The effectiveness and superiority of the algorithm are verified, and the performance of the FPGA is evaluated by the experiments.

Artificial Intelligence and Cardiac PET/Computed Tomography Imaging

Artificial intelligence is an important technology, with rapidly expanding applications for cardiac PET. We review the common terminology, including methods for training and testing, which are fundamental to understanding artificial intelligence. Next, we highlight applications to improve image acquisition, reconstruction, and segmentation. Computed tomographic imaging is commonly acquired in conjunction with PET and various artificial intelligence methods have been applied, including methods to automatically extract anatomic information or generate synthetic attenuation images. Last, we describe methods to automate disease diagnosis or risk stratification. This summary highlights the current and future clinical applications of artificial intelligence to cardiovascular PET imaging.

CT Image Features of the FBP Reconstruction Algorithm in the Evaluation of Fasting Blood Sugar Level of Diabetic Pulmonary Tuberculosis Patients and Early Diet Nursing

The study was aimed at exploring the application value of the CT image based on a filtered back projection (FBP) algorithm in the diagnosis of patients with diabetes complicated with tuberculosis and at analyzing the influence of dietary nursing on patients with diabetes complicated with tuberculosis. In this study, the FBP algorithm was used to optimize CT images to effectively obtain reconstructed ROI images. Then, the deviation from measurement values of reconstructed images at different pixel levels was analyzed. 138 patients with diabetes complicated with tuberculosis were selected as research subjects to compare the number of lung segments involved and the CT imaging manifestations at different fasting glucose levels. All patients were divided into the control group (routine drug treatment) and observation group (diet intervention on the basis of drug treatment) by random number table method, and the effect of different nursing methods on the improvement of patients' clinical symptoms was discussed. The results showed that the distance measurement value decreased with the increase in pixel level, there was no significant difference in the number of lung segments involved in patients with different fasting glucose levels (P > 0.05), and there were statistically significant differences in the incidence of segmental lobar shadow, bronchial air sign, wall-less cavity, thick-walled cavity, pulmonary multiple cavity, and bronchial tuberculosis in patients with different fasting glucose levels (P < 0.05). Compared with the control group, 2 h postprandial blood glucose level in the observation group was significantly improved (P < 0.05), there was a statistical significance in the number with reduced pleural effusion and the number with reduced tuberculosis foci in the two groups (P < 0.05), and the level of hemoglobin in the observation group was 7.1 ± 1.26, significantly lower than that in the control group (8.91 ± 2.03, P < 0.05). It suggested that the changes of CT images based on the FBP reconstruction algorithm were correlated with fasting blood glucose level. Personalized diet nursing intervention can improve the clinical symptoms of patients, which provides a reference for the clinical diagnosis and treatment of patients with diabetes complicated with tuberculosis.

Computational approaches to detect small lesions in 18 F-FDG PET/CT scans

Purpose

When physicians interpret ¹⁸F‐FDG PET/CT scans, they rely on their subjective visual impression of the presence of small lesions, the criteria for which may vary among readers. Our investigation used physical phantom scans to evaluate whether image texture analysis metrics reliably correspond to visual criteria used to identify lesions and accurately differentiate background regions from sub‐centimeter simulated lesions.

Methods

Routinely collected quality assurance test data were processed retrospectively for 65 different ¹⁸F‐FDG PET scans performed of standardized phantoms on eight different PET/CT systems. Phantoms included 8‐, 12‐, 16‐, and 25‐mm diameter cylinders embedded in a cylindrical water bath, prepared with 2.5:1 activity‐to‐background ratio emulating typical whole‐body PET protocols. Voxel values in cylinder regions and background regions were sampled to compute several classes of image metrics. Two experienced physicists, blinded to quantified image metrics and to each other's readings, independently graded cylinder visibility on a 5‐level scale (0 = definitely not visible to 4 = definitely visible).

Results

The three largest cylinders were visible in 100% of cases with a mean visibility score of 3.3 ± 1.2, while the smallest 8‐mm cylinder was visible in 58% of cases with a significantly lower mean visibility score of 1.5±1.1 (P < 0.0001). By ROC analysis, the polynomial‐fit signal‐to‐noise ratio was the most accurate at discriminating 8‐mm cylinders from the background, with accuracy greater than visual detection (93% ± 2% versus 76% ± 4%, P = 0.0001), and better sensitivity (94% versus 58%, P < 0.0001).

Conclusion

Image texture analysis metrics are more sensitive than visual impressions for detecting sub‐centimeter simulated lesions. Therefore, image texture analysis metrics are potentially clinically useful for ¹⁸F‐FDG PET/CT studies.

GapFill-Recon Net: A Cascade Network for simultaneously PET Gap Filling and Image Reconstruction

PET image reconstruction from incomplete data, such as the gap between adjacent detector blocks generally introduces partial projection data loss, is an important and challenging problem in medical imaging. This work proposes an efficient convolutional neural network (CNN) framework, called GapFill-Recon Net, that jointly reconstructs PET images and their associated sinogram data. GapFill-Recon Net including two blocks: the Gap-Filling block first address the sinogram gap and the Image-Recon block maps the filled sinogram onto the final image directly. A total of 43,660 pairs of synthetic 2D PET sinograms with gaps and images generated from the MOBY phantom are utilized for network training, testing and validation. Whole-body mouse Monte Carlo (MC) simulated data are also used for evaluation. The experimental results show that the reconstructed image quality of GapFill-Recon Net outperforms filtered back-projection (FBP) and maximum likelihood expectation maximization (MLEM) in terms of the structural similarity index metric (SSIM), relative root mean squared error (rRMSE), and peak signal-to-noise ratio (PSNR). Moreover, the reconstruction speed is equivalent to that of FBP and was nearly 83 times faster than that of MLEM. In conclusion, compared with the traditional reconstruction algorithm, GapFill-Recon Net achieves relatively optimal performance in image quality and reconstruction speed, which effectively achieves a balance between efficiency and performance.

Prediction of an oxygen extraction fraction map by convolutional neural network: validation of input data among MR and PET images

Purpose

Oxygen extraction fraction (OEF) is a biomarker for the viability of brain tissue in ischemic stroke. However, acquisition of the OEF map using positron emission tomography (PET) with oxygen-15 gas is uncomfortable for patients because of the long fixation time, invasive arterial sampling, and radiation exposure. We aimed to predict the OEF map from magnetic resonance (MR) and PET images using a deep convolutional neural network (CNN) and to demonstrate which PET and MR images are optimal as inputs for the prediction of OEF maps.

Methods

Cerebral blood flow at rest (CBF) and during stress (sCBF), cerebral blood volume (CBV) maps acquired from oxygen-15 PET, and routine MR images (T1-, T2-, and T2*-weighted images) for 113 patients with steno-occlusive disease were learned with U-Net. MR and PET images acquired from the other 25 patients were used as test data. We compared the predicted OEF maps and intraclass correlation (ICC) with the real OEF values among combinations of MRI, CBF, CBV, and sCBF.

Results

Among the combinations of input images, OEF maps predicted by the model learned with MRI, CBF, CBV, and sCBF maps were the most similar to the real OEF maps (ICC: 0.597 ± 0.082). However, the contrast of predicted OEF maps was lower than that of real OEF maps.

Conclusion

These results suggest that the deep CNN learned useful features from CBF, sCBF, CBV, and MR images and predict qualitatively realistic OEF maps. These findings suggest that the deep CNN model can shorten the fixation time for ¹⁵O PET by skipping ¹⁵O₂ scans. Further training with a larger data set is required to predict accurate OEF maps quantitatively.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11548-021-02356-7.