The fabrication process consists of salt leaching and heat press moulding. In this method, the salt particles are utilized as porogen materials. A mixture of the biodegradable polymer and the salt was shaped into microneedles by moulding.
This form factor has the potential to make a better replacement to glucose monitoring products (commonly used by people with diabetes), so they can get a continuous readout of their glucose levels.
Rule-based methods are often used for assigning fiber orientation to cardiacanatomical models. However, existing methods have been developed using datamostly from the left ventricle. As a consequence, fiber information obtainedfrom rule-based methods often does not match histological data in other areasof the heart such as the right ventricle, having a negative impact in cardiacsimulations beyond the left ventricle. In this work, we present a rule-basedmethod where fiber orientation is separately modeled in each ventriclefollowing observations from histology. This allows to create detailed fiberorientation in specific regions such as the endocardium of the right ventricle,the interventricular septum and the outflow tracts. We also carried outelectrophysiological simulations involving these structures and with differentfiber configurations. In particular, we built a modelling pipeline for creatingpatient-specific volumetric meshes of biventricular geometries, including theoutflow tracts, and subsequently simulate the electrical wavefront propagationin outflow tract ventricular arrhythmias with different origins for the ectopicfocus. The resulting simulations with the proposed rule-based method showed avery good agreement with clinical parameters such as the 10 ms isochrone ratioin a cohort of nine patients suffering from this type of arrhythmia. Thedeveloped modelling pipeline confirms its potential for an in silicoidentification of the site of origin in outflow tract ventricular arrhythmiasbefore clinical intervention.
The aim of this study was to investigate the impact of decay data provided bythe newly developed stochastic atomic relaxation model BrIccEmis on dose pointkernels (DPKs - radial dose distribution around a unit point source) andS-values (absorbed dose per unit cumulated activity) of 14 Auger electron (AE)emitting radionuclides, namely 67Ga, 80mBr, 89Zr, 90Nb, 99mTc, 111In, 117mSn,119Sb, 123I, 124I, 125I, 135La, 195mPt and 201Tl. Radiation spectra were basedon the nuclear decay data from the medical internal radiation dose (MIRD)RADTABS program and the BrIccEmis code, assuming both an isolated-atom andcondensed-phase approach. DPKs were simulated with the PENELOPE Monte Carlo(MC) code using event-by-event electron and photon transport. S-values forconcentric spherical cells of various sizes were derived from these DPKS usingappropriate geometric reduction factors. The number of Auger and Coster-Kronig(CK) electrons and x-ray photons released per nuclear decay (yield) fromMIRD-RADTABS were consistently higher than those calculated using BrIccEmis.DPKs for the electron spectra from BrIccEmis were considerably different fromMIRD-RADTABS in the first few hundred nanometres from a point source where mostof the Auger electrons are stopped. S-values were, however, not significantlyimpacted as the differences in DPKS in the sub-micrometre dimension werequickly diminished in larger dimensions. Overestimation in the total AE energyoutput by MIRD-RADTABS leads to higher predicted energy deposition by AEemitting radionuclides, especially in the immediate vicinity of the decayingradionuclides. This should be taken into account when MIRD-RADTABS data areused to simulate biological damage at nanoscale dimensions.
To benchmark a Monte Carlo model of the Auger cascade that has been developedat the Australian National University (ANU) against the literature data. Themodel is applicable to any Auger-electron emitting radionuclide with nuclearstructure data in the format of the Evaluated Nuclear Structure Data File(ENSDF). Sch\"onfeld's algorithms and the BrIcc code were incorporated toobtain initial vacancy distributions due to electron capture (EC) and internalconversion (IC), respectively. Atomic transition probabilities were adoptedfrom the Evaluated Atomic Data Library (EADL) for elements with atomic number,Z = 1-100. Atomic transition energies were evaluated using a relativisticDirac-Fock method. An energy-restriction protocol was implemented to eliminateenergetically forbidden transitions from the simulations. Calculated initialvacancy distributions and average energy spectra of 123I, 124I and 125I werecompared with the literature data. In addition, simulated kinetic energyspectra and frequency distributions of the number of emitted electrons andphotons of the three iodine radionuclides are presented. Some examples ofradiation spectra of individual decays are also given. Good agreement with thepublished data was achieved except for the outer-shell Auger and Coster-Kronigtransitions. Nevertheless, the model needs to be compared with experimentaldata in a future study.
