Search

The second Machine Learning for Health (ML4H) symposium was held both virtually and in-person on November 28, 2022, in New Orleans, Louisiana, USA (Parziale et al.,2022). The symposium included research roundtable sessions to foster discussions between participants and senior researchers on timely and relevant topics for the ML4H community. Encouraged by the successful virtual roundtables in the previous year (Roy et al.,2021), we organized nine in-person and four virtual roundtables at ML4H 2022 (Parziale et al.,2022). A roundtable session included invited senior chairs (with substantial experience in the field), junior chairs (responsible for facilitating the discussion), and attendees from di- verse backgrounds with interest in the session’s topic. This document explains the organization process we used and compiles the takeaways from the roundtable discussions, including recent advances, applications, and open challenges for each topic. We conclude with a summary and lessons learned across all roundtables.

We report a PET detector module (DM) designed for developing a preclinical positron emission tomography (PET)-electron paramagnetic resonance imaging (EPRI) hybrid system. The DM consists of a linear array of eight detector units, each of which is a 12×12 array of lutetium–yttrium oxyorthosilicate (LYSO) crystals read by a 4×4 silicon photomultiplier (SiPM) array. The crystal size is approximately 1.0×1.0×10 mm3. All surfaces of the crystal are polished; except for that coupled to SiPMs they are also covered with BaSO4 to reduce light loss. The pitches of the LYSO and SiPM arrays are about 1.05 mm and 3.2 mm, respectively. The front face of the resulting DM is about 1.28×10.24 cm2 in extext and its thickness is approximately 1.8 cm. A highly multiplexing readout is devised to produce only six outputs for a DM, including two outputs that are derived from the SiPM cathode signals for determining the event time and the active DU, and four outputs that are derived from the SiPM anode signals for determining the event energy and the active crystal within the active DU. At present, these outputs are acquired by waveform sampling and analyzed offline. We have successfully developed two DMs, both showing well discriminated DUs and crystals, and an average energy resolution of about 15%. Even though time-of-flight (ToF) is not needed for the proposed system, our data shows that the DM can potentially achieve a 300-400 ps ToF resolution.

Multiparametric magnetic resonance imaging (mpMRI) is increasingly used for risk stratification and localization of prostate cancer (PCa). Thanks to the great success of deep learning models in computer vision, the potential application for early detection of PCa using mpMRI is imminent.Deep learning analysis of the PROSTATEx dataset.In this study, we show a simple convolutional neural network (CNN) with mpMRI can achieve high performance for detection of clinically significant PCa (csPCa), depending on the pulse sequences used. The mpMRI model with T2-ADC-DWI achieved 0.90 AUC score in the held-out test set, not significantly better than the model using KConvolutional neural networks incorporating multiple pulse sequences show high performance for detection of clinically-significant prostate cancer, and the model including dynamic contrast-enhanced information correlates best with grade.

Advancing continuous health monitoring beyond vital signs to biochemistry will revolutionize personalized medicine. Herein, we report a biosensing platform to achieve remote biochemical monitoring using microparticle-based biosensors and optical coherence tomography (OCT). Stimuli-responsive, polymeric microparticles were designed to serve as freely dispersible biorecognition units, wherein binding with a target biochemical induces volumetric changes of the microparticle. Analytical approaches to detect these submicron changes in 3D using OCT were devised by modeling the microparticle as an optical cavity, enabling estimations far below the resolution of the OCT system. As a proof of concept, we demonstrated the 3D spatiotemporal monitoring of glucose-responsive microparticles distributed throughout a tissue mimic in response to dynamically fluctuating levels of glucose. Deep learning was further implemented using 3D convolutional neural networks to automate the vast processing of the continuous stream of three-dimensional time series data, resulting in a robust end-to-end pipeline with immense potential for continuous in vivo biochemical monitoring.

