Direct Parametric Image Reconstruction in Reduced Parameter Space for Rapid Multi-Tracer PET Imaging

Kinetic model-informed deep learning for multiplexed PET image separation

BACKGROUND

Multiplexed positron emission tomography (mPET) imaging can measure physiological and pathological information from different tracers simultaneously in a single scan. Separation of the multiplexed PET signals within a single PET scan is challenging due to the fact that each tracer gives rise to indistinguishable 511 keV photon pairs, and thus no unique energy information for differentiating the source of each photon pair.

METHODS

Recently, many applications of deep learning for mPET image separation have been concentrated on pure data-driven methods, e.g., training a neural network to separate mPET images into single-tracer dynamic/static images. These methods use over-parameterized networks with only a very weak inductive prior. In this work, we improve the inductive prior of the deep network by incorporating a general kinetic model based on spectral analysis. The model is incorporated, along with deep networks, into an unrolled image-space version of an iterative fully 4D PET reconstruction algorithm.

RESULTS

The performance of the proposed method was evaluated on a simulated brain image dataset for dual-tracer [18 F]FDG+[11 C]MET PET image separation. The results demonstrate that the proposed method can achieve separation performance comparable to that obtained with single-tracer imaging. In addition, the proposed method outperformed the model-based separation methods (the conventional voxel-wise multi-tracer compartment modeling method (v-MTCM) and the image-space dual-tracer version of the fully 4D PET image reconstruction algorithm (IS-F4D)), as well as a pure data-driven separation [using a convolutional encoder-decoder (CED)], with fewer training examples.

CONCLUSIONS

This work proposes a kinetic model-informed unrolled deep learning method for mPET image separation. In simulation studies, the method proved able to outperform both the conventional v-MTCM method and a pure data-driven CED with less training data.

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.

Deep learned triple-tracer multiplexed PET myocardial image separation

Introduction

In multiplexed positron emission tomography (mPET) imaging, physiological and pathological information from different radiotracers can be observed simultaneously in a single dynamic PET scan. The separation of mPET signals within a single PET scan is challenging due to the fact that the PET scanner measures the sum of the PET signals of all the tracers. The conventional multi-tracer compartment modeling (MTCM) method requires staggered injections and assumes that the arterial input functions (AIFs) of each tracer are known.

Methods

In this work, we propose a deep learning-based method to separate triple-tracer PET images without explicitly knowing the AIFs. A dynamic triple-tracer noisy MLEM reconstruction was used as the network input, and dynamic single-tracer noisy MLEM reconstructions were used as training labels.

Results

A simulation study was performed to evaluate the performance of the proposed framework on triple-tracer ([ F 18 ]FDG+ Rb 82 +[ Tc 99m ]sestamibi) PET myocardial imaging. The results show that the proposed methodology substantially reduced the noise level compared to the results obtained from single-tracer imaging. Additionally, it achieved lower bias and standard deviation in the separated single-tracer images compared to the MTCM-based method at both the voxel and region of interest (ROI) levels.

Discussion

As compared to MTCM separation, the proposed method uses spatiotemporal information for separation, which improves the separation performance at both the voxel and ROI levels. The simulation study also demonstrates the feasibility and potential of the proposed DL-based method for the application to pre-clinical and clinical studies.

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.

Machine learning methods for tracer kinetic modelling

Tracer kinetic modelling based on dynamic PET is an important field of Nuclear Medicine for quantitative functional imaging. Yet, its implementation in clinical routine has been constrained by its complexity and computational costs. Machine learning poses an opportunity to improve modelling processes in terms of arterial input function prediction, the prediction of kinetic modelling parameters and model selection in both clinical and preclinical studies while reducing processing time. Moreover, it can help improving kinetic modelling data used in downstream tasks such as tumor detection. In this review, we introduce the basics of tracer kinetic modelling and present a literature review of original works and conference papers using machine learning methods in this field.