Functional behavior of breast cancer - representing underlying biology - canbe analyzed using MRI. The most widely used breast MR imaging protocol isdynamic contrast-enhanced T1-weighted imaging. The cancer enhances on dynamiccontrast-enhanced MR imaging because the contrast agent leaks from the leakyvessels into the interstitial space. The contrast agent subsequently leaks backinto the vascular space, creating a washout effect. The normal parenchymaltissue of the breast can also enhance after contrast injection. Thisenhancement generally increases over time. Typically, a radiologist assessesthis background parenchymal enhancement (BPE) using the Breast ImagingReporting and Data System (BI-RADS). According to the BI-RADS, BPE refers tothe volume of enhancement and the intensity of enhancement and is divided infour incremental categories: minimal, mild, moderate, and marked. Researchers have developed semi-automatic and automatic methods to extractproperties of BPE from MR images. For clarity, in this syllabus the BI-RADSdefinition will be referred to as BPE, whereas the computer-extractedproperties will not. Both BPE and computer-extracted parenchymal enhancementproperties have been linked to screening and diagnosis, hormone status and age,risk of development of breast cancer, response monitoring, and prognosis.
Magnetic resonance imaging is capable of producing volumetric images withoutionizing radiation. Nonetheless, long acquisitions lead to prohibitively longexams. Compressed sensing (CS) can enable faster scanning via sub-sampling withreduced artifacts. However, CS requires significantly higher reconstructioncomputation, limiting current clinical applications to 2D/3D orlimited-resolution dynamic imaging. Here we analyze the practical limitationsto T2 Shuffling, a four-dimensional CS-based acquisition, which provides sharp3D-isotropic-resolution and multi-contrast images in a single scan. Ourimprovements to the pipeline on a single machine provide a 3x overallreconstruction speedup, which allowed us to add algorithmic changes improvingimage quality. Using four machines, we achieved additional 2.1x improvementthrough distributed parallelization. Our solution reduced the reconstructiontime in the hospital to 90 seconds on a 4-node cluster, enabling its useclinically. To understand the implications of scaling this application, wesimulated running our reconstructions with a multiple scanner setup typical inhospitals.
Ning Meng, Peng Zhang, Junfeng Li, Jun He, Jin Zhu
Published: Sep 2018
Background --The objective of this study was to examine the association ofroutine blood test results with coronary heart disease (CHD) risk, toincorporate them into coronary prediction models and to compare thediscrimination properties of this approach with other prediction functions.Methods and Results --This work was designed as a retrospective, single-centerstudy of a hospital-based cohort. The 5060 CHD patients (2365 men and 2695women) were 1 to 97 years old at baseline with 8 years (2009-2017) of medicalrecords, 5051 health check-ups and 5075 cases of other diseases. We developed atwo-layer Gradient Boosting Decision Tree(GBDT) model based on routine blooddata to predict the risk of coronary heart disease, which could identify 86% ofpeople with coronary heart disease. We built a dataset with 15,000 routineblood tests results. Using this dataset, we trained the two-layer GBDT model toclassify healthy status, coronary heart disease and other diseases. As a resultof the classification after machine learning, we found that the sensitivity ofdetecting the health data was approximately 93% for all data, and thesensitivity of detecting CHD was 93% for disease data that included coronaryheart disease. On this basis, we further visualized the correlation betweenroutine blood results and related data items, and there was an obvious patternin health and coronary heart disease in all data presentations, which can beused for clinical reference. Finally, we briefly analyzed the results abovefrom the perspective of pathophysiology. Conclusions --Routine blood dataprovides more information about CHD than what we already know through thecorrelation between test results and related data items. A simple coronarydisease prediction model was developed using a GBDT algorithm, which will allowphysicians to predict CHD risk in patients without overt CHD.
