Fast maximum intensity projections of large medical data sets by exploiting hierarchical memory architectures.

Maximum intensity projections (MIPs) are an important visualization technique for angiographic data sets. Efficient data inspection requires frame rates of at least five frames per second at preserved image quality. Despite the advances in computer technology, this task remains a challenge. On the one hand, the sizes of computed tomography and magnetic resonance images are increasing rapidly. On the other hand, rendering algorithms do not automatically benefit from the advances in processor technology, especially for large data sets. This is due to the faster evolving processing power and the slower evolving memory access speed, which is bridged by hierarchical cache memory architectures. In this paper, we investigate memory access optimization methods and use them for generating MIPs on general-purpose central processing units (CPUs) and graphics processing units (GPUs), respectively. These methods can work on any level of the memory hierarchy, and we show that properly combined methods can optimize memory access on multiple levels of the hierarchy at the same time. We present performance measurements to compare different algorithm variants and illustrate the influence of the respective techniques. On current hardware, the efficient handling of the memory hierarchy for CPUs improves the rendering performance by a factor of 3 to 4. On GPUs, we observed that the effect is even larger, especially for large data sets. The methods can easily be adjusted to different hardware specifics, although their impact can vary considerably. They can also be used for other rendering techniques than MIPs, and their use for more general image processing task could be investigated in the future.     

Mixed Mode VLSI Implementation of a Neural Associative Memory

A mixed mode digital/analog special purpose VLSI hardware implementation of an associative memory with neural architecture is presented. The memory concept is based on a matrix architecture with binary storage elements holding the connection weights. To enhance the processing speed analog circuit techniques are applied to implement the algorithm for the association. To keep the memory density as high as possible two design strategies are considered. First, the number of transistors per storage element is kept to a minimum. In this paper a circuit technique that uses a single 6-transistor cell for weight storage and analog signal processing is proposed. Second, the device precision has been chosen to a moderate level to save area as much as possible. Since device mismatch limits the performance of analog circuits, the impact of device precision on the circuit performance is explicitly discussed. It is shown that the device precision limits the number of rows activated in parallel. Since the input vector as well as the output vector are considered to be sparsely coded it is concluded, that even for large matrices the proposed circuit technique is appropriate and ultra large scale integration with a large number of connection weights is feasible.     

Transport as a consequence of state-dependent diffusion

Overdamped particles subject to a drift in a force field with sinusoidal space dependence and also a sinusoidally modulated space-dependent diffusion, with the same period as the drift, experience a net driving force. The resulting current depends on the amplitude of the modulation of the diffusion and is a periodic function of the phase difference between the sinusoidal drift and the sinusoidal modulation of the diffusion. For small modulation amplitudes a particle subject to state-dependent noise behaves the same way as a particle subject to thermal noise but with a drift which, in addition to the sinusoidal term, contains a net force term.Dedicated to Professor Harry Thomas on the occasion of his 60th birthday     

Analytic calculation of scaling dimensions: Tricritical hard squares and critical hard hexagons

The finite-size corrections, central chargesc, and scaling dimensionsx of tricritical hard squares and critical hard hexagons are calculated analytically. This is achieved by solving the special functional equation or inversion identity satisfied by the commuting row transfer matrices of these lattice models at criticality. The results are expressed in terms of Rogers dilogarithms. For tricritical hard squares we obtainc=7/10,x=3/40, 1/5, 7/8, 6/5 and for hard hexagons we obtainc=4/5,x=2/15, 4/5, 17/15, 4/3, 9/5, in accord with the predictions of conformal and modular invariance.     

