Toshiba (South Korea)

Magnetic Resonance Imaging: A Preview. Classical of a Single Nucleus to a Magnetic Field. Rotating Reference Frames and Resonance. Magnetization, Relaxation and the Bloch Equation. The Quantum Mechanical Basis of Precession and Excitation. The Quantum Mechanical Basis of Thermal Equilibrium and Longitudinal Relaxation. Signal Detection Concepts. Introductory Signal Acquisition Methods: Free Induction Decay, Spin Echoes, Inversion Recovery and Spectroscopy. One-Dimensional Fourier Imaging, k-Space and Gradient Echoes. Multi-Dimensional Fourier Imaging and Slice Excitation. The Continuous and Discrete Fourier Transforms. Sampling and Aliasing in Image Reconstruction. Filtering and Resolution in Fourier Transform Image Reconstruction. Projection Reconstruction of Images. Signal, Contrast and Noise. A Closer Look at Radiofrequency Pulses. Water/Fat Separation Techniques. Fast Imaging in the Steady State. Segmented k-Space and Echo Planar Imaging. Magnetic Field Inhomogeneity Effects and T-2 Dephasing. Random Walks, Relaxation and Diffusion. Spin Density, T-1 and T-2 Quantification Methods in MR Imaging. Motion Artifacts and Flow Compensation. MR Angiography and Flow Quantification. Magnetic Properties of Tissues: Theory and Measurement. Sequence Design, Artifacts and Nomenclature. Introduction to MRI Coils and Magnets. Appendices. Index.

A new method is proposed to subtract the count of scattered photons from that acquired with a photopeak window at each pixel in each planar image of single-photon emission computed tomography (SPECT). The subtraction is carried out using two sets of data: one set is acquired with a main window centered at photopeak energy and the other is acquired with two subwindows on both sides of the main window. The scattered photons included in the main window are estimated from the counts acquired with the subwindows and then they are subtracted from the count acquired with the main windows. Since the subtraction is performed at each pixel in each planar image, the proposed method has the potential to be more precise than conventional methods. For three different activity distributions in cylinder phantoms, simulation tests gave good agreement between the activity distributions reconstructed from unscattered photons and those from the corrected data.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

A 4 mm/sup 2/, two-dimensional (2-D) 8/spl times/8 discrete cosine transform (DCT) core processor for HDTV-resolution video compression/decompression in a 0.3-/spl mu/m CMOS triple-well, double-metal technology operates at 150 MHz from a 0.9-V power supply and consumes 10 mW, only 2% power dissipation of a previous 3.3-V design. Circuit techniques for dynamically varying threshold voltage (VT scheme) are introduced to reduce active power dissipation with negligible overhead in speed, standby power dissipation, and chip area. A way to explore V/sub DD/-V/sub th/ design space is also studied.

Eight synergistic processor units enable the Cell Broadband Engine's breakthrough performance. The SPU architecture implements a novel, pervasively data-parallel architecture combining scalar and SIMD processing on a wide data path. A large number of SPUs per chip provide high thread-level parallelism. The streamlined architecture provides an efficient multithreaded execution environment for both scalar and SIMD threads and represents a reaffirmation of the RISC principles of combining leading edge architecture and compiler optimizations. These design decisions have enabled the Cell BE to deliver unprecedented supercomputer-class compute power for consumer applications

The spatial distribution of scattered photons varies depending on many factors such as object size and source distribution. We propose a triple-energy window (TEW) scatter compensation method for determining position-dependent Compton scatter. We estimated the count of primary photons at each pixel in the acquired image using the 24% main window centered at the photo peak energy and 3 keV scatter rejection windows on both sides of the main window. We conducted a physical evaluation of this method using phantoms and also applied this method to patients in a clinical trial. The TEW method performed Compton scatter compensation with good accuracy.

