Peter J Julyan

One-pass list-mode EM algorithm for high-resolution 3-D PET image reconstruction into large arrays

High-resolution three-dimensional (3-D) positron emission tomography (PET) scanners with high count rate performance, such as the quad-high density avalanche chamber (HIDAC), place new demands on image reconstruction algorithms due to the large quantities of high-precision list-mode data which are produced. Therefore, a reconstruction algorithm is required which can, in a practical time frame, reconstruct into very large image arrays (submillimeter voxels, which range over a large field of view) whilst preferably retaining the precision of the data. This work presents an algorithm which meets these demands: one-pass list-mode expectation maximization (OPL-EM) algorithm. The algorithm operates directly on list-mode data, passes through the data once only, accounts for finite resolution effects in the system model, and can also include regularization. The algorithm performs multiple image updates during its single pass through the list-mode data, corresponding to the number of subsets that the data have been split into. The algorithm has been assessed using list-mode data from a quad-HIDAC and is compared to the analytic reconstruction method 3-D reprojection (RP) with 3-D filtered backprojection.

EM algorithm system modeling by image-space techniques for PET reconstruction

Methodology for PET system modeling using image-space techniques in the expectation maximization (EM) algorithm is presented. The approach, applicable to both list-mode data and projection data, is of particular significance to EM algorithm implementations which otherwise only use basic system models (such as those which calculate the system matrix elements on the fly). A basic version of the proposed technique can be implemented using image-space convolution, in order to include resolution effects into the system matrix, so that the EM algorithm gradually recovers the modeled resolution with each update. The improved system modeling (achieved by inclusion of two convolutions per iteration) results in both enhanced resolution and lower noise, and there is often no need for regularization-other than to limit the number of iterations. Tests have been performed with simulated list-mode data and also with measured projection data from a GE Advance PET scanner, for both [/sup 18/F]-FDG and [/sup 124/I]-NaI. The method demonstrates improved image quality in all cases when compared to the conventional FBP and EM methods presently used for clinical data (which do not include resolution modeling). The benefits of this approach for /sup 124/I (which has a low positron yield and a large positron range, usually resulting in noisier and poorer resolution images) are particularly noticeable.

Phase I Evaluation of a Fully Human Anti–αv Integrin Monoclonal Antibody (CNTO 95) in Patients with Advanced Solid Tumors

Purpose: A fully human monoclonal antibody to anti–αv integrins (CNTO 95) has been shown to inhibit angiogenesis and tumor growth in preclinical studies. We assessed the safety and pharmacokinetics of CNTO 95 in patients with advanced refractory solid tumors. Experimental Design: In this phase I trial, CNTO 95 (0.1, 0.3, 1.0, 3.0, and 10.0 mg/kg) was infused on days 0, 28, 35, and 42, and clinical assessments, dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI), and [18F]-2-fluorodeoxyglucose positron emission tomography (FDG-PET) were done. Patients achieving stable disease or better were eligible for extended dosing every 3 weeks for up to 12 months. Results: Among the 24 enrolled patients, CNTO 95 was associated with one episode of grade III and four episodes of grade II infusion-related fever (all responded to acetaminophen). Of the six patients who received extended dosing, one patient (10.0 mg/kg), with cutaneous angiosarcoma, had a 9-month partial response. Pre- and post-treatment lesion biopsies confirmed tumor cell αv integrin expression, as well as CNTO 95 penetration of the tumor and localization to tumor cells in association with reduced bcl-2 expression. A lesion in one patient (10.0 mg/kg) with stable ovarian carcinosarcoma was no longer detectable by FDG-PET by day 49. Exposure to CNTO 95 seemed to increase in a greater-than-dose-proportional manner; dose-dependent mean half-life ranged from 0.26 to 6.7 days. Conclusions: CNTO 95 was generally well tolerated. Six patients received extended therapy, including one patient with a prolonged response. Biopsy data confirmed tumor localization and pharmacodynamic activity.

Bias in iterative reconstruction of low-statistics PET data: benefits of a resolution model.

