Gary Wisniewski

Evolution and Localization of Postictal Blood Flow Changes in Partial Seizures Demonstrated by SPECT: Use of Quantitative Difference Images

Abstract Ictal single photon emission computed tomography (SPECT) is a sensitive means of seizure localization in partial epilepsy. The purpose of this study is to enhance the diagnostic yield of postictal SPECT for epilepsy and to overcome the difficulties in its conventional interpretation. We developed a method for coregistration, normalization, and subtraction of interictal from ictal SPECT to reveal positive difference images (and ictal from interictal to reveal negative difference images) of cerebral blood flow (CBF) during seizures. This method also allows quantification of regions of greatest percent increase and decrease, normalized to seizure duration. We studied 12 patients who had confirmed epileptogenic regions by surgery and pathology (seven temporal, five extratemporal). In all 12 patients, with either early (during ictus) or late injection (after seizure termination), the difference images showed increases or decreases respectively in blood flow consistent with the epileptogenic region. We highlight the localizing value of the negative difference change in perfusion (state of hypoperfusion) that persists for up to 100 seconds after seizure termination and corresponds to the epileptogenic region. Application of this technique for the quantification of perfusion changes during seizure or immediately after seizure cessation enhances the diagnostic yield of SPECT in both temporal and extratemporal epilepsy.

Understanding Fourier space and filter selection.

ConclusionBy appreciating the technical basis of Fourier representations and filtering, a better interpretation of SPECT images can be gained. The ramp filter has no selectable parameters and is the required filter used in tomography. The resulting noisy reconstructed image is smoothed by using a filter where the cut-off and order can be selected. By relating frequencies to sizes of structures in the image, reasonable values for the cut-off frequency for the smoothing filter can be selected. Most important, the realization that the ramp-filtered image (without the smoothing filter applied) is the most correct image means that subsequent smoothing can slightly degrade the numeric correctness of the reconstructed image. Smoothing filters make the image easier for clinical interpretation because noisy structures are difficult for the human eye to perceive. Selection of the smoothing filter to maximize noise reduction and image structure preservation is accomplished by matching the cut-off frequency to the image noise or camera resolution. This understanding should reduce the amount of time spent searching for a smoothing filter, which is used routinely in clinical imaging studies.

Human biodistribution and dosimetry of iodine-123-fluoroalkyl analogs of β-CIT

Abstract.Two new N-ω-fluoroalkyl analogs of [123I]2β-carbomethoxy-3β-(4-iodophenyl)tropane ([123I]β-CIT), the fluoroethyl and fluoropropyl compounds ([123I]FE-CIT and [123I]FP-CIT, respectively), have been shown to have faster kinetics and better selectivity for the dopamine transporter than [123I]β-CIT. We examined the organ biodistribution and radiation safety of these two compounds in six healthy volunteers who received an injection with each of the two compounds 2 weeks apart. Data were obtained on the Strichman 860 whole-body scanner. Transmission scans were obtained in all subjects prior to the injection of the radiotracer with a line source and used to derive organ-specific attenuation correction factors. Whole-body planar images were acquired every hour for the first 6 h, and at 24 h. Attenuation-corrected regional conjugate counts were converted into units of activity using a calibration factor obtained for each subject by dividing whole-body conjugate decay-corrected counts from the first acquisition by the injected activity. Radiation dose estimates were on average higher for [123I]CIT-FE than for [123I]CIT-FP, with the lower large intestine receiving the highest exposure: 0.15±13% mGy/MBq (mean ±COV) and 0.12±14% mGy/MBq for [123I]FE-CIT and [123I]FP-CIT, respectively, followed by the upper large intestine and the spleen.

Human biodistribution and dosimetry of the SPECT D2 dopamine receptor radioligand [123I]IBF

The research discussed in this article aimed to characterize better the biodistribution, excretion and radiation dosimetry of the single photon emission computed tomography (SPECT) D2 Dopamine receptor radioligand [123I]IBF. Following administration of 111 +/- 12 MBq [123I]IBF, seven healthy human subjects were scanned serially with a whole body imager over a 48-h period. Transmission images were obtained with a scanning line source for attenuation correction of the emission images. Urine was collected for 48 h to measure the fraction of activity voided by the renal system. Radiation absorbed dose estimates were performed using biokinetic modeling and the Medical Internal Radiation Dose (MIRD) schema. Highest absorbed doses were to the kidney (0.13 +/- 0.02 mGy/MBq) and urinary bladder wall (0.11 +/- 0.01 mGy/MBq). The effective dose equivalent was 0.041 +/- 0.005 mSv/MBq. Peak brain uptake represented 8% of the injected activity. Rapid urinary excretion minimized the absorbed dose to most tissues. The mean cumulative urinary excretion fraction was 69%. Thus [123I]IBF is a promising SPECT agent for imaging the D2 dopamine receptor in humans with high brain uptake and favorable dosimetry.

