Toshiba (Germany)

The aim of this study was to report initial clinical experience with a 320-slice CT scanner and to perform an image quality evaluation. 26 patients with presumptive cerebrovascular pathology underwent 320-slice CT. Single-rotation CT of the head, incremental CT angiography (three-dimensional (3D) CTA) as well as four-dimensional whole-brain CTA (4D CTA) and whole-brain CT perfusion (CTP) were performed and the resulting images were assessed for quality and compared with those obtained with 64-slice CT protocols. 320-slice CT neuroimaging could be performed in all cases. The image quality of 320-slice CT of the head and 3D CTA was inferior to that of the 64-slice protocols. The image quality of 4D 320-slice CTA was rated as inferior to both 320- and 64-slice 3D CTA. 4D CTA-CTP imaging added information with pivotal clinical implications. 320-slice CT neuroimaging is feasible technique that permits whole-brain 4D imaging and has the potential to identify pathologies with altered haemodynamics. However, image quality is a limitation of this technique at present.

Basic flip-flop structures are compared with the main emphasis on CMOS ASIC implementations. Flip-flop properties are analyzed by means of simplified models, some structural approaches for optimized metastable behavior are discussed. A special integrated test circuit which facilitates accurate and reproducible measurements is presented. The circuit has been used for carrying out metastability measurements in a wide temperature and voltage range to predict circuit parameters for worst-case designs. Results from measurements and circuit simulation indicate that different criteria for optimizing flip-flop performance should be used for synchronizers and for those applications where the observation of timing constraints imposed on flip-flop input signals can be guaranteed. These results can help in determining the reliability of existing synchronizer and arbiter designs. By means of special synchronizer cells the reliability of asynchronous interfaces can be improved significantly, enabling the system design to gain speed and flexibility in communication between independently clocked submodules.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

An integrated 1024/spl times/1024 CMOS image sensor with programmable region-of-interest (ROI) readout and multiexposure technique has been developed and successfully tested. Size and position of the ROI is programmed based on multiples of a minimum readout kernel of 32/spl times/32 pixels. Since the dynamic range of the irradiance normally exceeds the electrical dynamic range of the imager that can be covered using a single integration time, a multiexposure technique has been implemented in the imager. Subsequent sensor images are acquired using different integration times and recomputed to form a single composite image. A newly developed algorithm performing the recomputation is presented. The chip has been realized in a 0.5-/spl mu/m n-well standard CMOS process. The pixel pitch is 10 /spl mu/m/sup 2/ and the total chip area is 164 mm/sup 2/.

A new integrated system to design and generate a configurable embedded processor for multimedia applications has been developed. The system, "Media embedded Processor Integrator", provides a distinctive feature that generates development tools, such as compilers and simulators, not only for the configurable embedded processor but also for its template based extensible VLIW co-processor. This paper describes the architecture and the function of the "Media embedded Processor Integrator" especially focusing on how the system treats the VLIW co-processor extension. In order to determine an ISA for a 3-way VLIW co-processor for image recognition as an example, several different sets of ISA were evaluated and compared for the best performance using corresponding compilers and simulators, which were generated by the system. The system greatly contributed to reduce this entire ISA definition process.

The use of the CT scanner for cardiac imaging is mainly influenced by the spatial and temporal resolution that can be achieved with the applied technologies and procedures. The data acquisition with 16 x 0.5 mm scan slice thickness and a special multisegment image reconstruction procedure are a new combination for accurate imaging of the cardiac morphology. A 0.5 mm slice thickness and an overlapping pitch < 0.35 generate an isotropic image voxel of 0.35 x 0.35 x 0.35 mm. The object size of a coronary artery with a diameter of 2.5 mm amounts to a relative spatial blurring factor K (d) of approximately 15 %. The segment reconstruction with 4 segments from 4 consecutive cardiac cycles requires the optimum acquisition time of 50 ms for one frame. The relative exposure factor K (t) with reference to the R-R interval is an appropriate measure to validate the influence of coronary artery movement on the image quality at different heart rates. This relative exposure varies between 10 % and 20 % for a heart rate of 40 to 140 beats per minutes (bpm) and its mean is approximated by a linear trend function with K (t) = 14 %. A constant value in this linear trend function means a constant "blurring" of the imaged coronary arteries, independent of the actual heart rate. Thus, computed tomographic examinations can be carried out for heart rates between 40 and 140 bpm without using beta-blocking medication. Case studies of the 3D reconstruction and curved reformatting of coronary arteries with stents and calcifications show the achievable image quality at different heart rates.