To identify the optimal threshold insup18/supF-fluoromisonidazole (FMISO) PET images to accurately locate tumor hypoxia by using electron paramagnetic resonance imaging (pOsub2/subEPRI) as ground truth for hypoxia, defined by pOsub2/sub[Formula: see text] 10 mmHg.Tumor hypoxia images in mouse models of SCCVII squamous cell carcinoma (n = 16) were acquired in a hybrid PET/EPRI imaging system 2 h post-injection of FMISO. T2-weighted MRI was used to delineate tumor and muscle tissue. Dynamic contrast enhanced (DCE) MRI parametric images of Ksuptrans/supand vsube/subwere generated to model tumor vascular properties. Images from PET/EPR/MRI were co-registered and resampled to isotropic 0.5 mm voxel resolution for analysis. PET images were converted to standardized uptake value (SUV) and tumor-to-muscle ratio (TMR) units. FMISO uptake thresholds were evaluated using receiver operating characteristic (ROC) curve analysis to find the optimal FMISO threshold and unit with maximum overall hypoxia similarity (OHS) with pOsub2/subEPRI, where OHS = 1 shows perfect overlap and OHS = 0 shows no overlap. The means of dice similarity coefficient, normalized Hausdorff distance, and accuracy were used to define the OHS. Monotonic relationships between EPRI/PET/DCE-MRI were evaluated with the Spearman correlation coefficient ([Formula: see text]) to quantify association of vasculature on hypoxia imaged with both FMISO PET and pOsub2/subEPRI.FMISO PET thresholds to define hypoxia with maximum OHS (both OHS = 0.728 [Formula: see text] 0.2) were SUV [Formula: see text] 1.4 [Formula: see text] SUVsubmean/suband SUV [Formula: see text] 0.6 [Formula: see text] SUVsubmax/sub. Weak-to-moderate correlations (|[Formula: see text]|lt; 0.70) were observed between PET/EPRI hypoxia images with vascular permeability (Ksuptrans/sup) or fractional extracellular-extravascular space (vsube/sub) from DCE-MRI.This is the first in vivo comparison of FMISO uptake with pOsub2/subEPRI to identify the optimal FMISO threshold to define tumor hypoxia, which may successfully direct hypoxic tumor boosts in patients, thereby enhancing tumor control.

A recent trend in PET instrumentation is the use of silicon photomultipliers (SiPMs) for high resolution and time-of-flight (TOF) detection. Due to its small size, a PET system can use a large number of SiPMs and hence effective and scalable multiplexing readout methods become important. Unfortunately, multiplexing readout generally degrades the fast timing properties necessary for TOF, especially at high channel reduction. Previously, we developed a stripline (SL)-based readout method for PET that uses a time-based multiplexing mechanism. This method maintains fast timing by design and has been successfully used for TOF PET detectors. In this article, we present a more systematic study in which we examine how two important design parameters of the readout—the number of inputs on an SL ( $n_{\mathrm {SL}}$ ) and the pathlength between adjacent input positions ( $\Delta \ell $ )—affect its detection performance properties for PET. Our result shows that up to $n_{\mathrm {SL}}=32$ the readout can achieve accurate pixel discrimination and causes little degradation in the energy resolution. The TOF resolution is compromised mildly and a coincidence resolving time on the order of 300-ps FWHM can be achieved for lutetium–yttrium oxyorthosilicate- and SiPM-based detectors. We also discuss strategies in using the readout to further reduce the number of electronic channels that a PET system would otherwise need.

We present a rotation-equivariant self-supervised learning framework for the sparse deconvolution of non-negative scalar fields on the unit sphere. Spherical signals with multiple peaks naturally arise in Diffusion MRI (dMRI), where each voxel consists of one or more signal sources corresponding to anisotropic tissue structure such as white matter. Due to spatial and spectral partial voluming, clinically-feasible dMRI struggles to resolve crossing-fiber white matter configurations, leading to extensive development in spherical deconvolution methodology to recover underlying fiber directions. However, these methods are typically linear and struggle with small crossing-angles and partial volume fraction estimation. In this work, we improve on current methodologies by nonlinearly estimating fiber structures via self-supervised spherical convolutional networks with guaranteed equivariance to spherical rotation. We perform validation via extensive single and multi-shell synthetic benchmarks demonstrating competitive performance against common baselines. We further show improved downstream performance on fiber tractography measures on the Tractometer benchmark dataset. Finally, we show downstream improvements in terms of tractography and partial volume estimation on a multi-shell dataset of human subjects.

Current deep learning approaches for diffusion MRI modeling circumvent the need for densely-sampled diffusion-weighted images (DWIs) by directly predicting microstructural indices from sparsely-sampled DWIs. However, they implicitly make unrealistic assumptions of static $q$-space sampling during training and reconstruction. Further, such approaches can restrict downstream usage of variably sampled DWIs for usages including the estimation of microstructural indices or tractography. We propose a generative adversarial translation framework for high-quality DWI synthesis with arbitrary $q$-space sampling given commonly acquired structural images (e.g., B0, T1, T2). Our translation network linearly modulates its internal representations conditioned on continuous $q$-space information, thus removing the need for fixed sampling schemes. Moreover, this approach enables downstream estimation of high-quality microstructural maps from arbitrarily subsampled DWIs, which may be particularly important in cases with sparsely sampled DWIs. Across several recent methodologies, the proposed approach yields improved DWI synthesis accuracy and fidelity with enhanced downstream utility as quantified by the accuracy of scalar microstructure indices estimated from the synthesized images. Code is available at https://github.com/mengweiren/q-space-conditioned-dwi-synthesis., Comment: Accepted by MICCAI 2021. Project page: https://heejongkim.com/dwi-synthesis; Code: https://github.com/mengweiren/q-space-conditioned-dwi-synthesis