Synthesis of Prospective Multiple Time Points F-18 FDG PET Images from a Single Scan Using a Supervised Generative Adversarial Network

The cumulative activity map estimation are essential tools for patient specific dosimetry with high accuracy, which is estimated using biokinetic models instead of patient dynamic data or the number of static PET scans, owing to economical and time-consuming points of view. In the era of deep learning applications in medicine, the pix-to-pix (p2 p) GAN neural networks play a significant role in image translation between imaging modalities. In this pilot study, we extended the p2 p GAN networks to generate PET images of patients at different times according to a 60 min scan time after the injection of F-18 FDG. In this regard, the study was conducted in two sections: phantom and patient studies. In the phantom study section, the SSIM, PSNR, and MSE metric results of the generated images varied from 0.98-0.99, 31-34 and 1-2 respectively and the fine-tuned Resnet-50 network classified the different timing images with high performance. In the patient study, these values varied from 0.88-0.93, 36-41 and 1.7-2.2, respectively and the classification network classified the generated images in the true group with high accuracy. The results of phantom studies showed high values of evaluation metrics owing to ideal image quality conditions. However, in the patient study, promising results were achieved which showed that the image quality and training data number affected the network performance. This study aims to assess the feasibility of p2 p GAN network application for different timing image generation.

Direct reconstruction for simultaneous dual-tracer PET imaging based on multi-task learning

BACKGROUND

Simultaneous dual-tracer positron emission tomography (PET) imaging can observe two molecular targets in a single scan, which is conducive to disease diagnosis and tracking. Since the signals emitted by different tracers are the same, it is crucial to separate each single tracer from the mixed signals. The current study proposed a novel deep learning-based method to reconstruct single-tracer activity distributions from the dual-tracer sinogram.

METHODS

We proposed the Multi-task CNN, a three-dimensional convolutional neural network (CNN) based on a framework of multi-task learning. One common encoder extracted features from the dual-tracer dynamic sinogram, followed by two distinct and parallel decoders which reconstructed the single-tracer dynamic images of two tracers separately. The model was evaluated by mean squared error (MSE), multiscale structural similarity (MS-SSIM) index and peak signal-to-noise ratio (PSNR) on simulated data and real animal data, and compared to the filtered back-projection method based on deep learning (FBP-CNN).

RESULTS

In the simulation experiments, the Multi-task CNN reconstructed single-tracer images with lower MSE, higher MS-SSIM and PSNR than FBP-CNN, and was more robust to the changes in individual difference, tracer combination and scanning protocol. In the experiment of rats with an orthotopic xenograft glioma model, the Multi-task CNN reconstructions also showed higher qualities than FBP-CNN reconstructions.

CONCLUSIONS

The proposed Multi-task CNN could effectively reconstruct the dynamic activity images of two single tracers from the dual-tracer dynamic sinogram, which was potential in the direct reconstruction for real simultaneous dual-tracer PET imaging data in future.

Temporal information guided dynamic dual-tracer PET signal separation network

PURPOSE

The difficulty of dynamic dual-tracer positron emission tomography (PET) technology is to separate the complete single-tracer information from mixed dual-tracer. Traditional methods cannot separate single injection single-scan dynamic dual-tracer PET images. In this paper, we propose a deep learning framework based on gated recurrent unit (GRU) network and evaluate its performance with simulation experiments and realistic monkey data.

METHODS

The proposed single-scan dynamic dual-tracer PET image separation network consists of three parts, including encoder, separation and decoder module. Encoder part is to map the mixed time activity curves (TACs) from the low-dimensional space to the high-dimensional space to get mixed weight vector matrix. Separation part is to capture the temporal information of mixed weight vector matrix using bi-directional GRU (bi-GRU) layer to obtain the single-tracer masks, and the decoding part remaps the high-dimensional single-tracer weight vector matrix to the low-dimensional space to obtain two separated single tracers.

RESULTS

In the simulation experiments under different tracers, phantoms, noise levels, arterial input function (AIF) and k-parameter with Gaussian random, compared to the stacked auto encoder (SAE) network and traditional background subtraction method, GRU-based network has better performance with low bias and mean squared error (MSE). In the realistic study, the image results of GRU network have higher mean structural similarity (MSSIM), and peak signal to noise ratio (PSNR).

CONCLUSIONS

This study demonstrates the feasibility of temporal information guided neural network in single-injection single-scan dynamic dual-tracer PET images separation. The GRU-based network uses TAC temporal information without AIFs to make the separation results more robust and accurate, which significantly outperforms state-of-the-art method qualitatively and quantitatively. This article is protected by copyright. All rights reserved.

Mathematical Models for FDG Kinetics in Cancer: A Review

Compartmental analysis is the mathematical framework for the modelling of tracer kinetics in dynamical Positron Emission Tomography. This paper provides a review of how compartmental models are constructed and numerically optimized. Specific focus is given on the identifiability and sensitivity issues and on the impact of complex physiological conditions on the mathematical properties of the models.