OBJECTIVE: Postnatal outcome of fetuses with hypoplastic left heart syndrome (HLHS) is determined mainly by right ventricular function. Our study examines whether there are differences in right ventricular function during gestation of fetuses with HLHS compared with healthy fetuses. METHODS: A prospective study was conducted including 20 fetuses with HLHS and 20 gestational age-matched controls. Peak systolic and diastolic right ventricular free wall velocities were assessed using color tissue Doppler imaging (c-TDI). Subsequently, isovolumic time intervals, ejection time (ET'), E'/A' ratio and tissue Doppler-derived myocardial performance index (MPI') were calculated. Possible changes to c-TDI indices during the course of pregnancy in both the HLHS group and the control group were investigated. RESULTS: Examination of right ventricular function revealed significantly lower E' velocities (13.6 vs 21.0 cm/s; P = 0.017) and E'/A' ratios (0.55 vs 0.76; P = 0.012) and prolonged isovolumic contraction time (ICT') (57.0 vs 45.7 ms; P = 0.008) in the HLHS group compared with healthy fetuses. Furthermore, isovolumic relaxation time and MPI' increased significantly with gestational age in HLHS fetuses but not in controls. Values for systolic and diastolic peak velocities (E', A', S'), ET' and ICT' did not change significantly during gestation in either group. CONCLUSION: Right ventricular function in HLHS is altered as early as in fetal life, well before palliative surgery is performed. Future research should provide further insight into ventricular remodeling during gestation in cases of HLHS. Copyright © 2015 ISUOG. Published by John Wiley & Sons Ltd.

We propose for the first time a systematic evaluation of the performance and underlying trade-off of the use of ternary Hf <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1-x</sub> Al <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">x</sub> O <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">y</sub> oxides for RRAM application. We show that intermixing HfO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> and Al <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> O <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sub> deposition cycles in a standard ALD process not only prevents crystallization of active layer but also significantly improve intrinsic retention and disturb-immunity properties at the expense of a small increase of programming voltage. We attribute this beneficial effect to the reduced V <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">o</sub> diffusivity caused by stronger Al-O bonds as suggested by ab-initio simulations.

PURPOSE: To evaluate the potential of multisegmental reconstruction and variable gantry rotation time for reducing motion-induced artifacts in coronary artery imaging by multislice helical CT. MATERIALS AND METHODS: The data sets of 20 patients (8 with HR < 60 bpm, 12 with HR > 60 bpm) were analyzed. The patients underwent multislice helical CT (Aquilion 8, Toshiba, Otawara, Japan) using the following parameters: 0.5 mm slice thickness, 250 mA, 120 kV, pitch of 0.25 and variable gantry rotation times of 400, 500, or 600 msec. Images were generated by halfscan and multisegmental reconstruction. In 9 coronary segments of each patient, the presence and severity of motion artifacts were assessed and graded on a scale between 5 (no artifacts) and 1 (heaviest artifacts). RESULTS: Diagnostically relevant motion artifacts were rare at low heart rates (< 60 bpm) for both types of image reconstruction (4 % of all segments). Higher heart rates (> 60 bpm) were associated with an increase in motion artifacts on halfscan reconstructions (33% of all segments, p < 0.05) but not on multisegmental reconstructions (4% of all segments). At low heart rates mean image quality did not differ between multisegmental and halfscan reconstruction (4.28 +/- 0.37 vs. 4.22 +/- 0.41; p > 0.05), whereas at higher heart rates image quality was better for multisegmental reconstruction than for halfscan reconstruction (4.23 +/- 0.47 vs. 3.11 +/- 0.63; p < 0.05). CONCLUSION: Multisegmental reconstruction with variable gantry rotation times suppresses motion artifacts and thus improves assessment of the coronary arteries in patients with higher heart rates.

