Application of PET-MRI registration techniques to cat brain imaging

Automated analysis of small animal PET studies through deformable registration to an atlas

Purpose

This work aims to develop a methodology for automated atlas-guided analysis of small animal positron emission tomography (PET) data through deformable registration to an anatomical mouse model.

Methods

A non-rigid registration technique is used to put into correspondence relevant anatomical regions of rodent CT images from combined PET/CT studies to corresponding CT images of the Digimouse anatomical mouse model. The latter provides a pre-segmented atlas consisting of 21 anatomical regions suitable for automated quantitative analysis. Image registration is performed using a package based on the Insight Toolkit allowing the implementation of various image registration algorithms. The optimal parameters obtained for deformable registration were applied to simulated and experimental mouse PET/CT studies. The accuracy of the image registration procedure was assessed by segmenting mouse CT images into seven regions: brain, lungs, heart, kidneys, bladder, skeleton and the rest of the body. This was accomplished prior to image registration using a semi-automated algorithm. Each mouse segmentation was transformed using the parameters obtained during CT to CT image registration. The resulting segmentation was compared with the original Digimouse atlas to quantify image registration accuracy using established metrics such as the Dice coefficient and Hausdorff distance. PET images were then transformed using the same technique and automated quantitative analysis of tracer uptake performed.

Results

The Dice coefficient and Hausdorff distance show fair to excellent agreement and a mean registration mismatch distance of about 6 mm. The results demonstrate good quantification accuracy in most of the regions, especially the brain, but not in the bladder, as expected. Normalized mean activity estimates were preserved between the reference and automated quantification techniques with relative errors below 10 % in most of the organs considered.

Conclusion

The proposed automated quantification technique is reliable, robust and suitable for fast quantification of preclinical PET data in large serial studies.

Correlation of biological aggressiveness assessed by 11C-methionine PET and hypoxic burden assessed by 18F-fluoromisonidazole PET in newly diagnosed glioblastoma

PURPOSE

Glioblastoma multiforme (GBM) is characterized by tissue hypoxia associated with resistance to radiotherapy and chemotherapy. To clarify the biological link between hypoxia and tumour-induced neovascularization and tumour aggressiveness, we analysed detailed volumetric and spatial information of viable hypoxic tissue assessed by (18)F-fluoromisonidazole (FMISO) PET relative to neovascularization in Gd-enhanced MRI and tumour aggressiveness by L-methyl-(11)C-methionine (MET) PET in newly diagnosed GBMs.

METHODS

Ten patients with newly diagnosed GBMs were investigated with FMISO PET, MET PET and Gd-enhanced MRI before surgery. Tumour volumes were calculated by performing a three-dimensional threshold-based volume of interest (VOI) analysis for metabolically active volume on MET PET (MET uptake indices of ≥1.3 and ≥1.5) and Gd-enhanced volume on MRI. FMISO PET was scaled to the blood FMISO activity to create tumour to blood (T/B) images. The hypoxic volume (HV) was defined as the region with T/B greater than 1.2. PET and MR images of each patient were coregistered to analyse the spatial location of viable hypoxic tissue relative to neovascularization and active tumour extension.

RESULTS

Metabolically active tumour volumes defined using MET uptake indices of ≥1.3 and ≥1.5 and the volumes of Gd enhancement showed a strong correlation (r = 0.86, p < 0.01 for an index of ≥1.3 and r = 0.77, p < 0.05 for an index of ≥1.5). The HVs were also excellently correlated with the volumes of Gd enhancement (r = 0.94, p < 0.01). The metabolically active tumour volumes as defined by a MET uptake index of ≥1.3 and the HVs exhibited a strong correlation (r = 0.87, p < 0.01). On superimposed images, the metabolically active area on MET PET defined by a MET uptake index of ≥1.3 was usually larger than the area of the Gd enhancement and about 20-30% of the MET area extended outside the area of the enhancement. On the other hand, the surface area of viable hypoxic tissue with a T/B cutoff of ≥1.2 on FMISO PET did not substantially differ from the area of the Gd enhancement.

CONCLUSION

The volumetric analysis demonstrates that the viable hypoxic tissue assessed by FMISO PET is related to the neovascularization in Gd-enhanced MRI and the tumour aggressiveness by MET PET in newly diagnosed GBMs. The spatial analysis shows that the metabolically active tumour may be substantially underestimated by Gd-enhanced MRI. Complementary use of MET and FMISO to Gd-enhanced MRI may improve the understanding of tumour biology and lead to the most efficient delineation of tumour volume and treatment strategy.