Advancing continuous health monitoring beyond vital signs to biochemistry will revolutionize personalized medicine. Herein, we report a novel platform to achieve remote biochemical monitoring using microparticle-based biosensors and optical coherence tomography (OCT). Stimuli-responsive, polymeric microparticles were designed to serve as freely-dispersible biorecognition units, wherein binding with a target biochemical induces volumetric changes of the microparticle. Analytical approaches to detect these sub-micron changes in 3D using OCT were devised by modeling the microparticle as an optical cavity, enabling estimations far below the resolution of the OCT system. As a proof of concept, we demonstrated the 3D spatiotemporal monitoring of glucose-responsive microparticles distributed throughout a tissue-mimic in response to dynamically-fluctuating levels of glucose. Deep learning was further implemented using 3D convolutional neural networks to automate the vast processing of the continuous stream of three-dimensional time series data, resulting in a robust end-to-end pipeline with immense potential for continuous in vivo biochemical monitoring.

Previously we reported the development of a positron emission tomography (PET) and electron paramagnetic resonance (EPR) combined imaging system. The combined imaging system aimed to investigate the potential of PET hypoxia imaging by using EPR oxygen imaging as a reference. Simultaneous PET/EPR data acquisition is important to make sure of recording the same biological changes in both imaging modalities as well as to shorten imaging time. Our current system does not have magnetic or RF shielding. During the initial simultaneous PET/EPR operation, we observed that the EPR RF pulsing induced spurious noise events in the PET, and significantly affected the PET detector performance. We developed a method to reject the EPR RF noise events in off-line data processing. The method is to exploit the waveform sampling capable PET data acquisition implemented in a multi-voltage-threshold (MVT) board, and to use the pulse shape difference found in the MVT waveforms between scintillation and RF noise events. Experiments were conducted to evaluate the effectiveness of the proposed method in rejecting RF noise events. Preliminary results indicate that the rejection method works effectively to enable simultaneous data acquisition of the PET/EPR system.

Microparticle biosensors embedded in 3D tissue-mimics were spatiotemporally monitored in a completely automated pipeline using deep learning. Localization and estimation of microparticle response from OCT imaging was achieved using 3D convolutional neural networks. The automated pipeline was demonstrated using glucose-responsive microparticles as a proof-of-concept. Keywords: Optical coherence tomography, convolutional neural networks, biosensors, microparticles

PURPOSE: To enhance the spatial accuracy of fluorine 18 ((18)F) misonidazole (MISO) PET imaging of hypoxia by using dynamic contrast-enhanced (DCE) MR images as a basis for modifying PET images and by using electron paramagnetic resonance (EPR) partial oxygen pressure (pO(2)) as the reference standard. MATERIALS AND METHODS: Mice (n = 10) with leg-borne MCa4 mammary carcinomas underwent EPR imaging, T2-weighted and DCE MRI, and (18)F-MISO PET/CT. Images were registered to the same space for analysis. The thresholds of hypoxia for PET and EPR images were tumor-to-muscle ratios greater than or equal to 2.2 mm Hg and less than or equal to 14 mm Hg, respectively. The Dice similarity coefficient (DSC) and Hausdorff distance (d(H)) were used to quantify the three-dimensional overlap of hypoxia between pO(2) EPR and (18)F-MISO PET images. A training subset (n = 6) was used to calculate optimal DCE MRI weighting coefficients to relate EPR to the PET signal; the group average weights were then applied to all tumors (from six training mice and four test mice). The DSC and d(H) were calculated before and after DCE MRI–corrected PET images were obtained to quantify the improvement in overlap with EPR pO(2) images for measuring tumor hypoxia. RESULTS: The means and standard deviations of the DSC and d(H) between hypoxic regions in original PET and EPR images were 0.35 mm ± 0.23 and 5.70 mm ± 1.7, respectively, for images of all 10 mice. After implementing a preliminary DCE MRI correction to PET data, the DSC increased to 0.86 mm ± 0.18 and the d(H) decreased to 2.29 mm ± 0.70, showing significant improvement (P < .001) for images of all 10 mice. Specifically, for images of the four independent test mice, the DSC improved with correction from 0.19 ± 0.28 to 0.80 ± 0.29 (P = .02), and the d(H) improved from 6.40 mm ± 2.5 to 1.95 mm ± 0.63 (P = .01). CONCLUSION: Using EPR information as a reference standard, DCE MRI information can be used to correct (18)F-MISO PET information to more accurately reflect areas of hypoxia. Keywords: Animal Studies, Molecular Imaging, Molecular Imaging-Cancer, PET/CT, MR-Dynamic Contrast Enhanced, MR-Imaging, PET/MR, Breast, Oncology, Tumor Mircoenvironment, Electron Paramagnetic Resonance Supplemental material is available for this article. © RSNA, 2021