PURPOSE: We investigated prostate cancer (ca.) development after transurethral resection of the prostate (TURP). PATIENTS AND METHODS: From 1995 to 2003, 430 patients (pts.) received TURP at Toshiba Rinkan Hospital. Of them, 23 pts. (5.3%) had incidental carcinoma (Stage A), which developed into clinically significant ca. after 1 to 5 years in 5 (22% of Stage A, 1.2% of TURP). In 13 (3.2%) of 407 Non-Stage A pts. (who had no ca. initially), prostate ca. developed after 1 to 7 yrs. A total of 21 pts. (including 3 Stage A pts. diagnosed before 1994) underwent radical prostatectomy. Stage A pts. received regular needle biopsy of prostate (Pbx). Non-Stage A pts. were followed by yearly PSA measurement and digital rectal examination (DRE). Detailed histopathological studies were done on 21 radical prostatectomy specimens. RESULTS: Clinically significant ca. developed in 8 Stage A pts. (all A2) after 1 to 14 yrs. Long term (5 or 10 years) MAB therapy changed moderately-differentiated adenocarcinoma (AC) to poorly-differentiated AC in 2 pts. during follow-up. When ca. developed PSA increased in only 3 of them, DRE was positive just in 1 pt. Tumor invasion was observed mainly in transition zone (TZ), especially anterior to urethra. In spite of no capsular penetration, surgical margin was positive in 2 pts. PSA failure occurred in another 2 pts. Thirteen Non-Stage A pts. showed aggressive ca. (6 moderately-differentiated AC, 6 poorly-differentiated AC, and 1 ductal carcinoma which showed metastasis later), most of which invaded widely in peripheral zone (PZ). Pbx before TURP was done to reveal that there was no cancer in 11 pts. Capsular penetration was seen in 4 pts. Surgical margin was positive in 4 pts. PSA (8.6 +/- 4.0 ng/ml) decreased after TURP but was kept in high level (4.8 +/- 2.2 ng/ml) after 1 year and increased (8.7 +/- 4.5) when cancer was diagnosed in all 13 pts. DRE was positive in 38% of them. Interval between TURP and diagnosis was short in pts. who had cancer of high Gleason Score (GS) or large prostate. CONCLUSIONS: As significant cancer developed in 22% of Stage A pts. (1.2% of TURP) in long term follow-up, regular Pbx (to get TZ tissue) is mandatory regardless of PSA value or DRE. Aggressive cancer developed in 3.2% of Non-Stage A pts. (3.0% of TURP). Pts. with high PSA or abnormal DRE after TURP must receive needle biopsy actively. Considering that more than 4% of TURP pts. eventually require radical prostatectomy, relatively younger pts. who received TURP have to be carefully followed for a long period.

Starting from an evaluation of recent and future image processing algorithm's properties, this paper proposes a new class of parallel DSP architectures adapting the concept of simultaneous multithreading (SMT) to signal processing applications. This concept allows to enable parallelization resources on thread level, which are unused by most recent media-professors and video-DSPs. A customizable simulator to explore the architecture's parameters dependent on algorithmic properties and implementation constraints is presented. Coarse estimations for the realization costs in terms of silicon area are derived. First simulated performance figures for selected image processing algorithms show that SMT architectures are suitable to increase the processor's overall utilization and can achieve a speed-up beyond the limits of VLIW and superscalar architectures.

Electrical motors are the most important source of mechanical energy in the industrial world. Their modeling traditionally relies on a physics-based approach, which aims at taking their complex internal dynamics into account. In this paper, we explore the feasibility of modeling the dynamics of an electrical motor by following a data-driven approach, which uses only its inputs and outputs and does not make any assumption on its internal behaviour. We propose a novel encoder-decoder architecture which benefits from recurrent skip connections. We also propose a novel loss function that takes into account the complexity of electrical motor quantities and helps in avoiding model bias. We show that the proposed architecture can achieve a good learning performance on our high-frequency high-variance datasets. Two datasets are considered: the first one is generated using a simulator based on the physics of an induction motor and the second one is recorded from an industrial electrical motor. We benchmark our solution using variants of traditional neural networks like feedforward, convolutional, and recurrent networks. We evaluate various design choices of our architecture and compare it to the baselines. We show the domain adaptation capability of our model to learn dynamics just from simulated data by testing it on the raw sensor data. We finally show the effect of signal complexity on the proposed method ability to model temporal dynamics.

