• Journal of Veterinary Clinics
• /
• v.30 no.6
• /
• pp.456-458
• /
• 2013

The aim of this study is to evaluate the reference value for electrocardiogram in healthy captive raccoon dogs. Forty-one free-ranging adult raccoon dogs rescued from Wildlife rescue centre, Kangwon National University were enrolled in this study. The 6-lead electrocardiogram was obtained in all raccoon dogs without any chemical restraints. The mean heart rate was $146.10{\pm}43.31$ beats/min (95% confidence interval 132.84~159.36 beats/min). The mean respiration rate was $35.73{\pm}11.56$ breaths/min (95% confidence interval 32.19~39.27 breaths/min). The mean systolic blood pressure was $136{\pm}29.26$ mmHg (95% confidence interval 127.99~145.91 mmHg). Electrocardiographical features were also evaluated in all raccoon dogs. The mean duration and amplitude of P-wave were $38.2{\pm}4.0$ ms (range 28-40 ms) and $0.128{\pm}0.039$ mV (range 0.09~0.20). The mean duration and amplitude of QRS complexes were $48.5{\pm}7.2ms$ (range 36-60 ms) and $1.330{\pm}0.650$ mV (range 0.15~2.30). The range of the mean electrical (QRS) axis was $-91^{\circ}{\sim}+96^{\circ}$ ($10^{\circ}{\sim}60^{\circ}$; 95% of confidence interval). The mean corrected QT (QTc) interval was $273.7{\pm}32.7ms$ (range 212-333 ms), while the mean PR interval was $76.1{\pm}10.0ms$ (range 50-82 ms). To the authors' knowledge, this is the first study to provide references in electrocardiogram (ECG) in healthy captive raccoon dogs.

• Korean Chemical Engineering Research
• /
• v.45 no.5
• /
• pp.466-472
• /
• 2007

A technology platform for wafer-level three-dimensional integration circuits (3D-ICs) is presented, and that uses wafer bonding with low-k polymeric adhesives and Cu damascene inter-wafer interconnects. In this work, one of such technical platforms is explained and characterized using a test vehicle of inter-wafer 3D via-chain structures. Electrical and mechanical characterizations of the structure are performed using continuously connected 3D via-chains. Evaluation results of the wafer bonding, which is a necessary process for stacking the wafers and uses low-k dielectrics as polymeric adhesive, are also presented through the wafer bonding between a glass wafer and a silicon wafer. After wafer bonding, three evaluations are conducted; (1) the fraction of bonded area is measured through the optical inspection, (2) the qualitative bond strength test to inspect the separation of the bonded wafers is taken by a razor blade, and (3) the quantitative bond strength is measured by a four point bending. To date, benzocyclobutene (BCB), $Flare^{TM}$, methylsilsesquioxane (MSSQ) and parylene-N were considered as bonding adhesives. Of the candidates, BCB and $Flare^{TM}$ were determined as adhesives after screening tests. By comparing BCB and $Flare^{TM}$, it was deduced that BCB is better as a baseline adhesive. It was because although wafer pairs bonded using $Flare^{TM}$ has a higher bond strength than those using BCB, wafer pairs bonded using BCB is still higher than that at the interface between Cu and porous low-k interlevel dielectrics (ILD), indicating almost 100% of bonded area routinely.

• Journal of IKEEE
• /
• v.25 no.3
• /
• pp.493-500
• /
• 2021

In this paper, we propose the development of deep learning structure to improve quality of polygonal containers. The deep learning structure consists of a convolution layer, a bottleneck layer, a fully connect layer, and a softmax layer. The convolution layer is a layer that obtains a feature image by performing a convolution 3x3 operation on the input image or the feature image of the previous layer with several feature filters. The bottleneck layer selects only the optimal features among the features on the feature image extracted through the convolution layer, reduces the channel to a convolution 1x1 ReLU, and performs a convolution 3x3 ReLU. The global average pooling operation performed after going through the bottleneck layer reduces the size of the feature image by selecting only the optimal features among the features of the feature image extracted through the convolution layer. The fully connect layer outputs the output data through 6 fully connect layers. The softmax layer multiplies and multiplies the value between the value of the input layer node and the target node to be calculated, and converts it into a value between 0 and 1 through an activation function. After the learning is completed, the recognition process classifies non-circular glass bottles by performing image acquisition using a camera, measuring position detection, and non-circular glass bottle classification using deep learning as in the learning process. In order to evaluate the performance of the deep learning structure to improve quality of polygonal containers, as a result of an experiment at an authorized testing institute, it was calculated to be at the same level as the world's highest level with 99% good/defective discrimination accuracy. Inspection time averaged 1.7 seconds, which was calculated within the operating time standards of production processes using non-circular machine vision systems. Therefore, the effectiveness of the performance of the deep learning structure to improve quality of polygonal containers proposed in this paper was proven.