Building of atlases plays a crucial role in the analysis of brain images. In scenarios where early growth, aging or disease trajectories are of key importance, longitudinal atlases become necessary as references, most often created from cross-sectional data. New opportunities will be offered by creating longitudinal brain atlases from longitudinal subject-specific image data, where explicit modeling of subject’s variability in slope and intercept leads to a more robust estimation of average trajectories but also to estimates of confidence bounds. This work focuses on a framework to build a continuous 4D atlas from longitudinal high angular resolution diffusion images (HARDI) where, unlike atlases of derived scalar diffusion indices such as FA, statistics on dODFs is preserved. Multi-scalar images obtained from DW images are used for geometric alignment, and linear mixed-effects modeling from longitudinal diffusion orientation distribution functions (dODF) leads to estimation of continuous dODF changes. The proposed method is applied to a longitudinal dataset of HARDI images from healthy developing infants in the age range of 3 to 36 months. Verification of mixed-effects modeling is obtained by voxel-wise goodness of fit calculations. To demonstrate the potential of our method, we display changes of longitudinal atlas using dODF and derived generalized fractional anisotropy (GFA) of dODF. We also investigate white matter maturation patterns in genu, body, and splenium of the corpus callosum. The framework can be used to build an average dODF atlas from HARDI data and to derive subject-specific and population-based longitudinal change trajectories.

We present a combined imaging system of positron emission tomography (PET) and electron paramagnetic resonance imaging (EPRI) developed for assessing tumor hypoxia. Tumor hypoxia has been known to be resistant for cancer treatment such as radiation therapy. Therefore, precise quantitative measurement of tumor hypoxia is important for effective cancer treatment planning in radiation therapy. In clinic, PET imaging by using 18F-fluoromisonidazole (F-MISO) have been performed to delineate tumor hypoxia. However, the effectiveness of PET imaging based hypoxia targeting in radiation therapy was not yet firmly established. EPRI is a non-invasive imaging method capable of measuring the partial pressure of oxygen (pO 2 ) in tissue precisely. However, EPRI is not available for clinical use currently. The combined imaging system is to use EPRI as a gold standard to measure the pO 2 in hypoxic tumor in animal, and aims to evaluate the validity of PET imaging in tumor hypoxia by simultaneously acquired EPRI. The PET component of the combined system, which was originally developed as an insert for MRI, consists of total 14 detector modules. Each PET detector module uses of 8x4 LYSO arrays (3x3x10 mm3) coupled to Hamamatsu S13361 MPPC arrays (3.2 mm pitch). The detector modules are encased in a cylindrical plastic holder with 60 mm inner diameter and 115 mm outer diameter in a ring configuration. The axial field of view of the PET is 25.6 mm. The EPRI component of the system is built with two permanent magnets (25 mT), three gradient field coils for spatial encoding (the maximum 15 mT/m) and a RF resonator. The PET is installed to fit into the two EPR magnets (12 cm distance apart) so that the PET detector center is aligned to the magnet center. The compatibility of two imaging modalities were tested by using a 22Na point source and imaging phantoms. Animal imaging with F-MISO and FDG were also performed by running the PET/EPR simultaneously. Initial imaging results are reported.