PURPOSE: Myocardial function (MF) of the systemic right ventricle (RV) influences the postnatal course of neonates with hypoplastic left heart syndrome (HLHS). Our study examines whether the presence of endocardial fibroelastosis of the left ventricle (LV EFE) influences MF of the RV in HLHS fetuses. MATERIALS AND METHODS: A prospective study was conducted including 10 controls (group 1), 10 HLHS fetuses with (group 2) and 10 without LV EFE (group 3) - all matched for gestational age. M-mode was used to assess tricuspid plane systolic excursion (TAPSE) and the shortening fraction (SF). PW-Doppler-derived and PW-TDI-derived velocities were assessed. E/A, E/e', e'/a' ratios and the myocardial performance index (mpi') were calculated. RESULTS: The examination of MF revealed significantly lower s' velocities (p < 0.05) and higher values for SF in group 2 compared to group 3. e'/a' ratio, et' (ejection time), E wave velocity, E/e' and SF showed significantly higher values in group 2 compared to group 1. In group 2 a' velocity increased significantly over gestational age. In group 3 but not in group 2, TAPSE increased during gestation. CONCLUSION: These significant differences in MF between the groups might lend support to the notion of negative ventricular-ventricular interaction in the case of HLHS with LV EFE possibly influencing surgical outcomes.

This paper focuses on the quantitative analysis of deep neural networks used in data-driven modeling of induction motor dynamics. With the availability of a large amount of data generated by industrial sensor networks, it is now possible to train deep neural networks. Recently researchers have started exploring the usage of such networks for physics modeling, online control, monitoring, and fault prediction in induction motor operations. We consider the problem of estimating speed and torque from currents and voltages of an induction motor. Neural networks provide quite good performance for this task when analysed from a machine learning perspective using standard metrics. We show, however, that there are some caveats in using machine learning metrics to analyze a neural network model when applied to induction motor problems. Given the mission- critical nature of induction motor operations, the performance of neural networks has to be validated from an electrical engineering point of view. To this end, we evaluate several traditional neural network architectures and recent state of the art architectures on dynamic and quasi-static benchmarks using electrical engineering metrics.

PURPOSE: To assess the time intervals of the cardiac cycle in healthy fetuses in the second and third trimester using color tissue Doppler imaging (cTDI) and to evaluate the influence of different sizes of sample gates on time interval values. MATERIALS AND METHODS: Time intervals were measured from the cTDI-derived Doppler waveform using a small and large region of interest (ROI) in healthy fetuses. RESULTS: 40 fetuses were included. The median gestational age at examination was 26 + 1 (range: 20 + 5 - 34 + 5) weeks. The median frame rate was 116/s (100 - 161/s) and the median heart rate 143 (range: 125 - 158) beats per minute (bpm). Using small and large ROIs, the second trimester right ventricular (RV) mean isovolumetric contraction times (ICTs) were 39.8 and 41.4 ms (p = 0.17), the mean ejection times (ETs) were 170.2 and 164.6 ms (p < 0.001), the mean isovolumetric relaxation times (IRTs) were 52.8 and 55.3 ms (p = 0.08), respectively. The left ventricular (LV) mean ICTs were 36.2 and 39.4 ms (p = 0.05), the mean ETs were 167.4 and 164.5 ms (p = 0.013), the mean IRTs were 53.9 and 57.1 ms (p = 0.05), respectively. The third trimester RV mean ICTs were 50.7 and 50.4 ms (p = 0.75), the mean ETs were 172.3 and 181.4 ms (p = 0.49), the mean IRTs were 50.2 and 54.6 ms (p = 0.03); the LV mean ICTs were 45.1 and 46.2 ms (p = 0.35), the mean ETs were 175.2 vs. 172.9 ms (p = 0.29), the mean IRTs were 47.1 and 50.0 ms (p = 0.01), respectively. CONCLUSION: Isovolumetric time intervals can be analyzed precisely and relatively independent of ROI size. In the near future, automatic time interval measurement using ultrasound systems will be feasible and the analysis of fetal myocardial function can become part of the clinical routine.