OBJECTIVE: Wingspan (Boston Scientific, Fremont, CA) is a self-expanding stent designed specifically for the treatment of symptomatic intracranial atheromatous disease. The current series reports the observed incidence of in-stent restenosis (ISR) and thrombosis on angiographic follow-up. METHODS: A prospective, intent-to-treat registry of patients in whom the Wingspan stent system was used to treat symptomatic intracranial atheromatous disease was maintained at five participating institutions. Clinical and angiographic follow-up results were recorded. ISR was defined as stenosis greater than 50% within or immediately adjacent (within 5 mm) to the implanted stents and absolute luminal loss greater than 20%. RESULTS: To date, follow-up imaging (average duration, 5.9 mo; range, 1.5-15.5 mo) is available for 84 lesions treated with the Wingspan stent (78 patients). Follow-up examinations consisted of 65 conventional angiograms, 17 computed tomographic angiograms, and two magnetic resonance angiograms. Of these lesions with follow-up, ISR was documented in 25 and complete thrombosis in four. Two of the 4 patients with stent thrombosis had lengthy lesions requiring more than one stent to bridge the diseased segment. ISR was more frequent (odds ratio, 4.7; 95% confidence intervals, 1.4-15.5) within the anterior circulation (42%) than the posterior circulation (13%). Of the 29 patients with ISR or thrombosis, eight were symptomatic (four with stroke, four with transient ischemic attack) and 15 were retreated. Of the retreatments, four were complicated by clinically silent in-stent dissections, two of which required the placement of a second stent. One was complicated by a postprocedural reperfusion hemorrhage. CONCLUSION: The ISR rate with the Wingspan stent is higher in our series than previously reported, occurring in 29.7% of patients. ISR was more frequent within the anterior circulation than the posterior circulation. Although typically asymptomatic (76% of patients in our series), ISR can cause neurological symptoms and may require target vessel revascularization.

Optimal process integration for array devices of bit-cost scalable (BiCS) flash memory is successfully developed. We adopt SiN-based gate dielectrics for the consistency with the 'gate-first' process which is unique to BiCS flash technology, and 'macaroni' body FETs for better controllability over the sub-threshold characteristics of depletion-mode poly-silicon transistors. With these technologies and newly devised 4F <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> cell array, BiCS flash becomes a promising candidate for future ultra-high density memory.

In order to realize ultra high density EPROM and Flash EEPROM, a NAND structure cell is proposed. This new structure is able to shrink cell size without scaling of device dimensions. The NAND structure cell realizes a cell as small as 6.43 µm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> using 1.0 µm design rule. As a result, cell area per bit can be reduced by 30% compared with that of a 4M bit EPROM using the conventional structure and the same design rule. It is confirmed that each bit in a NAND cell is able to be programmed selectively. This high performance NAND structure cell is applicable to high density nonvolatile memories as large as 8M bit EPROM and Flash-EEPROM or beyond.

We present the concept of logarithmic computation for neural networks. We explore how logarithmic encoding of non-uniformly distributed weights and activations is preferred over linear encoding at resolutions of 4 bits and less. Logarithmic encoding enables networks to 1) achieve higher classification accuracies than fixed-point at low resolutions and 2) eliminate bulky digital multipliers. We demonstrate our ideas in the hardware realization, LogNet, an inference engine using only bitshift-add convolutions and weights distributed across the computing fabric. The opportunities from hardware work in synergy with those from the algorithm domain.

Diffusion-weighted imaging (DWI) has recently been attempted in the abdominal region. We review diffusion-weighted images of the liver, especially from the technical point of view. We discuss selection of pulse sequence parameters, effects of anti-breathing motion technique, tips for measuring apparent diffusion coefficient (ADC), and utility of superparamagnetic iron oxide (SPIO), showing clinical cases, including those at 3T. Our current trial of new pulse sequencing, such as SPIO-mediated breath-holding black-blood fluid-attenuated inversion recovery (BH-BB-FLAIR), imaging is shown. Some prospects for the future in DWI of the liver are also stated.