Quantitative evaluation of anthropomorphic brain phantoms for standardization of amyloid PET SUVrs in alzheimer's Multicenter Therapeutic Trials

We therefore wish to assess the impact on brain atrophy as measured by the boundary shift integral (BSI) of changing from non-accelerated to accelerated MRI acquisitions over a 12-month interval.Methods: Accelerated and non-accelerated scans of 55 early mild cognitive impairment subjects, acquired in the same scanning session, were downloaded from ADNI. Brain regions were automatically delineated by Multi-Atlas Propagation and Segmentation, visually checked andmanually edited if necessary. BSIswere calculated using both a non-accelerated baseline scan and non-accelerated repeat scans (i.e. consistent acquisition), and a non-accelerated baseline scan and an accelerated repeat scan (i.e. changed acquisition). Results: Mean BSI from changed acquisition was 0.12%/year (p1⁄40.02) lower than consistent acquisition (see Table 1 and Fig. 1). Fig. 1(a) shows that most of the differences are close to zero, and 10 BSIs have differences > 0.5%/ year, which consists of all the 3 BSIs from site 21 (marked in red). There was no evidence that the standard deviations were different (p1⁄40.7). Conclusions: We found a relatively small (w0.1% of brain volume) effect on brain volume change in BSI when changing from non-accelerated to accelerated MRI during follow-up. This difference isw1/5 of the annual atrophy rate in normal aging in elderly subjects (w0.5%/year). However there did appear to be site-specific differences; this may reflect greater differences in the accelerated acquisition on that scanner. Going from non-accelerated baseline to acceleratedMRI for follow-up may have surprisingly little effect

Objective SUVr Determination using MRI Segmentation Maps in Florbetaben Beta-amyloid Brain PET Improves Discrimination of Alzheimer's and Controls

Background: The robustness of a PET tracer for visualising b-amyloid in vivo can be assessed by the following metrics: 1) Sensitivity and Specificity of the visual interpretation with clinical diagnosis as the gold standard, 2) The extent to which a number of independent readers agree on the same interpretation of the images, 3) Reader self consistency and 4) The confidence in which readers are able to make their interpretation. The performance of [F]flutemetamol was assessed against these metrics within the primary objective of the ALZ201 phase-II clinical trial. Methods: Five independent readers were trained on phase I images and classified 72 [F]flutemetamol randomised PET images of the whole brain into the categories of normal or abnormal appearance where abnormal refers to the ability of a reader to identify tracer binding in the grey matter. Readers also rated their confidence in their assessment on a 5 point scale. The clinical classification of these subjects was 27 probable Alzheimer’s Disease (pAD), 25 healthy volunteers (HV) and 20 amnestic Mild Cognitive Impairment (aMCI). The classification based on visual assessment was done blinded to the clinical status of the subjects and also to any quantitative evaluation. In a separate read, the same assessments and metrics were obtained for 20 of the pADs and all of the aMCI subjects who were also images with [C]PiB, typically on the same day as [F]flutemetamol. Results: Classification based on the visual assessment of [F]flutemetamol images categorised 25 of 27 AD as abnormal, 24 of 25 HV as normal and 9 of 20 MCI as abnormal. For the pAD & HV cohort this provided a sensitivity of 92.6% and specificity of 96.0%. Interreader agreement was measured and the reader agreement was 98.7% or a Fleiss’ kappa score of k 1⁄4 0.96. All readers re-read 8 different images and the self consistency was 97.5%. The readers rated their classification as highest confidence in the vast majority of all reads. There was exact concordance between the readers results for the subjects’ [F]flutemetamol and [C]PiB images. Conclusions: Visual assessment of [F]flutemetamol images is highly robust and consistent.