This paper presents the implementation and experimental evaluation of an advanced MIMO detector for wireless LAN systems. The proposed detector architecture is based on the well-known lattice-reduction aided MMSE method. Several optimizations at both algorithmic and architectural level are presented which result in an efficient VLSI design able to meet the timing requirements of a practical OFDM-based wireless LAN receiver while keeping complexity at moderate levels. Moreover, the detector offers built-in compensation for transmitter impairments such as nonlinear power amplifier characteristics, hence providing a full and cost-effective solution for practical systems. The described solution is implemented on an FPGA-based IEEE802.11n prototype and evaluation results comparing performance of both conventional MMSE and reduced-lattice detection under several propagation scenarios are presented. Experimental results show significantly lower error rates at the receiver for the advanced detector, or equivalently a lower number of required receiver antenna elements for a given performance target, hence resulting in lower cost, physical size and energy consumption <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup> .

OBJECTIVE: We sought to evaluate and validate a low-dose protocol for respiratory-gated multislice computed tomography (CT) for volume calculations in small ventilated neonatal animals as a model for the ventilated human neonatal lung. MATERIALS AND METHODS: Five mechanically ventilated newborn piglets were imaged in a multislice CT scanner (0.5-mm slice thickness, 4:16 pitch, 0.5 seconds rotation time, 120 kV) using a normal (100 mAs) and a reduced (10 mAs) dose protocol. All animals were scanned twice (at 100 and 10 mAs) at each of 3 different ventilator settings. Complete volume datasets were reconstructed throughout the respiratory cycle in increments of 10% using retrospective half-scan reconstruction. End-inspiratory volumes and volumes during maximal expiration (functional residual capacity) were calculated by a customized software and values for normal and reduced dose protocols were compared using Kolmogorov-Smirnov test and Bland-Altman plots. RESULTS: Two volume datasets (one normal and one reduced dose protocol) showed artifacts on the axial images, which could not be analyzed by the software. Those values were determined after manual segmentation and excluded from final analysis. The mean (+/-SD) end-inspiratory volumes and functional residual capacity were 34.3 +/- 10.1 mL and 25.3 +/- 8.0 mL for the normal-dose protocol versus 33.1 +/- 10.0 mL and 24.7 +/- 8.1 mL for the reduced-dose protocol, respectively. There was no statistically significant difference between normal and reduced dose protocol (KS-Test: D = 0.14 < Dmax). CONCLUSION: Lung volume calculation in ventilated newborn piglets (end-inspiratory volumes and functional residual capacity) can be performed using respiratory-gated multislice CT even at a substantially reduced dose (eg, to 10 mAs). This makes the technique a candidate for future pediatric use.

We propose a stereoscopic display system for multiple users. It allows three or more people to simultaneously observe individual stereoscopic image pairs from their own viewpoints. The system tracks the position of the user's heads and it generates distortion-free images for each user. The system consists of a normal display and a display mask with a hole in the center. The display mask is placed at a suitable distance over the display surface. By controlling the position of the image drawing area for each user according to the corresponding user's viewpoint, each user can observe the stereoscopic image pairs shown in an individual area of the display system with shutter glasses. On the other hand, no user is able to see the image drawing areas of the other users because these areas are adequately occluded by the display mask. Accordingly, the display system can simultaneously provide intelligible 3D stereoscopic images for three or more moving observers without flickering or distortion.