Ordered-subset expectation maximization (OSEM) is a widely used method of reconstructing PET data. Several authors have reported bias when reconstructing frames containing few counts via OSEM, although the level of bias reported varies substantially. Such bias may lead to errors in biological parameters as derived via dynamic PET. We examine low-statistics bias in OSEM reconstruction of patient data and estimate the subsequent errors in biological parameter estimates. Patient listmode data were acquired during a [11C]-DASB scan using a brain PET scanner, the high resolution research tomograph (HRRT). These data were sub-sampled to create many independent, low-count replicates. Each replicate was reconstructed with and without the use of an image based resolution model (PSF), from which bias and variance were calculated as a function of the noise equivalent counts (NEC). Time-activity curves were subsequently generated by Monte Carlo simulation and used to study the propagation of bias from the images into the biological parameters of interest, for which noise and bias were based on the NEC. The investigation was complemented by simulation of a PET scanner. Significant bias was observed when reconstructing data of low statistical quality, for both human and simulated data. For human data, this bias was substantially reduced by including a PSF model (e.g. caudate head, 1.7 M NEC, -5.5 % bias with PSF, -13 % bias without PSF). For the observed levels of bias, Monte Carlo simulations predicted biases in the binding potential of -4 and -10 % (with/without PSF). The use of the PSF changed the variance characteristics of the images, reducing variance at the voxel level for low to moderate numbers of iterations. We conclude that OSEM reconstruction of dynamic PET data can yield parameter estimates of acceptable accuracy (for DASB), despite producing biased images at low statistics. This is however dependent upon the application. The use of a resolution model is shown to reduce bias and is thus recommended. The most likely mechanism for this reduction is the suppression of noise. The magnitude of the bias for other tracers and methods of data analysis is yet to be evaluated.

Auditory cortical activation and speech perception in cochlear implant users: Effects of implant experience and duration of deafness

This study aimed to investigate the relationship between outcome following cochlear implantation and auditory cortical activation. It also studied the effects of length of implant use and duration of deafness on the auditory cortical activations. Cortical activity resulting from auditory stimulation was measured using [(18)F]FDG positron emission tomography. In a group of 18 experienced adult cochlear implant users, we found a positive correlation between speech perception and activations in both the primary and association auditory cortices. This correlation was present in a subgroup of experienced implant users but absent in a group of new implant users with similar speech perception abilities. There was a significant negative correlation between duration of deafness and auditory cortical activation. This study gives insights into the relationship between implant speech perception and auditory cortical activation and the influence of duration of preceding deafness and implant experience.

A comparison of PET imaging characteristics of various copper radioisotopes

PurposePET radiotracers which incorporate longer-lived radionuclides enable biological processes to be studied over many hours, at centres remote from a cyclotron. This paper examines the radioisotope characteristics, imaging performance, radiation dosimetry and production modes of the four copper radioisotopes, 60Cu, 61Cu, 62Cu and 64Cu, to assess their merits for different PET imaging applications. MethodsSpatial resolution, sensitivity, scatter fraction and noise-equivalent count rate (NEC) are predicted for 60Cu, 61Cu, 62Cu and 64Cu using a model incorporating radionuclide decay properties and scanner parameters for the GE Advance scanner. Dosimetry for 60Cu, 61Cu and 64Cu is performed using the MIRD model and published biodistribution data for copper(II) pyruvaldehyde bis(N4-methyl)thiosemicarbazone (Cu-PTSM). Results60Cu and 62Cu are characterised by shorter half-lives and higher sensitivity and NEC, making them more suitable for studying the faster kinetics of small molecules, such as Cu-PTSM. 61Cu and 64Cu have longer half-lives, enabling studies of the slower kinetics of cells and peptides and prolonged imaging to compensate for lower sensitivity, together with better spatial resolution, which partially compensates for loss of image contrast. 61Cu-PTSM and 64Cu-PTSM are associated with radiation doses similar to [18F]-fluorodeoxyglucose, whilst the doses for 60Cu-PTSM and 62Cu-PTSM are lower and more comparable with H215O. ConclusionThe physical and radiochemical characteristics of the four copper isotopes make each more suited to some imaging tasks than others. The results presented here assist in selecting the preferred radioisotope for a given imaging application, and illustrate a strategy which can be extended to the majority of novel PET tracers.