Preclinical Multimodality Imaging in Vivo

Multimodality small-animal molecular imaging has become increasingly important as transgenic and knockout mice are produced to model human diseases. With the ever-increasing number and importance of human disease models, particularly in rodents (mice and rats), the ability of high-resolution multimodality molecular imaging instrumentation to contribute unique information is becoming more common and necessary. Multimodality imaging with high spatial resolution and good sensitivity, which combines modalities and records sequentially or simultaneously complementary information, offers many advantages in certain research experiments. This article discusses the current trends and new horizons in preclinical multimodality imaging in-vivo and its role in biomedical research.

Deformable and rigid registration of MRI and microPET images for photodynamic therapy of cancer in mice

We are investigating imaging techniques to study the tumor response to photodynamic therapy (PDT). Positron emission tomography (PET) can provide physiological and functional information. High-resolution magnetic resonance imaging (MRI) can provide anatomical and morphological changes. Image registration can combine MRI and PET images for improved tumor monitoring. In this study, we acquired high-resolution MRI and microPET 18F-fluorodeoxyglucose (FDG) images from C3H mice with RIF-1 tumors that were treated with Pc 4-based PDT. We developed two registration methods for this application. For registration of the whole mouse body, we used an automatic three-dimensional, normalized mutual information algorithm. For tumor registration, we developed a finite element model (FEM)-based deformable registration scheme. To assess the quality of whole body registration, we performed slice-by-slice review of both image volumes; manually segmented feature organs, such as the left and right kidneys and the bladder, in each slice; and computed the distance between corresponding centroids. Over 40 volume registration experiments were performed with MRI and microPET images. The distance between corresponding centroids of organs was 1.5 +/- 0.4 mm which is about 2 pixels of microPET images. The mean volume overlap ratios for tumors were 94.7% and 86.3% for the deformable and rigid registration methods, respectively. Registration of high-resolution MRI and microPET images combines anatomical and functional information of the tumors and provides a useful tool for evaluating photodynamic therapy.

Quantitative evaluation of cerebral hemodynamics in patients with moyamoya disease by dynamic susceptibility contrast magnetic resonance imaging--comparison with positron emission tomography

We examined whether the degree of hemodynamic stress in patients with chronic occlusive cerebral vascular disease can be quantitatively evaluated with the use of perfusion-weighted magnetic resonance imaging (PWI). Thirty-six patients with moyamoya disease (mean age, 26.8 years; range, 18 to 59) underwent PWI and positron emission tomography (PET) within a month's interval. The PWI data were calculated by three different analytic methods. The cerebral blood flow (CBF) ratio, cerebral blood volume (CBV) ratio, and mean transit time (MTT) of the anterior circulation were calculated using the cerebellum as a control region and compared with PET data on the same three parameters and oxygen extraction fraction (OEF). Parametric maps of PWI attained a higher resolution than the PET maps and revealed focal perfusion failure on a gyrus-by-gyrus level. The relative CBV and MTT obtained with PWI showed significant linear correlations with the corresponding PET values (CBV, R2 = 0.47 to 0.58; MTT, R2 = 0.32 to 0.68). We also found that we could detect regions with abnormally elevated OEF and CBV based on the delay of PWI-measured MTT relative to the control region by defining a 2.0-sec delay as a threshold. The sensitivity and specificity were 92.3% and 100% in detecting regions with abnormally elevated OEF, and 20.0% and 100% in detecting regions with abnormally elevated CBV, respectively. Among the parameters obtained with PWI, our results suggested that the relative CBV value and delay of MTT might be quantitatively manipulated to assist in clinical decision-making for patients with moyamoya disease.

Automatic image registration of three-dimensional images of the head of cats and dogs by use of maximization of mutual information

OBJECTIVE

To validate mutual information criterion as a ready-to-use technique for automated alignment (ie, registration) of 3-dimensional (3-D) multimodal image data of the head of cats and dogs.

SAMPLE POPULATION

Corresponding 3-D magnetic resonance imaging (MRI) and computed tomography (CT) brain scans of a 6-month-old Doberman Pinscher with a brain cyst; CT images of the head of a European shorthair cat with a meningioma before and immediately, 3, and 6 months after surgical resection; and CT and corresponding stacked anatomic cryosection images of the entire head of a 2-year-old sexually intact female Beagle.

PROCEDURE

All images were matched retrospectively by use of an in-house computer program developed on the basis of a mutual information image registration algorithm. Accuracy of the resulting registrations was evaluated by visual inspection.

RESULTS

All registrations were judged to be highly accurate. Additional manual corrections were not necessary.

CONCLUSIONS AND CLINICAL RELEVANCE

Mutual information registration criterion can by applied to 3-D multimodal head images of cats and dogs for full automatic rigid-body image registration. The combination of such aligned images would considerably facilitate efforts of veterinary clinicians as indicated by its widespread use in brain surgery and radiation therapy of humans.