Understanding of early brain changes has the potential to investigate imaging biomarkers for pre-symptomatic diagnosis and thus opportunity for optimal therapeutic intervention, for example in early diagnosis of infants at risk to autism or altered development of infants to drug exposure. In this paper, we propose a framework to analyze longitudinal changes of structural connectivity in the early developing infant brain by exploring underlying network components of brain structural connectivity and its changes with age. Structural connectivity is a non-negative sparse network. Projective non-negative matrix factorization (PNMF) offers benefits in sparsity and learning fewer parameters for non-negative sparse data. The number of matrix subcomponents was estimated by automatic relevance determination PNMF (ARDPNMF) for brain connectivity networks for the given data. We apply linear mixed effect modeling on the resulting loadings from ARDPNMF to model longitudinal network component changes over time. The proposed framework was validated on a synthetic example generated by known linear mixed effects on loadings of the known number of bases with different levels of additive noises. Feasibility of the framework on real data has been demonstrated by analysis of structural connectivity networks of high angular resonance diffusion imaging (HARDI) data from an ongoing neuroimaging study of autism. A total of 139 image data sets from high-risk and low-risk subjects acquired at multiple time points have been processed. Results demonstrate the feasibility of the framework to analyze connectivity network properties as a function of age and the potential to eventually explore differences associated with risk status.

We report the development of a compact, highly modular SiPM-based PET detector that outputs digital data via a standard serial network interface and the use of such detectors in conjunction with a software platform to obtain a scalable system architecture for PET. The detector, approximately 2.5 cm wide and 10 cm long, is made of 4 LYSO/SiPM arrays arranged in a row. Each LYSO/SiPM array is consisted of a 13x13 LYSO array optically coupled to a 6x6 array of SensL MicroFC-30035 SiPMs. For the LYSO array, the crystal size is 1.89x1.89x13 mm3 and the pitch is 2.0 mm. For the SiPM array, the pixel size and pitch are 3.2 mm and 4.2 mm, respectively. The intrinsic resolution, energy resolution and coincidence time resolution of the detector are measured to be about 1.91 mm, 15% and 1.4 ns, respectively. Crystals in a detector exhibit rather uniform performance characteristics. Two demonstration systems are developed using the developed detectors and proposed architecture, including a single-ring system (~10.4 cm diameter, 10 cm length) using 12 detectors and a two-ring system (~19 cm diameter, 20 cm length) using 48 detectors. Imaging data are successfully obtained by using these systems and the results indicate that approximately 1 mm image resolution is achieved.

Precise quantitative delineation of tumor hypoxia is essential in radiation therapy treatment planning to improve the treatment efficacy by targeting hypoxic sub-volumes. We developed a combined imaging system of positron emission tomography (PET) and electron para-magnetic resonance imaging (EPRI) of molecular oxygen to investigate the accuracy of PET imaging in assessing tumor hypoxia. The PET/EPRI combined imaging system aims to use EPRI to precisely measure the oxygen partial pressure in tissues. This will evaluate the validity of PET hypoxic tumor imaging by (near) simultaneously acquired EPRI as ground truth. The combined imaging system was constructed by integrating a small animal PET scanner (inner ring diameter 62 mm and axial field of view 25.6 mm) and an EPRI subsystem (field strength 25 mT and resonant frequency 700 MHz). The compatibility between the PET and EPRI subsystems were tested with both phantom and animal imaging. Hypoxic imaging on a tumor mouse model using $^{18}$F-fluoromisonidazole radio-tracer was conducted with the developed PET/EPRI system. We report the development and initial imaging results obtained from the PET/EPRI combined imaging system., Comment: 12 pages, 6 figures

This study serves as a proof of concept for multimodal oxygen imaging in developing an algorithm to make fluorine 18 misonidazole PET more comparable with quantitative electron paramagnetic resonan...

In July 2014, a nematode species, Schistorophus cirripedesmi Rhizhikov and Khokhlova, 1964, was recovered from a bar-tailed godwit, Limosa lapponica baueri that was stored in a -20˚C freezer in the Chungnam Wild Animal Rescue Center. The bird was collected in 2012 from the coastal region of Pyeongtaek-si (City), Gyeonggi-do (Province) in the Republic of Korea, although the exact date is not clear. At necropsy, 9 nematodes were found in the gizzard of the bird. The parasites had 4 horn-like cephalic cuticular ornamentations. After morphometric comparison and morphological observations, including scanning electron microscopy, the nematodes were identified as S. cirripedesmi. This is the first description of a nematode species in a shorebird in Korea. This is also the first time this genus and species have been found in Korea.

The majority of parasitological studies of terrestrial snakes in Korea have focused on zoonotic parasites. However, in the present study, we describe 3 unrecorded nematode species recovered from 5 species of snakes (n=6) in Korea. The examined snakes, all confiscated from illegal hunters, were donated by the Chungnam Wild Animal Rescue Center and Korean Broadcasting System in July 2014 and February 2015. Light and scanning electron microscopies on the shapes of spicules that are either bent or straight (kalicephalids) and the presence of the intestinal cecum (ophidascarids) figured out 3 nematodes; Kalicephalus brachycephalus Maplestone, 1931, Kalicephalus sinensis Hsü, 1934, and Ophidascaris excavata Hsü and Hoeppli, 1934. These 3 species of nematode faunas are recorded for the first time in Korea.