For the first time, 4Gbit density STT-MRAM using perpendicular MTJ in compact cell was successfully demonstrated through the tight distributions for resistance and magnetic properties. This paper includes the results regarding parasitic resistance control process, MTJ process, and MTJ stack engineering. Both of successful 4Gb read and write operations were performed with high TMR, low Ic. This result will brighten the prospect of high-density STT-MRAM.

A novel approach for achieving high-performance Schottky-source/drain MOSFETs (SBTs: Schottky Barrier Transistors) is proposed. The dopant segregation (DS) technique is employed and significant modulation of Schottky barrier height is demonstrated. The DS-SBT fabricated with the current CoSi/sub 2/ process show competitive drive current and better short-channel-effect immunity compared to the conventional MOSFET. In conclusion the DS-Schottky junction is useful for the source/drain of advanced MOSFETs.

BACKGROUND/AIMS: Hemodynamics constitute a critical factor in the formation of intracranial aneurysms. However, little is known about how intracranial arteries respond to hemodynamic insult and how that response contributes to aneurysm formation. We examined early cellular responses at rabbit basilar termini exposed to hemodynamic insult that initiates aneurysmal remodeling. METHODS: Flow in the basilar artery was increased by bilateral carotid artery ligation. After 2 and 5 days, basilar terminus tissue was examined by immunohistochemistry and quantitative PCR. RESULTS: Within 2 days of flow increase, internal elastic lamina (IEL) was lost in the periapical region of the bifurcation, which experienced high wall shear stress and positive wall shear stress gradient. Overlying endothelium was still largely present in this region. IEL loss was associated with localized apoptosis and elevated expression of matrix metalloproteinases (MMPs) 2 and 9. A small number of inflammatory cells were sporadically scattered in the bifurcation adventitia and were not concentrated in regions of IEL loss and MMP elevation. Elevated MMP expression colocalized with smooth muscle α-actin in the media. CONCLUSION: The initial vascular response to aneurysm-initiating hemodynamic insult includes localized matrix degradation and cell apoptosis. Such destructive remodeling arises from intrinsic mural cells, rather than through inflammatory cell infiltration.

The Schottky contact is an important consideration in the development of semiconductor devices. This paper shows that a practical Schottky contact model is available for a unified device simulation of Schottky and ohmic contacts. The present model includes the thermionic emission at the metal/semiconductor interface and the spatially distributed tunneling calculated at each semiconductor around the interface. Simulation results of rectifying characteristics of Schottky barrier diodes (SBD's) and resistances under high impurity concentration conditions are reasonable, compared with measurements. As examples of application to actual devices, the influence of the contact resistance on salicided MOSFETs with source/drain extension and the immunity of Schottky barrier tunnel transistors (SBTTs) from the short-channel effect (SCE) are demonstrated.

PURPOSE: To prospectively compare the capability of quantitative first-pass perfusion 320-detector row computed tomography (CT) (ie, area-detector CT) with that of combined positron emission tomography and CT (PET/CT) for differentiation between malignant and benign pulmonary nodules. MATERIALS AND METHODS: This prospective study was approved by the institutional review board, and written informed consent was obtained from 50 consecutive patients with 76 pulmonary nodules. All patients underwent dynamic area-detector CT, PET/CT, and microbacterial and/or histopathologic examinations. All pulmonary nodules were divided into three groups: malignant nodules (n = 43), benign nodules with low biologic activity (n = 6), and benign nodules with high biologic activity (n = 27). For each dynamic area-detector CT data set, the perfusion derived by using the maximum slope model (PF(MS)), extraction fraction derived by using the Patlak plot model (EF(PP)), and blood volume derived by using the Patlak plot model (BV(PP)) were calculated. These parameters were statistically compared among the three nodule groups. Receiver operating characteristic (ROC) analyses were used to compare the diagnostic capability of the CT and PET/CT indexes. Finally, the sensitivity, specificity, and accuracy of each index were compared by using the McNemar test. RESULTS: All indexes in the malignant nodule group were significantly different from those in the low-biologic-activity benign nodule group (P < .05). Areas under the ROC curve for PF(MS) and EF(PP) were significantly larger than those for BV(PP) (P < .05) and maximal standard uptake value (SUV(max)) (P < .05). The specificity and accuracy of PF(MS) and EF(PP) were significantly higher than those of BV(PP) and SUV(max) (P < .05). CONCLUSION: Dynamic first-pass area-detector perfusion CT has the potential to be more specific and accurate than PET/CT for differentiating malignant from benign pulmonary nodules. SUPPLEMENTAL MATERIAL: http://radiology.rsna.org/lookup/suppl/doi:10.1148/radiol.10100245/-/DC1.