Technical note: A preliminary study of dual-tracer PET image reconstruction guided by FDG and/or MR kernels

PURPOSE

Clinically, single radiotracer positron emission tomography (PET) imaging is a commonly used examination method; however, since each radioactive tracer reflects the information of only one kind of cell, it easily causes false negatives or false positives in disease diagnosis. Therefore, reasonably combining two or more radiotracers is recommended to improve the accuracy of diagnosis and the sensitivity and specificity of the disease when conditions permit.

METHODS

This paper proposes incorporating ¹⁸ F-fluorodeoxyglucose (FDG) as a higher-quality PET image to guide the reconstruction of other lower-count ¹¹ C-methionine (MET) PET datasets to compensate for the lower image quality by a popular kernel algorithm. Specifically, the FDG prior is needed to extract kernel features, and these features were used to build a kernel matrix using a k-nearest-neighbor (kNN) search for MET image reconstruction. We created a 2-D brain phantom to validate the proposed method by simulating sinogram data containing Poisson random noise and quantitatively compared the performance of the proposed FDG-guided kernelized expectation maximization (KEM) method with the performance of Gaussian and non-local means (NLM) smoothed maximum likelihood expectation maximization (MLEM), MR-guided KEM, and multi-guided-S KEM algorithms. Mismatch experiments between FDG/MR and MET data were also carried out to investigate the outcomes of possible clinical situations.

RESULTS

In the simulation study, the proposed method outperformed the other algorithms by at least 3.11% in the signal-to-noise ratio (SNR) and 0.68% in the contrast recovery coefficient (CRC), and it reduced the mean absolute error (MAE) by 8.07%. Regarding the tumor in the reconstructed image, the proposed method contained more pathological information. Furthermore, the proposed method was still superior to the MR-guided KEM method in the mismatch experiments.

CONCLUSIONS

The proposed FDG-guided KEM algorithm can effectively utilize and compensate for the tissue metabolism information obtained from dual-tracer PET to maximize the advantages of PET imaging.

Rapid high-quality PET Patlak parametric image generation based on direct reconstruction and temporal nonlocal neural network

Parametric imaging based on dynamic positron emission tomography (PET) has wide applications in neurology. Compared to indirect methods, direct reconstruction methods, which reconstruct parametric images directly from the raw PET data, have superior image quality due to better noise modeling and richer information extracted from the PET raw data. For low-dose scenarios, the advantages of direct methods are more obvious. However, the wide adoption of direct reconstruction is inevitably impeded by the excessive computational demand and deficiency of the accessible raw data. In addition, motion modeling inside dynamic PET image reconstruction raises more computational challenges for direct reconstruction methods. In this work, we focused on the ¹⁸F-FDG Patlak model, and proposed a data-driven approach which can estimate the motion corrected full-dose direct Patlak images from the dynamic PET reconstruction series, based on a proposed novel temporal non-local convolutional neural network. During network training, direct reconstruction with motion correction based on full-dose dynamic PET sinograms was performed to obtain the training labels. The reconstructed full-dose /low-dose dynamic PET images were supplied as the network input. In addition, a temporal non-local block based on the dynamic PET images was proposed to better recover the structural information and reduce the image noise. During testing, the proposed network can directly output high-quality Patlak parametric images from the full-dose /low-dose dynamic PET images in seconds. Experiments based on 15 full-dose and 15 low-dose ¹⁸F-FDG brain datasets were conducted and analyzed to validate the feasibility of the proposed framework. Results show that the proposed framework can generate better image quality than reference methods.

Positron range-free and multi-isotope tomography of positron emitters

Despite improvements in small animal PET instruments, many tracers cannot be imaged at sufficiently high resolutions due to positron range, while multi-tracer PET is hampered by the fact that all annihilation photons have equal energies. Here we realize multi-isotope and sub-mm resolution PET of isotopes with several mm positron range by utilizing prompt gamma photons that are commonly neglected. A PET-SPECT-CT scanner (VECTor/CT, MILabs, The Netherlands) equipped with a high-energy cluster-pinhole collimator was used to image ¹²⁴I and a mix of ¹²⁴I and ¹⁸F in phantoms and mice. In addition to positrons (mean range 3.4 mm) ¹²⁴I emits large amounts of 603 keV prompt gammas that-aided by excellent energy discrimination of NaI-were selected to reconstruct ¹²⁴I images that are unaffected by positron range. Photons detected in the 511 keV window were used to reconstruct ¹⁸F images. Images were reconstructed iteratively using an energy dependent matrix for each isotope. Correction of ¹⁸F images for contamination with ¹²⁴I annihilation photons was performed by Monte Carlo based range modelling and scaling of the ¹²⁴I prompt gamma image before subtracting it from the ¹⁸F image. Additionally, prompt gamma imaging was tested for ⁸⁹Zr that emits very high-energy prompts (909 keV). In Derenzo resolution phantoms 0.75 mm rods were clearly discernable for ¹²⁴I, ⁸⁹Zr and for simultaneously acquired ¹²⁴I and ¹⁸F imaging. Image quantification in phantoms with reservoirs filled with both ¹²⁴I and ¹⁸F showed excellent separation of isotopes and high quantitative accuracy. Mouse imaging showed uptake of ¹²⁴I in tiny thyroid parts and simultaneously injected ¹⁸F-NaF in bone structures. The ability to obtain PET images at sub-mm resolution both for isotopes with several mm positron range and for multi-isotope PET adds to many other unique capabilities of VECTor's clustered pinhole imaging, including simultaneous sub-mm PET-SPECT and theranostic high energy SPECT.