The white matter tract-specific correlates of neuropsychological deficits are not fully established in patients with subcortical vascular cognitive impairment (SVCI), where white matter tract damage may be a critical factor in cognitive impairment. The purpose of this study is to investigate the tract-specific correlates of neuropsychological deficits in SVCI patients using tract-specific statistical analysis (TSSA). We prospectively recruited 114 SVCI patients, and 55 age-, gender-, and education-matched individuals with normal cognition (NC). All participants underwent diffusion weighted imaging and neuropsychological testing. We classified tractography results into fourteen major fiber tracts and analyzed group comparison and correlation with cognitive impairments. Relative to NC subjects, SVCI patients showed decreased fractional anisotropy values in bilateral anterior-thalamic radiation, cingulum, superior-longitudinal fasciculus, uncinate fasciculus, corticospinal tract, and left inferior-longitudinal fasciculus. Focal disruptions in specific tracts were associated with specific cognitive impairments. Our findings suggest that disconnection of specific white matter tracts, especially those neighboring and providing connections between gray matter regions important to certain cognitive functions, may contribute to specific cognitive impairments in SVCI.

Recently, we have developed a stripline (SL) readout method and amulti-voltage threshold (MVT) digitizer for PET. The SL readout is amultiplexing method for reading multiple input signals that is easy toimplement and can support timeof-flight (TOF) detection. The MVT methodsdigitally samples a PET signal pulse with respect to a number of voltagethresholds and can be implemented by using the field programmable gate array(FPGA). The use of these two methods can greatly simplify the electroniccomplexity of a TOF PET detector. In this paper, we investigate theperformance properties of the SL readout when applied to different numbersof silicon photomultipliers (SiPM) from SensL. Outputs of the SL readout areacquired by using a 72-ch FPGA MVT board. Our results show that up to 36pixels per SL can be discriminated, a pixel energy resolution of 11-21%can be obtained, and a coincidence resolving time (CRT) of 520 ps can beachieved. Previously, we obtained 320ps CRT when applied the MVT to a singlepixel made of LYSO and SiPM. Therefore, we believe the CRT can be furtherimproved after optimization.

We are developing a MR insert PET detector using SiPM for small animal imaging. Our detector design is based on the strip-line signal readout that we have developed to efficiently handle large amount signals for SiPM-based TOF PET. The stripline readout has an advantageous feature suitable for building PET integrated within MR. In strip-line readout, the signal digitization is accomplished remotely from the detection module, thereby the front-end electronics of the detector within MR is minimal; the interference between two systems, e.g., magnetic field distortion due to PET, can be minimized accordingly. The feasibility of the strip-line readout for PET/MR has been demonstrated in our previous study. The current prototype MR insert PET is comprised of 8 detector modules, which are installed in a plastic cylindrical supporting structure. Each detector module consists of 8x4 LYSO arrays (3x3x10 mm3each) and two Hamamatsu S13361–3050NE-04 MPPC arrays (4x4, 3.2 mm pitch); 32 SiPM signals are routed in 4 strip-lines so that 8 LYSO/SiPM signals in a row are connected to each strip-line. The signals from strip-lines are fed to a 72-ch multi-voltage threshold (MVT) digitizer, which digitizes the input by TDCs implemented in FPGA. Experiments have been carried out to evaluate the detector performance, and image reconstruction also has been performed for a simple shape object. In the paper, the detail on the development of the MR insert PET are explained, and preliminary test results are presented.

Signal multiplexing for SiPM-based TOF PET to achieve high multiplexing ratio and keep the fast timing performance is a challenging issue. We have developed a multiplexing method using a strip-line; multiple SiPMs are connected to a single micro-strip transmission line, and the position of hit SiPM is determined by the signal arrival time difference on the strip-line. The strip-line method is scalable; high multiplexing ratio can be easily achieved by increasing the number of SiPMs connected on a strip-line. To test the scalable feature of the strip-line readout systematically, we have implemented several strip-lines with different trace lengths and varied the number of SiPMs (Hamamatsu S13361-3050NE) on the strip-lines up to 32. Experimental tests have been conducted to assess the influences of the strip-lines on detector performance: position, energy, and coincidence time resolution. Preliminary results from the study are presented, and our ongoing efforts on strip-line readout for TOF PET are discussed.

