Andrew L. Goertzen

Although not classically considered together, there is emerging evidence that Alzheimer’s disease (AD) and epilepsy share a number of features and that each disease predisposes patients to developing the other. Using machine learning, we have previously developed an automated fluorodeoxyglucose positron emission tomography (FDG-PET) reading program (i.e., MAD), and demonstrated good sensitivity (84%) and specificity (95%) for differentiating AD patients versus healthy controls. In this retrospective chart review study, we investigated if epilepsy patients with/without mild cognitive symptoms also show AD-like metabolic patterns determined by the MAD algorithm. Scans from a total of 20 patients with epilepsy were included in this study. Because AD diagnoses are made late in life, only patients aged ≥40 years were considered. For the cognitively impaired patients, four of six were identified as MAD+ (i.e., the FDG-PET image is classified as AD-like by the MAD algorithm), while none of the five cognitively normal patients was identified as MAD+ (χ² = 8.148, p = 0.017). These results potentially suggest the usability of FDG-PET in prognosticating later dementia development in non-demented epilepsy patients, especially when combined with machine learning algorithms. A longitudinal follow-up study is warranted to assess the effectiveness of this approach.

Dementia is broadly characterized by cognitive and psychological dysfunction that significantly impairs daily functioning. Dementia has many causes including Alzheimer’s disease (AD), dementia with Lewy bodies (DLB), and frontotemporal lobar degeneration (FTLD). Detection and differential diagnosis in the early stages of dementia remains challenging. Fueled by AD Neuroimaging Initiatives (ADNI) (Data used in preparation of this article were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database. As such, the investigators within ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report.), a number of neuroimaging biomarkers for AD have been proposed, yet it remains to be seen whether these markers are also sensitive to other types of dementia. We assessed AD-related metabolic patterns in 27 patients with diverse forms of dementia (five had probable/possible AD while others had atypical cases) and 20 non-demented individuals. All participants had positron emission tomography (PET) scans on file. We used a pre-trained machine learning-based AD designation (MAD) framework to investigate the AD-related metabolic pattern among the participants under study. The MAD algorithm showed a sensitivity of 0.67 and specificity of 0.90 for distinguishing dementia patients from non-dementia participants. A total of 18/27 dementia patients and 2/20 non-dementia patients were identified as having AD-like patterns of metabolism. These results highlight that many underlying causes of dementia have similar hypometabolic pattern as AD and this similarity is an interesting avenue for future research.

Previously we have developed a first-generation PET insert prototype for small animal PET/MR imaging, which used resistor-based charge division multiplexing circuits and SensL B-series silicon photomultipliers (SiPMs). In this work we present results from a second-generation readout board with improved timing and count rate performance. Three detector boards were tested: the first-generation readout board with SensL SPMArray4B (SiPM-B), the second-generation readout board with SensL ArrayC-30035-16P-PCB (SiPM-C) using the ‘fast’ outputs for timing, and the second generation board using Hamamatsu S11361-3050AE-04 MPPC arrays. Timing data were obtained with detector modules in coincidence with a single-pixel SensL MicroFJ-SMA-30035 reference detector and acquired using standard NIM electronics, while count rate data were acquired using the OpenPET data acquisition electronics system. The full-width at half-maximum (FWHM) coincidence time resolution (CTR) for the SiPM-B, SiPM-C and MPPC designs were 2600 ± 200 ps, 550 ± 50 ps, and 570 ± 30 ps, respectively. OpenPET waveform capture determined the mean signal durations, measured as time above 10% of the maximum amplitude, were 1850 ± 150 ns, 600 ± 25 ns, and 350 ± 25 ns, respectively, where the short signal of the MPPC resulted in reduced pileup effects at higher count rates. Decaying source measurements showed a non-paralyzable dead time of 1.30–1.41 µ s for all three detectors tested, which was limited by the signal capture and processing time of the OpenPET system.

PET detectors commonly feature large numbers of multi-pixel photon counters (MPPCs), requiring data acquisition systems with significant signal multiplexing or using application specific integrated circuits (ASICs) for independent MPPC readout. We evaluate here two Hamamatsu C13500-4075LC-12 detector modules designed for time-of-flight (TOF) PET. Each module has a 12 × 12 lutetium fine silicate (LFS) crystal array (4.2×4.2×20 mm3) one-to-one coupled to a 12×12 array of 4 mm MPPCs. Each detector has 8 18-channel ASICs utilizing time-over-threshold (ToT) readout. At low event rates, the detectors have full-width at half-maximum (FWHM) energy resolution of 11.8%±0.4%, measured in linearized energy spectra, and 275 ps FWHM coincidence timing resolution (CTR) using energy window of ±1 photopeak FWHM. The 511 keV photopeak amplitude in the ToT spectrum changed 0.24%/∘C. There was <10% event loss up to 1450 kcps/detector. From count rates of 35-1450 kcps the 511 keV photopeak position varied by 1.2%, while photopeak FWHM in the ToT spectra increased by 54%, from 5.8% to 9.0%, corresponding to energy resolution changing from 11.8% to approximately 18.0%. Over this count rate range the CTR varied from 285 to 435 ps FWHM. The TOF-PET module performance suggests they are well suited to use in whole-body PET applications.

