• Journal of Chest Surgery
• /
• v.39 no.12 s.269
• /
• pp.879-890
• /
• 2006

Background: The dysfunction of multiple organs is found to be caused by reactive oxygen species as a major modulator of microvascular injury after hemorrhagic shock. Hemorrhagic shock, one of many causes inducing acute lung injury, is associated with increase in alveolocapillary permeability and characterized by edema, neutrophil infiltration, and hemorrhage in the interstitial and alveolar space. Aggressive and rapid fluid resuscitation potentially might increased the risk of pulmonary dysfunction by the interstitial edema. Therefore, in order to improve the pulmonary dysfunction induced by hemorrhagic shock, the present study was attempted to investigate how to reduce the inflammatory responses and edema in lung. Material and Method: Male Sprague-Dawley rats, weight 300 to 350 gm were anesthetized with ketamine(7 mg/kg) intramuscular Hemorrhagic Shock(HS) was induced by withdrawal of 3 mL/100 g over 10 min. through right jugular vein. Mean arterial pressure was then maintained at $35{\sim}40$ mmHg by further blood withdrawal. At 60 min. after HS, the shed blood and Ringer's solution or 5% albumin was infused to restore mean carotid arterial pressure over 80 mmHg. Rats were divided into three groups according to rectal temperature level($37^{\circ}C$[normothermia] vs $33^{\circ}C$[mild hypothermia]) and resuscitation fluid(lactate Ringer's solution vs 5% albumin solution). Group I consisted of rats with the normothermia and lactate Ringer's solution infusion. Group II consisted of rats with the systemic hypothermia and lactate Ringer's solution infusion. Group III consisted of rats with the systemic hypothermia and 5% albumin solution infusion. Hemodynamic parameters(heart rate, mean carotid arterial pressure), metabolism, and pulmonary tissue damage were observed for 4 hours. Result: In all experimental groups including 6 rats in group I, totally 26 rats were alive in 3rd stage. However, bleeding volume of group I in first stage was $3.2{\pm}0.5$ mL/100 g less than those of group II($3.9{\pm}0.8$ mL/100 g) and group III($4.1{\pm}0.7$ mL/100 g). Fluid volume infused in 2nd stage was $28.6{\pm}6.0$ mL(group I), $20.6{\pm}4.0$ mL(group II) and $14.7{\pm}2.7$ mL(group III), retrospectively in which there was statistically a significance between all groups(p<0.05). Plasma potassium level was markedly elevated in comparison with other groups(II and III), whereas glucose level was obviously reduced in 2nd stage of group I. Level of interleukine-8 in group I was obviously higher than that of group II or III(p<0.05). They were $1.834{\pm}437$ pg/mL(group I), $1,006{\pm}532$ pg/mL(group II), and $764{\pm}302$ pg/mL(group III), retrospectively. In histologic score, the score of group III($1.6{\pm}0.6$) was significantly lower than that of group I($2.8{\pm}1.2$)(p<0.05). Conclusion: In pressure-controlled hemorrhagic shock model, it is suggested that hypothermia might inhibit the direct damage of ischemic tissue through reduction of basic metabolic rate in shock state compared to normothermia. It seems that hypothermia should be benefit to recovery pulmonary function by reducing replaced fluid volume, inhibiting anti-inflammatory agent(IL-8) and leukocyte infiltration in state of ischemia-reperfusion injury. However, if is considered that other changes in pulmonary damage and inflammatory responses might induce by not only kinds of fluid solutions but also hypothermia, and that the detailed evaluation should be study.

• Journal of Intelligence and Information Systems
• /
• v.24 no.1
• /
• pp.205-225
• /
• 2018

Convolutional Neural Network (ConvNet) is one class of the powerful Deep Neural Network that can analyze and learn hierarchies of visual features. Originally, first neural network (Neocognitron) was introduced in the 80s. At that time, the neural network was not broadly used in both industry and academic field by cause of large-scale dataset shortage and low computational power. However, after a few decades later in 2012, Krizhevsky made a breakthrough on ILSVRC-12 visual recognition competition using Convolutional Neural Network. That breakthrough revived people interest in the neural network. The success of Convolutional Neural Network is achieved with two main factors. First of them is the emergence of advanced hardware (GPUs) for sufficient parallel computation. Second is the availability of large-scale datasets such as ImageNet (ILSVRC) dataset for training. Unfortunately, many new domains are bottlenecked by these factors. For most domains, it is difficult and requires lots of effort to gather large-scale dataset to train a ConvNet. Moreover, even if we have a large-scale dataset, training ConvNet from scratch is required expensive resource and time-consuming. These two obstacles can be solved by using transfer learning. Transfer learning is a method for transferring the knowledge from a source domain to new domain. There are two major Transfer learning cases. First one is ConvNet as fixed feature extractor, and the second one is Fine-tune the ConvNet on a new dataset. In the first case, using pre-trained ConvNet (such as on ImageNet) to compute feed-forward activations of the image into the ConvNet and extract activation features from specific layers. In the second case, replacing and retraining the ConvNet classifier on the new dataset, then fine-tune the weights of the pre-trained network with the backpropagation. In this paper, we focus on using multiple ConvNet layers as a fixed feature extractor only. However, applying features with high dimensional complexity that is directly extracted from multiple ConvNet layers is still a challenging problem. We observe that features extracted from multiple ConvNet layers address the different characteristics of the image which means better representation could be obtained by finding the optimal combination of multiple ConvNet layers. Based on that observation, we propose to employ multiple ConvNet layer representations for transfer learning instead of a single ConvNet layer representation. Overall, our primary pipeline has three steps. Firstly, images from target task are given as input to ConvNet, then that image will be feed-forwarded into pre-trained AlexNet, and the activation features from three fully connected convolutional layers are extracted. Secondly, activation features of three ConvNet layers are concatenated to obtain multiple ConvNet layers representation because it will gain more information about an image. When three fully connected layer features concatenated, the occurring image representation would have 9192 (4096+4096+1000) dimension features. However, features extracted from multiple ConvNet layers are redundant and noisy since they are extracted from the same ConvNet. Thus, a third step, we will use Principal Component Analysis (PCA) to select salient features before the training phase. When salient features are obtained, the classifier can classify image more accurately, and the performance of transfer learning can be improved. To evaluate proposed method, experiments are conducted in three standard datasets (Caltech-256, VOC07, and SUN397) to compare multiple ConvNet layer representations against single ConvNet layer representation by using PCA for feature selection and dimension reduction. Our experiments demonstrated the importance of feature selection for multiple ConvNet layer representation. Moreover, our proposed approach achieved 75.6% accuracy compared to 73.9% accuracy achieved by FC7 layer on the Caltech-256 dataset, 73.1% accuracy compared to 69.2% accuracy achieved by FC8 layer on the VOC07 dataset, 52.2% accuracy compared to 48.7% accuracy achieved by FC7 layer on the SUN397 dataset. We also showed that our proposed approach achieved superior performance, 2.8%, 2.1% and 3.1% accuracy improvement on Caltech-256, VOC07, and SUN397 dataset respectively compare to existing work.