Katie Dinelle

Accurate Event-Driven Motion Compensation in High-Resolution PET Incorporating Scattered and Random Events

With continuing improvements in spatial resolution of positron emission tomography (PET) scanners, small patient movements during PET imaging become a significant source of resolution degradation. This work develops and investigates a comprehensive formalism for accurate motion-compensated reconstruction which at the same time is very feasible in the context of high-resolution PET. In particular, this paper proposes an effective method to incorporate presence of scattered and random coincidences in the context of motion (which is similarly applicable to various other motion correction schemes). The overall reconstruction framework takes into consideration missing projection data which are not detected due to motion, and additionally, incorporates information from all detected events, including those which fall outside the field-of-view following motion correction. The proposed approach has been extensively validated using phantom experiments as well as realistic simulations of a new mathematical brain phantom developed in this work, and the results for a dynamic patient study are also presented.

System matrix modelling of externally tracked motion.

In high resolution emission tomography imaging, even small patient movements can considerably degrade image quality. This work investigates an approach to motion compensated reconstruction of motion-contaminated data, thus applicable to any scanner in the field (e.g. without list-mode acquisition capability), assuming externally-tracked motion information; it involves incorporation of the measured motion information into the system matrix of the EM algorithm. Furthermore, it is shown that the effect of motion-contamination of the attenuation factors should also be modeled and taken into account in the reconstruction task.

Direct intranigral administration of an ubiquitin proteasome system inhibitor in rat: Behavior, positron emission tomography, immunohistochemistry

Several independent lines of research suggest that disruption of the ubiquitin proteasome system (UPS) may play a role in the pathophysiology of Parkinsons disease. Direct intracerebral injection of UPS inhibitors (e.g. lactacystin) in animals has consistently produced important features of the disease. In this study, a range of lactacystin doses (0.5, 1, 2, 10 and 20 μg) were injected into the right substantia nigra in rats to determine the ideal dose required to produce a robust and specific lesion of the dopamine nigro-striatal system and motor deficits. Motor behavior, assessed with the tapered ledged beam task, was severely affected in animals that received high doses (10 and 20 μg) but only mild, impairments were observed in animals that received low doses (0.5, 1, and 2 μg). Positron emission tomography was performed with a dedicated small animal scanner on the rats following the injection of the radio-labeled tracer (±)[(11)C]dihydrotetrabenazine (DTBZ) which labels vesicular monoamine transporter type 2. Severe loss of [(11)C]DTBZ binding in the ipsilateral striatum was observed in the higher dose groups and mild loss was observed in the low dose groups. Stereological cell counting of tyrosine hydroxylase immunoreactive cells in the substantia nigra and the ventral tegmental area indicated a dose dependent loss of dopaminergic neurons. Significant correlations were found between the behavioral motor deficits, striatal [(11)C]DTBZ binding and cell counts of tyrosine hydroxylase immunoreactive cells. Taken together these results indicate that intranigral injection of lactacystin produces dose dependent effects on the dopamine nigro-striatal system and a dose of 10 μg will produce a consistent severe lesion.

Anterior brain glucose hypometabolism predates dementia in progranulin mutation carriers.

Objective: In this prospective cohort study, we investigated cerebral glucose metabolism reductions on [18F]-fluorodeoxyglucose (FDG)-PET in progranulin (GRN) mutation carriers prior to frontotemporal dementia (FTD) onset. Methods: Nine mutation carriers (age 51.5 ± 13.5 years) and 11 noncarriers (age 52.7 ± 9.5 years) from 5 families with FTD due to GRN mutations underwent brain scanning with FDG-PET and MRI and clinical evaluation. Normalized FDG uptake values were calculated with reference to the pons. PET images were analyzed with regions of interest (ROI) and statistical parametric mapping (SPM) approaches. Results: Compared with noncarriers, GRN mutation carriers had a lowered anterior-to-posterior (AP) ratio of FDG uptake (0.86 ± 0.09 vs 0.92 ± 0.05) and less left-right asymmetry, consistent with an overall pattern of right anterior cerebral hypometabolism. This pattern was observed regardless of whether they were deemed clinically symptomatic no dementia or asymptomatic. Individual ROIs with lowered FDG uptake included right anterior cingulate, insula, and gyrus rectus. SPM analysis supported and extended these findings, demonstrating abnormalities in the right and left medial frontal regions, right insular cortex, right precentral and middle frontal gyri, and right cerebellum. Right AP ratio was correlated with cognitive and clinical scores (modified Mini-Mental State Examination r = 0.74; Functional Rating Scale r = −0.73) but not age and years to estimated onset in mutation carriers. Conclusion: The frontotemporal lobar degenerative process associated with GRN mutations appears to begin many years prior to the average age at FTD onset (late 50s–early 60s). Right medial and ventral frontal cortex and insula may be affected in this process but the specific regional patterns associated with specific clinical variants remain to be elucidated.