FBP-Net for direct reconstruction of dynamic PET images

Dynamic positron emission tomography (PET) imaging can provide information about metabolic changes over time, used for kinetic analysis and auxiliary diagnosis. Existing deep learning-based reconstruction methods have too many trainable parameters and poor generalization, and require mass data to train the neural network. However, obtaining large amounts of medical data is expensive and time-consuming. To reduce the need for data and improve the generalization of network, we combined the filtered back-projection (FBP) algorithm with neural network, and proposed FBP-Net which could directly reconstruct PET images from sinograms instead of post-processing the rough reconstruction images obtained by traditional methods. The FBP-Net contained two parts: the FBP part and the denoiser part. The FBP part adaptively learned the frequency filter to realize the transformation from the detector domain to the image domain, and normalized the coarse reconstruction images obtained. The denoiser part merged the information of all time frames to improve the quality of dynamic PET reconstruction images, especially the early time frames. The proposed FBP-Net was performed on simulation and real dataset, and the results were compared with the state-of-art U-net and DeepPET. The results showed that FBP-Net did not tend to overfit the training set and had a stronger generalization.

PET Parametric Imaging: Past, Present, and Future

Positron emission tomography (PET) is actively used in a diverse range of applications in oncology, cardiology, and neurology. The use of PET in the clinical setting focuses on static (single time frame) imaging at a specific time-point post radiotracer injection and is typically considered as semi-quantitative; e.g. standardized uptake value (SUV) measures. In contrast, dynamic PET imaging requires increased acquisition times but has the advantage that it measures the full spatiotemporal distribution of a radiotracer and, in combination with tracer kinetic modeling, enables the generation of multiparametric images that more directly quantify underlying biological parameters of interest, such as blood flow, glucose metabolism, and receptor binding. Parametric images have the potential for improved detection and for more accurate and earlier therapeutic response assessment. Parametric imaging with dynamic PET has witnessed extensive research in the past four decades. In this paper, we provide an overview of past and present activities and discuss emerging opportunities in the field of parametric imaging for the future.

Explicit measurement of multi-tracer arterial input function for PET imaging using blood sampling spectroscopy

Background

Conventional PET imaging has usually been limited to a single tracer per scan. We propose a new technique for multi-tracer PET imaging that uses dynamic imaging and multi-tracer compartment modeling including an explicitly derived arterial input function (AIF) for each tracer using blood sampling spectroscopy. For that purpose, at least one of the co-injected tracers must be based on a non-pure positron emitter.

Methods

The proposed technique was validated in vivo by performing cardiac PET/CT studies on three healthy pigs injected with ¹⁸FDG (viability) and ⁶⁸Ga-DOTA (myocardial blood flow and extracellular volume fraction) during the same acquisition. Blood samples were collected during the PET scan, and separated AIF for each tracer was obtained by spectroscopic analysis. A multi-tracer compartment model was applied to the myocardium in order to obtain the distribution of each tracer at the end of the PET scan. Relative activities of both tracers and tracer uptake were obtained and compared with the values obtained by ex vivo analysis of excised myocardial tissue segments.

Results

A high correlation was obtained between multi-tracer PET results, and those obtained from ex vivo analysis (¹⁸FDG relative activity: r = 0.95, p < 0.0001; SUV: r = 0.98, p < 0.0001).

Conclusions

The proposed technique allows performing PET scans with two tracers during the same acquisition obtaining separate information for each tracer.