The erythroid differentiation regulator 1 (Erdr1), which is a novel and highly conserved factor, was recently reported to be negatively regulated by IL-18 and to play a crucial role as an antimetastatic factor. IL-18 is a proinflammatory cytokine that functions as an angiogenic mediator in inflammation. Rosacea is a chronic inflammatory skin disorder that is characterized by abnormal inflammation and vascular hyperactivity of the facial skin. To determine whether Erdr1 contributes to the regulation of the chronic inflammatory process in the development of rosacea, an immunohistochemical analysis was performed in healthy donors and patients with rosacea. In this study, we showed that Erdr1 was downregulated, whereas IL-18 was upregulated, in patients with rosacea, which led us to question the role of Erdr1 in this disorder. Moreover, a rosacea-like BALB/c mouse model was used to determine the role of Erdr1 in rosacea in vivo. LL-37 injection induced typical rosacea features, including erythema, telangiectasia and inflammation. Treatment with recombinant Erdr1 (rErdr1) resulted in a significant reduction of erythema, inflammatory cell infiltration (including CD4(+) and CD8(+) T cells), and microvessel density with vascular endothelial growth factor (VEGF). Taken together, our findings suggest that rErdr1 may be involved in attenuating the inflammation and angiogenesis associated with the pathogenesis of rosacea. Thus, these results provide new insight into the mechanism involved in this condition and indicate that rErdr1 could be a potential target for therapeutic intervention of rosacea.

The Trans - PET ® BioCaliburn ® SH system, recently introduced by the Raycan Technology Co. Ltd. (Suzhou, China), is a commercial positron emission tomography (PET) system designed for rodent imaging. The system contains 6 basic detector modules (BDMs) arranged on a 10.8 cm diameter ring to provide a transaxial field of view (FOV) of 6.5 cm and an axial FOV of 5.3 cm. In this paper, we report on its performance properties in accordance with the National Electrical Manufacturers Association (NEMA) 2008 NU-4 standards with modifications. The measured spatial resolution at the center of the FOV was 1.05 mm, 1.12 mm and 1.13 mm in the tangential, radial and axial directions, respectively. The measured system sensitivity was 3.29% for a point source at the center of the FOV when using a 350–650 keV energy window and a 5 ns coincidence time window. When a wider 250–750 keV energy window was used, it increased to 4.21%. For mouse- and rat-sized phantoms, the scatter fraction was 10.7% and 16.1%, respectively. The peak noise equivalent count rate were 36 kcps@8.52 MBq for the mouse-sized phantom and 16 kcps@6.79 MBq for the rat-sized phantom. The Derenzo phantom image showed that the system can resolve 1.0 mm diameter rods. The measured performance properties of the system indicate that the Trans - PET ® BioCaliburn ® SH is a versatile imaging device that can provide high spatial resolution for rodent imaging while offering competitive sensitivity and count-rate performance.

We report the development of a modularized compact positron emission tomography (PET) detector that outputs serial streams of digital samples of PET event pulses via an Ethernet interface using the UDP/IP protocol to enable rapid configuration of a PET system by connecting multiple such detectors via a network switch to a computer. Presently, the detector is [Formula: see text] in extent (excluding I/O connectors) and contains an [Formula: see text] array of [Formula: see text] one-to-one coupled lutetium-yttrium oxyorthosilicate/silicon photomultiplier pixels. It employs cross-wire and stripline readouts to merge the outputs of the 216 detector pixels to 24 channels. Signals at these channels are sampled using a built-in 24-ch, 4-level field programmable gate arrays-only multivoltage threshold digitizer. In the computer, software programs are implemented to analyze the digital samples to extract event information and to perform energy qualification and coincidence filtering. We have developed two such detectors. We show that all their pixels can be accurately discriminated and measure a crystal-level energy resolution of 14.4% to 19.4% and a detector-level coincidence time resolution of 1.67 ns FWHM. Preliminary imaging results suggests that a PET system based on the detectors can achieve an image resolution of [Formula: see text].