[18F]DPA-714: Direct Comparison with [11C]PK11195 in a Model of Cerebral Ischemia in Rats

Purpose Neuroinflammation is involved in several brain disorders and can be monitored through expression of the translocator protein 18 kDa (TSPO) on activated microglia. In recent years, several new PET radioligands for TSPO have been evaluated in disease models. [18F]DPA-714 is a TSPO radiotracer with great promise; however results vary between different experimental models of neuroinflammation. To further examine the potential of [18F]DPA-714, it was compared directly to [11C]PK11195 in experimental cerebral ischaemia in rats. Methods Under anaesthesia, the middle cerebral artery of adult rats was occluded for 60 min using the filament model. Rats were allowed recovery for 5 to 7 days before one hour dynamic PET scans with [11C]PK11195 and/or [18F]DPA-714 under anaesthesia. Results Uptake of [11C]PK11195 vs [18F]DPA-714 in the ischemic lesion was similar (core/contralateral ratio: 2.84±0.67 vs 2.28±0.34 respectively), but severity of the brain ischemia and hence ligand uptake in the lesion appeared to vary greatly between animals scanned with [11C]PK11195 or with [18F]DPA-714. To solve this issue of inter-individual variability, we performed a direct comparison of [11C]PK11195 and [18F]DPA-714 by scanning the same animals sequentially with both tracers within 24 h. In this direct comparison, the core/contralateral ratio (3.35±1.21 vs 4.66±2.50 for [11C]PK11195 vs [18F]DPA-714 respectively) showed a significantly better signal-to-noise ratio (1.6 (1.3–1.9, 95%CI) fold by linear regression) for [18F]DPA-714. Conclusions In a clinically relevant model of neuroinflammation, uptake for both radiotracers appeared to be similar at first, but a high variability was observed in our model. Therefore, to truly compare tracers in such models, we performed scans with both tracers in the same animals. By doing so, our result demonstrated that [18F]DPA-714 displayed a higher signal-to-noise ratio than [11C]PK11195. Our results suggest that, with the longer half-life of [18F] which facilitates distribution of the tracer across PET centre, [18F]DPA-714 is a good alternative for TSPO imaging.

Biodistribution, pharmacokinetics and metabolism of interleukin-1 receptor antagonist (IL-1RA) using [18F]-IL1RA and PET imaging in rats

Positron emission tomography (PET) has the potential to improve our understanding of the preclinical pharmacokinetics and metabolism of therapeutic agents, and is easily translated to clinical studies in humans. However, studies involving proteins radiolabelled with clinically relevant PET isotopes are currently limited. Here we illustrate the potential of PET imaging in a preclinical study of the biodistribution and metabolism of 18F‐labelled IL‐1 receptor antagonist ([18F]IL‐1RA) using a novel [18F]‐radiolabelling technique.

Accelerated list-mode EM algorithm

List-mode data preserves all sampling information in three-dimensional (3-D) PET imaging and can reduce storage requirements for short-time frame acquisitions. List-mode expectation maximization-maximum likelihood (EM-ML), which has been implemented in a number of forms (such as the EM algorithm for list-mode maximum likelihood, the FAIR algorithm and COSEM), is an obvious choice to reconstruct from such data sets when the statistics are low. However, these methods can be slow for large quantities of list-mode data and it is desirable to accelerate them. This work investigates the use of subsets in combination with a relaxation parameter for 3-D list-mode EM reconstructions. Results show that two iterations through the list-mode data are sufficient to yield good quality reconstructions. Furthermore, if counting statistics are good, just one iteration may prove sufficient, opening the way for real-time iterative reconstruction.

Mapping metabolic brain activation during human volitional swallowing: a positron emission tomography study using [18F]fluorodeoxyglucose.

We have previously shown that labelled water positron emission tomography (H215O PET) can be used to identify regional cerebral blood flow (rCBF) changes in the human brain during volitional swallowing. (18F) fluorodeoxyglucose (FDG PET), by comparison, uses a glucose analogue to quantitatively measure regional cerebral glucose metabolism (rCMRglc) rather than rCBF. The main advantage of FDG PET is improved spatial resolution, and because of its pharmacodynamic properties, activation can be performed external to the scanner, allowing subjects to assume more physiologic positions. We therefore conducted a study of the brains metabolic response while swallowing in the erect seated position, using FDG PET. Eight healthy male volunteers were studied with a randomised 2 scan paradigm of rest or water swallowing at 20-second intervals for 30 minutes. Data were analysed with SPM99 using multisubject conditions and covariates design. During swallowing, analysis identified increased rCMRglc (P<0.01) in the following areas: left sensorimotor cortex, cerebellum, thalamus, precuneus, anterior insula, left and right lateral postcentral gyrus, and left and right occipital cortex. Decreased rCMRglc were also seen in the right premotor cortex, right and left sensory and motor association cortices, left posterior insula and left cerebellum. Thus, FDG PET can be applied to measure the brain metabolic activity associated with volitional swallowing and has the advantage of normal task engagement. This has implications for future activation studies in patients, especially those suffering swallowing problems after brain injury.