Utilizing the publicly available neuroimaging database enabled by Alzheimer’s disease Neuroimaging Initiative (ADNI; http://adni.loni.usc.edu/), we have compared the performance of automated classification algorithms that differentiate AD vs. normal subjects using Positron Emission Tomography (PET) with fluorodeoxyglucose (FDG). General linear model, scaled subprofile modeling and support vector machines were examined. Among the tested classification methods, support vector machine with Iterative Single Data Algorithm produced the best performance, i.e., sensitivity (0.84) × specificity (0.95), by 10-fold cross-validation. We have applied the same classification algorithm to four different datasets from ADNI, Health Science Centre (Winnipeg, Canada), Dong-A University Hospital (Busan, S. Korea) and Asan Medical Centre (Seoul, S. Korea). Our data analyses confirmed that the support vector machine with Iterative Single Data Algorithm showed the best performance in prediction of future development of AD from the prodromal stage (mild cognitive impairment), and that it was also sensitive to other types of dementia such as Parkinson’s Disease Dementia and Dementia with Lewy Bodies, and that perfusion imaging using single photon emission computed tomography may achieve a similar accuracy to that of FDG-PET.

PET scanners suffer from a loss of resolution due to parallax error. Depth-of-interaction (DOI) detectors, such as dual-layer offset (DLO) detectors, mitigate the loss of resolution. We are designing a DLO brain dedicated PET insert that fits into the Siemens Magnetom 7T brain MR scanner. A wide range of DLO detector geometries was evaluated through Monte Carlo simulations in this work. The total detector thickness was varied from 10 to 30 mm and for each total thickness, various layer thickness ratios and crystal widths in the range of 0.5 to 4 mm were examined. The effects of each detector geometry on radial mispositioning of the incident photons, detection efficiency ratio between the two layers, coincidence response functions (CRFs), system spatial resolution, and sensitivity were studied. Optimum layer thickness ratio for each total thickness is discussed. Analysis of CRFs showed that the improvement in the CRF of a DLO PET scanner with a total detector thickness of larger than 10 mm is lower for detectors with a crystal width of smaller than 2 mm than for detectors with a larger crystal width. Estimated spatial resolution and sensitivity of a PET scanner with each suggested detector geometry are also reported.

Optical coupling is an important factor in detector design as it improves optical photon transmission by mitigating internal reflections at light-sharing boundaries. In this work we compare optical coupling materials, namely double-sided acrylic polymer tapes and silicone optical grease (SiG), in the context of positron emission tomography. Four double-sided tapes from 3 M of varying thicknesses (0.229 mm–1.016 mm) and adhesive materials (‘100MP’, ‘A100’, and ‘GPA’) were characterized with spectrophotometer measurements as well as photopeak amplitude and energy resolution measurements using lutetium-yttrium oxy-orthosilicate (LYSO) coupled to photomultiplier tubes (PMT) or silicon photomultipliers (SiPMs). Transmission spectra from the spectrophotometer showed over 80% transmission for all tapes at 420 nm and above, with 89.6% and 88.8% transmission for the 0.508 mm and 1.016 mm thick GPA tapes, respectively, at 420 nm. Measurements with single-pixel LYSO-PMT and 4 × 4 array (one-to-one coupled) LYSO-SiPM setups determined that SiG had the greatest photopeak amplitude, with tapes showing 2.1%–14.8% reduction in photopeak amplitude with respect to SiG. Energy resolution changed by less than 4% on a relative basis between tapes and SiG with PMT measurements, however for the SiPM array measurements the energy resolution improved from 15.6% ± 2.7% full-width at half-maximum to 11.4% ± 1.2% for SiG and 1 mm GPA respectively. Data acquired with dual-layer offset LYSO arrays (light sharing detector designs) demonstrated that a detector coupled with 1 mm thick GPA tape produced equivalent detector flood histograms to those from a design coupled with SiG and a 1 mm thick glass lightguide. No significant degradation in photopeak amplitude and energy resolution was observed over five months of measurements, indicating the tapes maintain their coupling integrity over several months. Though minimal photopeak amplitude degradation compared to SiG occurs, double-sided tapes are convenient alternatives for optical coupling materials since they diffuse light intrinsically, acting as a light guide, offer mechanical support and durability, are easily applied and removed from scintillators/photodetectors, and are relatively inexpensive and readily available.