We have developed a signal multiplexing method for a SiPM-based time-of-flight (TOF) positron emission tomography (PET). In the method, multiple SiPM signals are connected to a single strip-line, and the position of hit SiPM is decoded from the arrival time difference measured at the two ends of a strip-line. Since the position resolution on strip-line is directly proportional to the accuracy of the measured time, a high speed (≥ 1 GSPS) waveform sampling electronics, e.g., Domino Ring Sampler (DRS4), has been used to determine the time stamp of the strip-line signal as precise as possible: e.g., the electronic time resolution of the DRS4 is measured to be 5 ps. Although the DRS4 waveform sampling provides a good timing accuracy, its large latency time required for digitizing samples could be a limiting factor for high count rate operation. Therefore, in an effort to increase the counting performance, we have come up with a new design of the strip-line board that can work with commonly used time-to-digital converters (TDC) with 50-100 ps electronic time resolution, which provide high counting capability. Hamamatsu MPPC 4×4 arrays (S12642-0404PA-50, 3×3 mm2 area for each) are employed in the new strip-line board; 32 MPPC signals are routed to 4 strip-lines on the board. For verification purpose, we have conducted experimental tests to readout the current strip-line board by using the OpenPET electronics, which is a general purpose data acquisition developed for nuclear medicine research community. In the test, LYSO scintillator arrays (3×3×20 mm3 each) are coupled to MPPCs on the strip-line board, and the detector is irradiated by a 22Na source with 10 μCi activity. The initial results obtained by using OpenPET are very promising: each MPPC on the strip-line is well identified with 1.2 mm FWHM resolution, and 13.0% FWHM of energy resolution for 511 keV gamma. The results are comparable to the ones by using DRS4 waveform sampler running at 2.5 GSPS.

In this work we demonstrate, for the first time, the feasibility of making rapid fluorescence lifetime measurements with analog silicon photomultipliers (SiPMs) and waveform sampling techniques. Fluorophore excitation was provided by means of a low-power fiber-coupled pulsed laser diode operating at 405nm, whose beam had been expanded to a diameter of 2mm before impinging upon the sample. A portion of the sample's isotropic fluorescent emission was collected orthogonal to the incident excitation beam via an analog SiPM and analyzed by waveform sampling at 2.5 GS/s. Fluorophores for system evaluation were selected such that their fluorescence lifetimes differed significantly from one another but their excitation and emission wavelength's maxima did not, and included Coumarin 6, Coumarin 314, and Coumarin 343, each dissolved in ethanol. Measurements were made of a non-fluorescent, isotropically-scattering specimen to derive the system's response function. Single exponential fitting of the convolution of the system's response function with each dye's intrinsic fluorescence intensity function yielded fluorescence lifetimes in close agreement to those derived from measurements made on a commercial system.

We are developing a time-of-flight (TOF) positron emission tomography (PET) scanner for brain imaging using SiPMs. In our development, we have adopted a strip-line based signal multiplexing method for efficient channel reduction for SiPMs; the method is using a single strip-line on which multiple SiPMs are connected for sharing readout, and the position of hit SiPM is decoded from the arrival time difference measured at the two ends of the strip-line. We have conducted experimental tests for evaluating the performance of the detectors based on the strip-line signal readout. Hamamatsu MPPC array (S12642-0404PA-50, 3.2 mm pitch) is employed in the detector module; 32 MPPC signals from two MPPCs arrays are routed to 4 strip-lines on the current strip-line board. For the tests, two detectors are used in a coincidence setup, and an array of 4×4 LYSO scintillator (3×3×20 mm3, 3.2 mm pitch) is coupled to the MPPC array in each detector. A 22Na source with 3 μCi activity is used, and the waveform of the detectors signals are collected by using a Domino Ring Sampler (DRS4) evaluation board running at 2.5 giga sampling rate. The preliminary results show that the position on the strip-line is determined with ∼1.0 mm precision from the measured differential time on the strip-line. The average energy and coincidence time resolution are measured to be ∼13% FWHM and ∼726 ps FWHM, respectively, at 511 keV photo-peak. A systematic investigation of the effects of strip-line on detector performance is carried out in parallel by using a specially designed strip-line board. The test results is presented in the paper.

Over the last decade, marine debris has become a major factor affecting the coastal ecosystem of Korea. This study compiled information regarding how marine debris impacts wildlife in Korea. Cases of marine debris impacting wildlife were collected from experts of various fields and from local participants through an open access website from February 2010 to March 2012. A total of 21 species were affected by marine debris: 18 species of birds, 2 species of mammals, and 1 species of crustacean. Five threatened or protected species were identified: black-faced spoonbill, finless porpoise, water deer, whooper swan, and greater painted snipe. Recreational fishing gears were the types of debris that most frequently impacted wildlife, especially birds. Black tailed gulls were the most vulnerable species to recreational fishing hooks and lines. Although it was preliminary, this study revealed that recreational fishing activities should be prioritized when managing marine debris in Korea.