Automated reference-free detection of motion artifacts in magnetic resonance images

Objectives

Our objectives were to provide an automated method for spatially resolved detection and quantification of motion artifacts in MR images of the head and abdomen as well as a quality control of the trained architecture.

Materials and methods

T1-weighted MR images of the head and the upper abdomen were acquired in 16 healthy volunteers under rest and under motion. Images were divided into overlapping patches of different sizes achieving spatial separation. Using these patches as input data, a convolutional neural network (CNN) was trained to derive probability maps for the presence of motion artifacts. A deep visualization offers a human-interpretable quality control of the trained CNN. Results were visually assessed on probability maps and as classification accuracy on a per-patch, per-slice and per-volunteer basis.

Results

On visual assessment, a clear difference of probability maps was observed between data sets with and without motion. The overall accuracy of motion detection on a per-patch/per-volunteer basis reached 97%/100% in the head and 75%/100% in the abdomen, respectively.

Conclusion

Automated detection of motion artifacts in MRI is feasible with good accuracy in the head and abdomen. The proposed method provides quantification and localization of artifacts as well as a visualization of the learned content. It may be extended to other anatomic areas and used for quality assurance of MR images.

     

Nanoporous aluminum oxide membranes for filtration and biofunctionalization

Nanoporous aluminum oxide membranes with high open porosity are prepared by anodic oxidation. Conventional self-supporting as well as mechanically stabilized nanoporous membranes are produced from aluminum plates and microimprinted aluminum foils, respectively. The mechanically stabilized membranes are characterized by very thin membrane parts stabilized by surrounding thick bridges. The minimal thickness of these thin membranes with open pores on both sides is 1 microm, with a mean pore size of the parallel open pores of 185 nm. With these two kinds of membrane the flow rates for cross filtration can be tuned over a wide range. With the mechanically stabilized membranes, substantially higher flow rates are achieved and experiments that cannot be performed with thicker membranes become possible. The biofunctionalization of the pore walls with archaebacterial tetraether lipids is realized and proved using aminated semiconductor nanocrystals. The lipid layer deposited on the pore walls also changes the filtration properties.     

Rheological properties of aqueous α A12O3 suspensions: Influence of dispersant concentration

The effects of concentration of polyacrylic acid as a dispersant on rheological properties of aqueous alumina suspensions have been investigated under steady and oscillatory shear conditions. At solid volume fractions between 0.45 and 0.6, a high degree of particle stabilization is achieved when 0.2 wt% of polyacrylic acid is added. At lower dispersant concentrations, suspensions exhibit pronounced irreversible thixotropic behaviour, whereas at higher dispersant concentrations, time dependent effects on the flow properties are not detectable. When the saturation adsorption limit of the polyelectrolyte on Al₂O₃ is reached, further addition of the dispersant appreciably changes the flow behaviour, as well as the viscoelastic response of investigated suspensions. The data under steady shear are described by application of the generalized Casson model, and for the analysis of viscoelastic data the generalized Maxwell model is used.     

Multi-length scale modeling of martensitic transformations in stainless steels

In the present work a multi-length scale model is developed to study both the athermal and stress-assisted martensitic transformations in a single crystal of 301 type stainless steel. The microstructure evolution is simulated using elastoplastic phase-field simulations in three dimensions. The input data for the simulations is acquired from a combination of computational techniques and experimental works. The driving force for the transformation is calculated by using the CALPHAD technique and the elastic constants of the body-centered cubic phase are calculated by using ab initio method. The other input data is acquired from experimental works. The simulated microstructures resemble a lath-type martensitic microstructure, which is in good agreement with the experimental results obtained for a stainless steel of similar composition. The martensite habit plane predicted by the model is in accordance with experimental results. The Magee effect, i.e. formation of favorable martensite variants depending on the loading conditions, is observed in the simulations. The results also indicate that anisotropic loading conditions give rise to a significant anisotropy in the martensitic microstructure.     