【目的】透析患者の栄養状態と大脳白質病変（white matter lesions：WMLs）の信号強度との関連について検討した．【方法】昭和大学病院の関連施設で維持血液透析中の患者（n＝28）を対象とした．全ての患者で脳MRI（T1，T2強調画像，FLAIR画像）を施行し，同時期に血液検査を含む各種臨床検査を施行した．栄養状態の指標はbody mass index（BMI），血清アルブミン濃度（Alb），標準蛋白異化率（nPCR），末梢血リンパ球数（TLC）を選択した．BMI＜18.5（kg/m2），Alb＜3.5（g/dL），nPCR＜0.9（g/kg/day），TLC＜1,500（cells/mm3）の4項目のうち，該当する数が1項目以下の群をgroupI，2項目の群をgroupII，3項目以上の群をgroupIIIとした．WMLsのスコア化はMRI画像上のWMLsをFazekas分類に基づき，脳室周囲高信号域スコア（perivascular hyperintensity（PVH）score）と，深部白質高信号域スコア（deep subcortical white matter hyperintensity（DSWMH）score）に分けて行った．【結果】groupIII(PVH score 2.40±0.90，DSWMH score 2.40±0.90），groupII(PVH score 1.75±0.71，DSWMH score 2.00±0.76）はgroupI(PVH score 0.93±0.46，DSWMH score 0.87±0.74）と比較して高いPVH・DSWMH scoreを示した．重回帰分析を行った結果，低栄養指標（上記4項目）の該当項目数がPVH score（標準化回帰係数（標準化β)＝0.608，p＜0.001），DSWMH score（標準化β＝0.537，p＜0.005）に対する有意な説明変数となった．【結語】血液透析患者の栄養状態（特に低栄養）とWMLsの信号強度が関連する可能性が考えられた．

We report a case of iatrogenic vesical tuberculosis diagnosed 4 years after intravesical immunotherapy using Bacillus Calmette- Guérin (BCG) for the treatment of bladder carcinoma. A 72-year-old man underwent a transurethral resection (TUR) of multiple noninvasive urothelial carcinomas and intravesical BCG infusion (40 mg/week) for 7 weeks to prevent the recurrence of bladder carcinoma. BCG infusion therapy was terminated because of the appearance of Reiter's syndrome, including arthritis of the left toe joint, conjunctivitis and non-gonococcal urethritis as complications. The patient suffered from repeated cystitis, bladder atrophy and urethral stenosis. The cystitis improved with the administration of antibiotics (Levofloxacin) but persisted without a complete cure. Four years later, a cystoscopy revealed mucosal erosion and a white coating. An acid-fast bacteria examination of a urine sample using bacteria incubation and DNA PCR revealed the presence of Mycobacterium bovis. Finally, anti-tuberculosis therapy (INH＋REP＋EB) was initiated after the patient was diagnosed as having iatrogenic bladder tuberculosis resulting from BCG immunotherapy. The tuberculosis bacteria subsequently disappeared from the urine samples, and the gross appearance of the bladder mucosa improved. Bladder carcinoma has not recurred to date. Intravesical BCG infusion therapy has a good anti-tumor effect and can help prevent tumor recurrence after TUR therapy in case of noninvasive bladder carcinoma. However, there is a risk of severe complications arising from the BCG infusion. In the present case, an adequate bacteria examination was not performed, even though antibiotics were repeatedly administered for cystitis. In particular, the patient was not tested for the presence of acid-fast bacteria for 4 years after the intravesical BCG infusion therapy. Furthermore, among patients who received anti-bacteria therapy for repeated cystitis after BCG infusion, a bacteria examination including bacteria incubation, was not ordered in 19 out of 30 cases treated at our hospital over the past 5 years. In conclusion, bacteria examination, including tests for acid-fast bacteria, should be immediately performed when repeated and/or persistent cystitis occurs after BCG infusion therapy.

On January 6,2009,Jindongnan-Nanyang-Jingmen 1000 kV UHV AC pilot demonstration project was successfully put into operation,so far the system is running smoothly and equipment is in good condition.EPRI Toshiba Arrester Co.,Ltd.provides GIS and porcelain housed surge arresters.The research and development process and performance of GIS and porcelain housed surge arresters are reviewed detailedly.Research and development work,including high energy capacity ZnO elements,GIS and porcelain housed surge arresters,is based on the key technologies of Toshiba.High energy capacity ZnO element,which has sufficient safety margin,meets the requirements of 1000 kV GIS and porcelain housed surge arresters fully,and has actually applied to 500 kV and 750 kV porcelain housed surge arresters.Potential distribution of GIS surge arresters analytical results and measured results are consistent,and meet the requirements of bid documents and technology agreements.Porcelain housed surge arresters are an international ground-breaking products,which apply four key technologies with the actual operation performance of Toshiba.Especially the damping technology(device) significantly reduces the root stress of porcelain bushing to meet anti-earthquake requirements.On the basis of introducing the development process and results,we summarize research and design work,in order to accumulate experience and raise awareness,and better serve for other UHV AC projects.