Purpose: The presence of respiratory motion during radiation treatment leadsto degradation of the expected dose distribution, both for target coverage andhealthy-tissue sparing, particularly for techniques like pencil-beam scanningproton therapy which have dynamic delivery systems. While tools exist toestimate this degraded four-dimensional (4D) dose, they typically have one ormore deficiencies such as ... Methods: To quickly compute the 4D-dose, the three main tasks of thecalculator were run on graphics processing units (GPUs). These tasks were:simulating the delivery of the plan using measured delivery parameters todistribute the plan amongst 4DCT phases characterizing the patient breathing,using an in-house Monte Carlo simulation (MC) dose calculator to determine thedose delivered to each breathing phase, and accumulating the doses from thevarious breathing phases onto a single phase for evaluation. The accumulationwas performed by individually transferring the energy and mass of dose-gridsubvoxels, a technique models the transfer of dose in a more physicallyrealistic manner. The calculator was run ... Results: 4D doses were successfully computed for the three test cases withcomputation times ranging from 4-6 min on a server with eight NVIDIA Titan Xgraphics cards; the most time-consuming component was the MC dose engine. Thesubvoxel-based dose-accumulation technique produced stable 4D-dosedistributions at subvoxel scales of 0.5-1.0 mm without impairing the totalcomputation time. The uncertainties in the beam-delivery simulation ... Conclusions: A MC-based and GPU-accelerated 4D-dose calculator was developedto estimate the effects of respiratory motion on pencil-beam scanning protontherapy treatments. The calculator can currently be used ...
Imaging neuronal activity non-invasively in vivo is of tremendous interest,but current imaging techniques lack either functional contrast or necessarymicroscopic resolution. The retina is the only part of the central nervoussystem (CNS) that allows us direct optical access. Not only ophthalmicdiseases, but also many degenerative disorders of the CNS go along withpathological changes in the retina. Consequently, functional analysis ofretinal neurons could lead to an earlier and better diagnosis and understandingof those diseases. Recently, we showed that an activation of photoreceptorcells could be visualized in humans using a phase sensitive evaluation ofoptical coherence tomography data. The optical path length of the outersegments changes by a few hundred nanometers in response to opticalstimulation. Here, we show simultaneous imaging of the activation ofphotoreceptor and ganglion cells. The signals from the ganglion cells areten-fold smaller than those from the photoreceptor cells and were only visibleusing new algorithms for suppressing motion artifacts. This allowed us togenerate a wiring diagram showing functional connections between photoreceptorsand ganglion cells. We present a theoretical model that explains the observedintrinsic optical signals by osmotic volume changes, induced by ion influx orefflux. Since all neuronal activity is associated with ion fluxes, imagingosmotic induced size changes with nanometer precision should visualizeactivation in any neuron.
We present signature for planetary correlations following an analysis ofmonthly melanoma rates in USA for the period 1973 - 2011. A planetaryrelationship in medicine is observed for the first time. The statisticalsignificance is well above 5 sigmas, while various crosschecking makesystematics highly improbable as the cause. The observed planetary dependencein physics was suggestive for this investigation. Streaming invisible matterfrom the dark sector, whose flux can be occasionally enhanced towards the Earthvia planetary gravitational focusing, and, even much stronger by the Sun, itmay be the explanation for 1-10% of melanoma diagnoses. The derived shortestmelanoma periodicity of about 87.5 days points in its own right at a shortlatency period of about few months. Contrariwise, the present findingsstrengthen the previous physics claim of streams of invisible matter.