Size-dependent multispectral photoacoustic response of solid and hollow gold nanoparticles

Photoacoustic (PA) imaging attracts a great deal of attention as an innovative modality for longitudinal, non-invasive, functional and molecular imaging in oncology. Gold nanoparticles (AuNPs) are identified as superior, NIR-absorbing PA contrast agents for biomedical applications. Until now, no systematic comparison of the optical extinction and PA efficiency of water-soluble AuNPs of various geometries and small sizes has been performed.Here spherical AuNPs with core diameters of 1.0, 1.4 and 11.2?nm, nanorods with longitudinal/transversal elongation of 38/9 and 44/12?nm and hollow nanospheres with outer/inner diameters of 33/19, 57/30, 68/45 and 85/56?nm were synthesized. The diode laser set-up with excitations at 650, 808, 850 and 905?nm allowed us to correlate the molar PA signal intensity with the molar extinction of the respective AuNPs. Deviations were explained by differences in heat transfer from the particle to the medium and, for larger particles, by the scattering of light. The molar PA intensity of 1.0?nm AuNPs was comparable to the commonly used organic dye methylene blue, and rapidly increased with the lateral size of AuNPs.     

An agent-based concept for planning and control of the production of carbon fibre reinforced plastics aircraft structures with mobile production units

This article is concerned with the control of the production of aircraft primary structures made of carbon fibre reinforced plastics. To increase the parallelization of the production, a scenario that implies new design paradigms is discussed. In order to realize this parallelization, mobile production entities work concurrently at the same primary structure of the fuselage. The present article proposes a software architecture for the control of such a production system. It is proposed that components are organized hierarchically. In enhancement to previous proposals, two perspectives of hierarchy are used here: grouping by functionality and grouping by timing context. The core of the architecture is a market-based multi-agent system, where the agents may operate in multiple timing contexts. Such a design yields advantages in terms of fast integration of new functionalities and scale-up of the production system. In brief, this article introduces a top-level control architecture for parallelized production of large reliable CFRP structures in a scalable production system.     

Realization of minute-long steady-state H-mode discharges on EAST

In the 2016 EAST experimental campaign,a steady-state long-pulse H-mode discharge with an ITER-like tungsten divertor lasting longer than one minute has been obtained using only RF heating and current drive,through an integrated control of the wall conditioning,plasma configuration,divertor heat flux,particle exhaust,impurity management,and effective coupling of multiple RF heating and current drive sources at high injected power.The plasma current (Ip ～ 0.45 MA) was fully-noninductively driven (Vloop ＜ 0.0 V) by a combination of ～2.5 MW LHW,～0.4 MW ECH and ～0.8 MW ICRF.This result demonstrates the progress of physics and technology studies on EAST,and will benefit the physics basis for steady state operation of ITER and CFETR.     

C-Curves for Lengthening of Widmanstätten and Bainitic Ferrite

Widmanstätten ferrite and bainitic ferrite are both acicular and their lengthening rate in binary Fe-C alloys and low-alloyed steels under isothermal conditions is studied by searching the literature and through new measurements. As a function of temperature, the lengthening rate can be represented by a common curve for both kinds of acicular ferrite in contrast to the separate C-curves often presented in time-temperature-transformation (TTT) diagrams. The curves for Fe-C alloys with low carbon content show no obvious decrease in rate at low temperatures down to 623 K (350 °C). For alloys with higher carbon content, the expected decrease of rate as a function of temperature below a nose was observed. An attempt to explain the absence of a nose for low carbon contents by an increasing deviation from local equilibrium at high growth rates is presented. This explanation is based on a simple kinetic model, which predicts that the growth rates for Fe-C alloys with less than 0.3 mass pct carbon are high enough at low temperatures to make the carbon pileup, in front of the advancing tip of a ferrite plate, shrink below atomic dimensions, starting at about 600 K (323 °C).