We propose a novel Si dot memory whose floating gate consists of self-aligned doubly stacked Si dots. A lower Si dot exists immediately below an upper dot and lies between thin tunnel oxides. It is experimentally shown that charge retention is improved compared to the usual single-layer Si dot memory. A theoretical model considering quantum confinement and Coulomb blockade in the lower Si dot explains the experimental results consistently, and shows that charge retention is improved exponentially by lower dot size scaling. It is shown that the retention improvement by lower dot scaling is possible, keeping the same write/erase speed as single dot memory, when the tunnel oxide thickness is adjusted simultaneously.

This paper describes a technique for controlling the degree of expressivity of a desired emotional expression and/or speaking style of synthesized speech in an HMM-based speech synthesis framework. With this technique, multiple emotional expressions and speaking styles of speech are modeled in a single model by using a multiple-regression hidden semi-Markov model (MRHSMM). A set of control parameters, called the style vector, is defined, and each speech synthesis unit is modeled by using the MRHSMM, in which mean parameters of the state output and duration distributions are expressed by multiple-regression of the style vector. In the synthesis stage, the mean parameters of the synthesis units are modified by transforming an arbitrarily given style vector that corresponds to a point in a low-dimensional space, called style space, each of whose coordinates represents a certain specific speaking style or emotion of speech. The results of subjective evaluation tests show that style and its intensity can be controlled by changing the style vector.

A 60-mW MPEG4 video codec has been developed for mobile multimedia applications. This codec supports both the H.263 ITU-T recommendation and the simple profile of MPEG4 committee draft version 1 released in November 1997. It is composed of a 16-bit reduced instruction set computer processor and several dedicated hardware engines so as to satisfy both power efficiency and programmability. It performs 10 frames/s of encoding and decoding with quarter-common intermediate format at 30 MHz. Several innovative low-power techniques were employed in both architectural and circuit levels, and the final power dissipation is 60 mW at 30 MHz, which is only 30% of the power dissipation for a conventional CMOS design. The chip was fabricated in a 0.3-/spl mu/m CMOS with double-well and triple-metal technology. It contains 3 million transistors, including a 52-kB on-chip SRAM. Internal supply voltages of 2.5 and 1.75 V are generated by on-chip dc-dc converters from 3.3-V external supply voltage.

macroeconomics, corporate group, Japan, trade, antitrust enforcement, keiretsu

In this work, we propose a compact, physically based, analytical single-electron transistor (SET) model suitable for the design and analysis of realistic SET circuits. The model is derived on the basis of the “orthodox” theory of correlated single-electron tunneling and the steady-state master equation method. The SET inverter characteristics are successfully calculated using the model implemented in the simulation program with integrated circuit emphasis (SPICE). The hybrid circuit of SETs with metal-oxide-semiconductor field-effect transistors (MOSFETs) is also successfully simulated. By utilizing the model, it is clarified that the drain-voltage-induced shift of the gate voltage dependence of SET current reaches one-half of the drain voltage in the case of a completely symmetric SET.