Three-dimensional convolutional neural networks for simultaneous dual-tracer PET imaging

Dual-tracer positron emission tomography (PET) is a promising technique to measure the distribution of two tracers in the body by a single scan, which can improve the clinical accuracy of disease diagnosis and can also serve as a research tool for scientists. Most current research on dual-tracer PET reconstruction is based on mixed images pre-reconstructed by algorithms, which restricts the further improvement of the precision of reconstruction. In this study, we present a hybrid loss-guided deep learning based framework for dual-tracer PET imaging using sinogram data, which can achieve reconstruction by naturally unifying two functions: the reconstruction of the mixed images and the separation for individual tracers. Combined with volumetric dual-tracer images, we adopted a three-dimensional (3D) convolutional neural network (CNN) to learn full features, including spatial information and temporal information simultaneously. In addition, an auxiliary loss layer was introduced to guide the reconstruction of the dual tracers. We used Monte Carlo simulations with data augmentation to generate sufficient datasets for training and testing. The results were analyzed by the bias and variance both spatially (different regions of interest) and temporally (different frames). The analysis verified the feasibility of the 3D CNN framework for dual-tracer reconstruction. Furthermore, we compared the reconstruction results with a deep belief network (DBN), which is another deep learning based technique for the separation of dual-tracer images based on time-activity curves (TACs). The comparison results provide insights into the superior features and performance of the 3D CNN. Furthermore, we tested the [¹¹C]FMZ-[¹¹C]DTBZ images with three total-counts levels ([Formula: see text], [Formula: see text], [Formula: see text]), which indicate different noise ratios. The analysis results demonstrate that our method can successfully recover the respective distribution of lower total counts with nearly the same accuracy as that of the higher total counts in the total counts range we applied, which also also indicates the proposed 3D CNN framework is more robust to noise compared with DBN.

Simultaneous estimation and segmentation from projection data in dynamic PET

PURPOSE

Dynamic positron emission tomography (PET) is known for its ability to extract spatiotemporal information of a radio tracer in living tissue. Information of different functional regions based on an accurate reconstruction of the activity images and kinetic parametric images has been widely studied and can be useful in research and clinical setting for diagnosis and other quantitative tasks. In this paper, our purpose is to present a novel framework for estimating the kinetic parametric images directly from the raw measurement data together with a simultaneous segmentation accomplished through kinetic parameters clustering.

METHOD

An iterative framework is proposed to estimate the kinetic parameter image, activity map and do the segmentation simultaneously from the complete dynamic PET projection data. The clustering process is applied to the kinetic parameter variable rather than to the traditional activity distribution so as to achieve accurate discrimination between different functional areas. Prior information such as total variation regularization is incorporated to reduce the noise in the PET images and a sparseness constraint is integrated to guarantee the solution for kinetic parameters due to the over complete dictionary. Alternating direction method of multipliers (ADMM) method is used to solve the optimization problem. The proposed algorithm was validated with experiments on Monte Carlo-simulated phantoms and real patient data. Symbol error rate (SER) was defined to evaluate the performance of clustering. Bias and variance of the reconstruction activity images were calculated based on ground truth. Relative mean square error (MSE) was used to evaluate parametric results quantitatively.

RESULT

In brain phantom experiment, when counting rate is 1 × 10⁶ , the bias (variance) of our method is 0.1270 (0.0281), which is lower than maximum likelihood expectation maximization (MLEM) 0.1637 (0.0410) and direct estimation without segmentation (DE) 0.1511 (0.0326). In the Zubal phantom experiment, our method has the lowest bias (variance) 0.1559 (0.0354) with 1 × 10⁵ counting rate, compared with DE 0.1820 (0.0435) and MLEM 0.3043 (0.0644). As for classification, the SER of our method is 18.87% which is the lowest among MLEM + k-means, DE + k-means, and kinetic spectral clustering (KSC). Brain data with MR reference and real patient results also show that the proposed method can get images with clear structure by visual inspection.

CONCLUSION

In this paper, we presented a joint reconstruction framework for simultaneously estimating the activity distribution, parametric images, and parameter-based segmentation of the ROIs into different functional areas. Total variation regularization is performed on the activity distribution domain to suppress noise and preserve the edges between ROIs. An over complete dictionary for time activity curve basis is constructed. SER, bias, variance, and MSE were calculated to show the effectiveness of the proposed method.