PET neuroreceptor imaging as predictor of severe cerebral ischemic insult

Measurement of the adenosine A1 receptor (A1-R) with positron emission tomography (PET) using a newly developed positron ligand, [1-methyl-11C]8-dicyclopropylmethyl-1-methyl-3-propylxanthine (MPDX). were performed in a cat middle cerebral artery (MCA) occlusion and reperfusion. Eighteen adult cats underwent PET measurement of; 1) cerebral blood flow (CBF). 2) A1-R, 3) central benzodiazepine receptor (BDZ-R) and 4) glucose metabolism with 15O labeled water, MPDX, 11C-flumazenil (FMZ) and 18F-fluorodeoxyglucose (FDG), respectively. The CBF, A1-R, BDZ-R and FDG uptake were serially measured after 60 min occlusion of MCA in this order. MPDX binding and FMZ binding, but not CBF and FDG uptake, were significantly reduced in the groups with severer ischemic insult than in the groups with no or milder insults. Of the two receptor ligands, the reduction rate of the MPDX binding to A1-Rs was larger in a group that caused fatal ischemic insult. The newly developed PET in vivo imaging technique using MPDX was suitable in evaluating the function of adenosine and A1-Rs in relation to cerebral ischemia.

Quantitative in vivo measurement of central benzodiazepine receptors in the brain of cats by use of positron-emission tomography and [11C]flumazenil

OBJECTIVE

To map central benzodiazepine receptors (BZRs) in the brain of cats by use of positron-emission tomography (PET) and [11C]flumazenil.

ANIMALS

6 male cats that weighed between 2.0 and 3.6 kg.

PROCEDURE

Brain images obtained by PET evaluation of [11C]flumazenil were superimposed on T2-weighted magnetic-resonance imaging (MRI) scans of the same cats. Detailed anatomic regions, such as the cerebral cortex, striatum, thalamus, midbrain, and cerebellum, on the PET images were evident by PET-MRI registration. Regional binding of [11C]flumazenil to BZRs was quantitatively measured by use of a model with 2 tissue compartments and 4 variables.

RESULTS

The highest value for distribution volume was observed in the cerebral cortex, and the lowest value was found in the midbrain of cats.

CONCLUSIONS AND CLINICAL RELEVANCE

Binding of [11C]flumazenil to BZRs in the brain of cats can be quantitatively measured by use of PET with the aid of PET-MRI registration. It is difficult to diagnose changes in these neuroreceptors within the field of current veterinary science. In the future, PET should prove useful for investigating and diagnosing brain disorders in animals in clinical settings.

Non-invasive imaging methods for the characterization of the pathophysiology of brain ischemia

Non-invasive imaging methods are increasingly used to study the evolution and therapy of brain diseases under both clinical and experimental conditions. In the animal experiment, these methods can be supplemented by invasive tissue assays to allow precise characterization of the underlying pathophysiology. Based on such an approach, this review evaluates the importance of in vivo nuclear magnetic resonance (NMR) and positron emission tomography (PET) for the understanding of the pathophysiology of brain ischemia.

Mapping adenosine A(1) receptors in the cat brain by positron emission tomography with [(11)C]MPDX

We evaluated the potential of [(11)C]MPDX as a radioligand for mapping adenosine A(1) receptors in comparison with previously proposed [(11)C]KF15372 in cat brain by PET. Two tracers showed the same brain distribution. Brain uptake of [(11)C]MPDX (Ki = 4.2 nM) was much higher and washed out faster than that of [(11)C]KF15372 (Ki = 3.0 nM), and was blocked by carrier-loading or displaced with an A(1) antagonist. The regional A(1) receptor distribution evaluated with kinetic analysis is consistent with that previously measured in vitro. [(11)C]MPDX PET has a potential for mapping adenosine A(1) receptors in brain.

A device for feline head positioning and stabilization during magnetic resonance imaging

Minimization of head movement and reproduction of standard head positions are essential for reliable brain functional magnetic resonance imaging. Devices for stabilization and alignment of feline preparations are not available currently. We describe a system that involves minimal surgery, allows for both acute and chronic atraumatic positioning, and has the potential to be used for unanesthetized animals. The device uses non-metallic materials and stabilizes the head by means of an apparatus that fixes the head with nylon screws and dental cement in the frontal sinuses. Application of the head-stabilizing device decreases head movements by more than a factor of ten. Anatomical images show that this device provides 3 dimensional head placement at a precision comparable to that of a stereotactic frame, i.e. within 1 mm.