Portable and Automatic Mössbauer Analysis

A portable Mössbauer spectrometer, developed for extraterrestrial applications, opens up new industrial applications of MBS. But for industrial applications, an available tool for fast data analysis is also required, and it should be easy to handle. The analysis of Mössbauer spectra and their parameters is a barrier for the popularity of this wide-applicable spectroscopic technique in industry. Based on experience, the analysis of a Mössbauer spectrum is time-consuming and requires the dedication of a specialist. However, the analysis of Mössbauer spectra, from the fitting to the identification of the sample phases, can be faster using by genetic algorithms, fuzzy logic and artificial neural networks. Industrial applications are very specific ones and the data analysis can be performed using these algorithms. In combination with an automatic analysis, the Mössbauer spectrometer can be used as a probe instrument which covers the main industrial needs for an on-line monitoring of its products, processes and case studies. Some of these real industrial applications will be discussed.     

Changes in the kurtosis parameter of super-Gaussian beams passing through a spherically aberrated lens

Based on the Collins formula and irradiance moments definition, the propagation of the kurtosis parameter of super-Gaussian beams through a spherically aberrated lens is studied. Detailed numerical results are given. It is shown that, as compared with aberration-free super-Gaussian beams, the profile of spherically aberrated super-Gaussian beams can be leptokurtic, mesokurtic or platykurtic, depending on the aberration and propagation distance. The results for the spherically aberrated Gaussian beams can be dealt with as a special case treated in this paper. Finally, the advantage of our method is pointed out.     

Linear demosaicing inspired by the human visual system.

There is an analogy between single-chip color cameras and the human visual system in that these two systems acquire only one limited wavelength sensitivity band per spatial location. We have exploited this analogy, defining a model that characterizes a one-color per spatial position image as a coding into luminance and chrominance of the corresponding three colors per spatial position image. Luminance is defined with full spatial resolution while chrominance contains subsampled opponent colors. Moreover, luminance and chrominance follow a particular arrangement in the Fourier domain, allowing for demosaicing by spatial frequency filtering. This model shows that visual artifacts after demosaicing are due to aliasing between luminance and chrominance and could be solved using a preprocessing filter. This approach also gives new insights for the representation of single-color per spatial location images and enables formal and controllable procedures to design demosaicing algorithms that perform well compared to concurrent approaches, as demonstrated by experiments.     

Diffraction coefficients related to cylindrically curved soft-hard surfaces

An asymptotic solution to the problem of the diffraction of high-frequency waves by the edge of a cylindrically curved surface whose convex and concave faces present soft and hard boundary conditions is obtained. The results permit us to define new diffraction coefficients related to this kind of mixed boundary conditions. When the radius of the curvature becomes infinitely large, the edgediffraction coefficient reduces to this known for the softhard half-plane.     

Dielectric relaxation analysis of starch oligomers and polymers with respect to their chain length

Starch belongs to the polyglucan group. This type of polysaccharide shows a broad β-relaxation process in dielectric spectra at low temperatures, which has its molecular origin in orientational motions of sugar rings via glucosidic linkages. This chain dynamic was investigated for α(1,4)-linked starch oligomers with well-defined chain lengths of 2, 3, 4, 6, and 7 anhydroglucose units (AGUs) and for α(1,4)-polyglucans with average degrees of polymerization of 5, 10, 56, 70, and so forth (up to 3000; calculated from the mean molecular weight). The activation energy (E_a) of the segmental chain motion was lowest for dimeric maltose (E_a = 49.4 ± 1.3 kJ/mol), and this was followed by passage through a maximum at a degree of polymerization of 6 (E_a = 60.8 ± 1.8 kJ/mol). Subsequently, E_a leveled off at a value of about 52 ± 1.5 kJ/mol for chains containing more than 100 repeating units. The results were compared with the values of cellulose-like oligomers and polymers bearing a β(1,4)-linkage. Interestingly, the shape of the E_a dependency on the chain length of the molecules was qualitatively the same for both systems, whereas quantitatively the starch-like substances generally showed higher E_a values. Additionally, and for comparison, three cyclodextrins were measured by dielectric relaxation spectroscopy. The ringlike molecules, with 6, 7, and 8 α(1,4)-linked AGUs, showed moderately different types of dielectric spectra. © 2003 Wiley Periodicals, Inc. J Polym Sci Part B: Polym Phys 42: 188–197, 2004