4D-PET reconstruction using a spline-residue model with spatial and temporal roughness penalties

4D reconstruction of dynamic positron emission tomography (dPET) data can improve the signal-to-noise ratio in reconstructed image sequences by fitting smooth temporal functions to the voxel time-activity-curves (TACs) during the reconstruction, though the optimal choice of function remains an open question. We propose a spline-residue model, which describes TACs as weighted sums of convolutions of the arterial input function with cubic B-spline basis functions. Convolution with the input function constrains the spline-residue model at early time-points, potentially enhancing noise suppression in early time-frames, while still allowing a wide range of TAC descriptions over the entire imaged time-course, thus limiting bias. Spline-residue based 4D-reconstruction is compared to that of a conventional (non-4D) maximum a posteriori (MAP) algorithm, and to 4D-reconstructions based on adaptive-knot cubic B-splines, the spectral model and an irreversible two-tissue compartment ('2C3K') model. 4D reconstructions were carried out using a nested-MAP algorithm including spatial and temporal roughness penalties. The algorithms were tested using Monte-Carlo simulated scanner data, generated for a digital thoracic phantom with uptake kinetics based on a dynamic [18F]-Fluromisonidazole scan of a non-small cell lung cancer patient. For every algorithm, parametric maps were calculated by fitting each voxel TAC within a sub-region of the reconstructed images with the 2C3K model. Compared to conventional MAP reconstruction, spline-residue-based 4D reconstruction achieved  >50% improvements for five of the eight combinations of the four kinetics parameters for which parametric maps were created with the bias and noise measures used to analyse them, and produced better results for 5/8 combinations than any of the other reconstruction algorithms studied, while spectral model-based 4D reconstruction produced the best results for 2/8. 2C3K model-based 4D reconstruction generated the most biased parametric maps. Inclusion of a temporal roughness penalty function improved the performance of 4D reconstruction based on the cubic B-spline, spectral and spline-residue models.

Automated Whole-Body Bone Lesion Detection for Multiple Myeloma on 68Ga-Pentixafor PET/CT Imaging Using Deep Learning Methods

The identification of bone lesions is crucial in the diagnostic assessment of multiple myeloma (MM). 68Ga-Pentixafor PET/CT can capture the abnormal molecular expression of CXCR-4 in addition to anatomical changes. However, whole-body detection of dozens of lesions on hybrid imaging is tedious and error prone. It is even more difficult to identify lesions with a large heterogeneity. This study employed deep learning methods to automatically combine characteristics of PET and CT for whole-body MM bone lesion detection in a 3D manner. Two convolutional neural networks (CNNs), V-Net and W-Net, were adopted to segment and detect the lesions. The feasibility of deep learning for lesion detection on 68Ga-Pentixafor PET/CT was first verified on digital phantoms generated using realistic PET simulation methods. Then the proposed methods were evaluated on real 68Ga-Pentixafor PET/CT scans of MM patients. The preliminary results showed that deep learning method can leverage multimodal information for spatial feature representation, and W-Net obtained the best result for segmentation and lesion detection. It also outperformed traditional machine learning methods such as random forest classifier (RF), k-Nearest Neighbors (k-NN), and support vector machine (SVM). The proof-of-concept study encourages further development of deep learning approach for MM lesion detection in population study.

Direct Parametric Reconstruction With Joint Motion Estimation/Correction for Dynamic Brain PET Data

Direct reconstruction of parametric images from raw photon counts has been shown to improve the quantitative analysis of dynamic positron emission tomography (PET) data. However it suffers from subject motion which is inevitable during the typical acquisition time of 1-2 hours. In this work we propose a framework to jointly estimate subject head motion and reconstruct the motion-corrected parametric images directly from raw PET data, so that the effects of distorted tissue-to-voxel mapping due to subject motion can be reduced in reconstructing the parametric images with motion-compensated attenuation correction and spatially aligned temporal PET data. The proposed approach is formulated within the maximum likelihood framework, and efficient solutions are derived for estimating subject motion and kinetic parameters from raw PET photon count data. Results from evaluations on simulated [¹¹C]raclopride data using the Zubal brain phantom and real clinical [¹⁸F]florbetapir data of a patient with Alzheimer's disease show that the proposed joint direct parametric reconstruction motion correction approach can improve the accuracy of quantifying dynamic PET data with large subject motion.

Update on novel trends in PET/CT technology and its clinical applications

After a brief history of the major evolutions of positron emission tomography since its introduction in 1972, this article reviews the recent improvements and novel trends in positron emission tomography with a special focus on the time of flight that is currently the major research topic. Novel emerging acquisition modalities, such as dual tracer acquisition, inline hadron therapy dose imaging and yttrium-90 imaging are reviewed.