Positron emission tomography kinetic modeling algorithms for small animal dopaminergic system imaging

Small animal positron emission tomography (PET) imaging allows in vivo quantification of lesion‐ or treatment‐induced neurochemical changes in animal models of disease. Important for quantification are the kinetic modeling methods used to determine biologically‐relevant parameters of tracer‐tissue interaction. In this work, we evaluate modeling algorithms for the dopaminergic tracers 11C‐dihydrotetrabenazine (DTBZ), 11C‐methylphenidate (MP), and 11C‐raclopride (RAC), used to image the dopaminergic system in the unilateral 6‐hydroxydopamine lesioned rat model of Parkinsons disease. For the presynaptic tracers, PET measures are compared with autoradiographic binding measurements using DTBZ and [3H]WIN 35,428 (WIN). We independently developed a new variant of the tissue‐input Logan graphical modeling method, and compared its performance with the simplified Logan graphical method and the simplified reference tissue with basis functions method (SRTM), for region of interest (ROI) averaged time activity curves (TACs) and parametric imaging. The modified graphical method was found to be effectively unbiased by target tissue noise and has advantages for parametric imaging, while all tested methods were equivalent for ROI‐averaged data. Synapse 64:200–208, 2010.

PBB3 imaging in Parkinsonian disorders: Evidence for binding to tau and other proteins

Background and Objectives: To study selective regional binding for tau pathology in vivo, using PET with [11C]PBB3 in PSP patients, and other conditions not typically associated with tauopathy.

A scan without evidence is not evidence of absence: Scans without evidence of dopaminergic deficit in a symptomatic leucine‐rich repeat kinase 2 mutation carrier

The basis for SWEDD is unclear, with most cases representing PD mimics but some later developing PD with a dopaminergic deficit.

Motion estimation for functional medical imaging studies using a stereo video head pose tracking system

The accuracy of functional medical imaging modalities used to assess brain functions are known to be easily degraded by patients’ head movement, due to the extensive duration and the increasing resolution of the scanner. In positron emission tomography these corruptions can cause tracer concentrations to appear blurred or originating from erroneous locations on the final image. Many researches have been using external markers to provide an accurate motion measurement. In this work, we provide a marker-less framework that can track the patient’s 3D head pose during the scan. The framework approaches the problem by combining stereo vision, feature point detection, and the Unscented Kalman Filter. By utilizing features directly available on a patient’s face for tracking, the work eliminates the need of markers common to most current approaches, and therefore effectively minimizes any scanning preparation times and patients’ discomfort. Initial visual inspections show this approach is able to retrieve final transformation parameters matching the extracted feature points with the actual head motion. This framework can be extended to any imaging modality that is affected by patients’ movement.

Methods for Parkinson’s rat model PET image analysis with regions of interest

Accurate methods are required for analysis of microPET dopamine (DA) receptor or transporter images of unilaterally 6-hydroxydopamine-lesioned rat models of Parkinsons disease. Heavily lesioned striata and the cerebellum do not appear distinctly in PET images when presynaptic tracers such as [11C]-(+)-dihydrotetrabenazine (DTBZ) are used, and are difficult targets on which to place reliably regions of interest (ROIs) without additional guidance. Registration of a brain atlas to DA receptor/transporter images significantly improves reproducibility and reliability of ROI-based analyses, as measured by discrepancy between calculated binding potentials (BP) of repeated scans of the same animal, and correlation with autoradiographic binding measurements with the same tracer (DTBZ). Averaging over 3 or 5 axial planes to generate time activity curves gives equivalent reproducibility and reliability. Scan-to-scan coregistration with automated image registration (AIR) can be successful with appropriate masking. Coregistered image analysis produces statistically equivalent results to separately placing ROIs on images that are being directly compared, though coregistered images require only one set of ROIs to be placed, reducing analysis effort.

Quality control protocol for frame-to-frame PET motion correction

Subject motion during Position Emission Tomography (PET) brain scans can reduce image quality, and may lead to incorrect biological outcome measures, especially during analysis of dynamic data sets. This is particularly relevant when imaging with state-of-the-art scanners such as the High Resolution Research Tomograph (HRRT, Siemens Medical Solutions). Motion correction via frame-to-frame image realignment is simpler to implement and requires fewer computing resources than methods that correct for motion during data reconstruction and has been shown to significantly improve the accuracy of dynamically-derived biological variables. However, an ongoing problem is a lack of objective criteria to validate the accuracy of frame-to-frame realignment. Visual inspection of realigned images is a common method of validation but requires a significant amount of operator time and results may vary from one operator to another. This work presents a quality control protocol that automatically flags inadequate realignments based on the comparison of motion transformation matrices obtained from two independent sources: the Polaris Vicra optical tracking device and the image based realignment algorithm AIR (Automated Image Registration). A metric was computed to determine the difference between the transformations from both methods. Realignments were accepted or flagged based on the value of the metric. Since the two methods rely on independent motion assessment tools, the chance of both algorithms giving consistently wrong estimates is low. Human test cases show that the quality control protocol is capable of correctly identifying both acceptable and incorrect realigned images, thus providing an objective quality control metric. Implementation of the protocol reduces the number of images requiring visual inspection by 72% and operator time required by 50%, decreasing both operator labour and operator-dependent biases.