In 47 patients with liver cirrhosis, we performed dynamic MRI with a multisection FLASH technique that enabled us to obtain 13 T 1-weighted images of the entire liver within a single breath hold. Computed tomographic arterial portography (CTAP), US, CT, angiography (AOG) and MRI (spin echo [SE] and dynamic MRI) were performed in all 47 patients. Except for cyst, hemangioma and metastatic tumor, 104 focal nodules less than 3 cm in diameter were detected. These 104 focal lesions were divided into three groups according to the pattern of CTAP: 69 portal supply negative, 11 portal supply decreased, and 24 portal supply normal. In the portal supply negative group, 63 lesions (91%) were detected by dynamic MRI, which was superior to other modalities (US 77%, CT 41%, AOG 70%, MRI-SE 61%). The superiority of dynamic MRI resulted from its excellent ability to detect liver lesions less than 1 cm in diameter. We confirmed histologically that dynamic MRI had almost the same ability to detect hepatocellular carcinoma (HCC) as CTAP. Dynamic MRI should be clinically useful as a noninvasive examination for the detection of HCC.
The purpose of this study is to produce the Cu-Ni-Si copper alloy strip by using a vertical type twin-roll casting process. The demand of Cu-Ni-Si copper alloy sheet called Corson alloy sheet has been increasing for small electronic devices because of these remarkable properties such as good bending workability, high strength and high conductivity. However it has difficulty for Cu-Ni-Si copper alloy to produce a thin strip by conventional hot rolling process, thus it needs many process because it has to repeat cold rolling and annealing. Therefore new producing process for Cu-Ni-Si copper alloy strip is required. In this study, Cu-5Ni-1.25Si and Cu-10Ni-2.5Si alloy strip were casted by vertical twin-roll process. Cast ability and surface conditions were investigated. It was possible to produce continuously these Cu-Ni-Si copper alloy strip by using a vertical type twin-roll casting process.
Many researchers of both domestic and oversea promote the report in relation to the liquid ring pump. However, there exist only few reports that are discussed about the visualization of the liquid ring between the liquid as the working fluid and pressured gas at present. What is worse, the content of such current discussed report of the visualization of the liquid ring is only visualized the cross sectional images of the inside of the liquid ring pump by taking a photograph or other media.Therefore, our report "The Visualization of Liquid Ring inside a Liquid Ring Pump" treats this research theme by visualizing and discussing the visualization of the liquid ring between the liquid as the working fluid and pressured gas systematically, and figures out the characteristics of the liquid ring pump under various conditions for more meaningful development of it.
In this paper, we discuss an obstacle to training in SpikeProp[1], which is a type of supervised learning algorithms for spiking neural networks. In the original publication of SpikeProp, weights with mixed signs are suspected to cause failures of training. We pointed out the cause of it through some experiments. Weights with mixed signs make the dynamics of the unit's activity twisted, and the twisted dynamics break the assumption that SpikeProp algorithm is based on. Therefore, it causes surges in training processes. They would mean an underlying problem on training processes.
