• The Journal of the Korean bone and joint tumor society
• /
• v.14 no.2
• /
• pp.95-105
• /
• 2008

Purpose: To identify which is the best procedure in recycling autograft according to the resection & reconstruction type and recycling methods, and so when the recycling autograft is used in limb salvage surgery. Materials and Methods: We have treated fifty-eight patients (34 male, 24 female; age range 5 to 74 years, mean age 36.5 years), who had the malignant musculoskeletal tumors, with recycling autograft (47 patients with extracoporeal irradiation, 11 patients with pasteurization) from December 1995 to February 2006. The resection and reconstruction type was 3 cases with fragmentary, 8 intercalary, 23 rAPC (recycling-Autograft-Prosthesis composite), 18 osteoarticular, 5 total joint and 1 soft tissue (achilles tendon). The result was evaluated by the radiologic union at junctional site, the functional score by musculoskeletal tumor society score and complications according to the resection & reconstruction type and recycling methods. Results: The junctional union was obtained at 15.0 months in extracoporeal irradiation and 12.6 months in pasteurization. Also the mean radiologic union was shown at 6.0 months in fragmentary, 12.8 months in intercalary, 10 months in rAPC, 23.3 months in osteoarticular and 15.6 months in total joint. The functional score was 65.5% in fragmentary, 60.8% in intercalary, 62.8% in APC (except pelvis), 66.0% in osteoarticular and 66.6% in total joint. We have experienced 1 infection, 1 prutrusio acetabuli in pasteurization (18.1%) and other 22 complications (3 deep infections, 8 nonunions, 2 fractures, 2 epiphyseal problems, 5 joint instabilities, 2 local recurrence) in extracoporeal irradiation (46.8%). Also we have experienced 3 complications (3 nonunions) in intercalary (37.5%), 9 complications (4 nonunions, 1 deep infection, 1 periprosthetic fracture, 1 epiphyseal problem, 1 local recurrence, 1 protrusio acetabuli) in rAPC (50.0%), 6 complications (2 deep infections, 2 nonunions, 1 epiphyseal problem, 1 pathologic fracture) in osteoarticular (33.3%), 5 complications (5 joint instabilities) in total joint (100%) and 1 complication(1 local recurrence) in soft tissue (100%). Conclusion: In our experience, according to the resection & reconstruction type fragmentary and intercalary may have several advantages such as good radiologic and functional result and low rate of complication. And it seems that rAPC was available in case which have no sufficient residual bone stock. Also the pasteurization may have more advantages than that of the extracorporeal irradiation.

• Journal of Intelligence and Information Systems
• /
• v.18 no.3
• /
• pp.185-202
• /
• 2012

.Since the value of information has been realized in the information society, the usage and collection of information has become important. A facial expression that contains thousands of information as an artistic painting can be described in thousands of words. Followed by the idea, there has recently been a number of attempts to provide customers and companies with an intelligent service, which enables the perception of human emotions through one's facial expressions. For example, MIT Media Lab, the leading organization in this research area, has developed the human emotion prediction model, and has applied their studies to the commercial business. In the academic area, a number of the conventional methods such as Multiple Regression Analysis (MRA) or Artificial Neural Networks (ANN) have been applied to predict human emotion in prior studies. However, MRA is generally criticized because of its low prediction accuracy. This is inevitable since MRA can only explain the linear relationship between the dependent variables and the independent variable. To mitigate the limitations of MRA, some studies like Jung and Kim (2012) have used ANN as the alternative, and they reported that ANN generated more accurate prediction than the statistical methods like MRA. However, it has also been criticized due to over fitting and the difficulty of the network design (e.g. setting the number of the layers and the number of the nodes in the hidden layers). Under this background, we propose a novel model using Support Vector Regression (SVR) in order to increase the prediction accuracy. SVR is an extensive version of Support Vector Machine (SVM) designated to solve the regression problems. The model produced by SVR only depends on a subset of the training data, because the cost function for building the model ignores any training data that is close (within a threshold ${\varepsilon}$) to the model prediction. Using SVR, we tried to build a model that can measure the level of arousal and valence from the facial features. To validate the usefulness of the proposed model, we collected the data of facial reactions when providing appropriate visual stimulating contents, and extracted the features from the data. Next, the steps of the preprocessing were taken to choose statistically significant variables. In total, 297 cases were used for the experiment. As the comparative models, we also applied MRA and ANN to the same data set. For SVR, we adopted '${\varepsilon}$-insensitive loss function', and 'grid search' technique to find the optimal values of the parameters like C, d, ${\sigma}^2$, and ${\varepsilon}$. In the case of ANN, we adopted a standard three-layer backpropagation network, which has a single hidden layer. The learning rate and momentum rate of ANN were set to 10%, and we used sigmoid function as the transfer function of hidden and output nodes. We performed the experiments repeatedly by varying the number of nodes in the hidden layer to n/2, n, 3n/2, and 2n, where n is the number of the input variables. The stopping condition for ANN was set to 50,000 learning events. And, we used MAE (Mean Absolute Error) as the measure for performance comparison. From the experiment, we found that SVR achieved the highest prediction accuracy for the hold-out data set compared to MRA and ANN. Regardless of the target variables (the level of arousal, or the level of positive / negative valence), SVR showed the best performance for the hold-out data set. ANN also outperformed MRA, however, it showed the considerably lower prediction accuracy than SVR for both target variables. The findings of our research are expected to be useful to the researchers or practitioners who are willing to build the models for recognizing human emotions.

• Journal of Intelligence and Information Systems
• /
• v.19 no.4
• /
• pp.123-132
• /
• 2013

As the smartphones are equipped with various sensors such as the accelerometer, GPS, gravity sensor, gyros, ambient light sensor, proximity sensor, and so on, there have been many research works on making use of these sensors to create valuable applications. Human activity recognition is one such application that is motivated by various welfare applications such as the support for the elderly, measurement of calorie consumption, analysis of lifestyles, analysis of exercise patterns, and so on. One of the challenges faced when using the smartphone sensors for activity recognition is that the number of sensors used should be minimized to save the battery power. When the number of sensors used are restricted, it is difficult to realize a highly accurate activity recognizer or a classifier because it is hard to distinguish between subtly different activities relying on only limited information. The difficulty gets especially severe when the number of different activity classes to be distinguished is very large. In this paper, we show that a fairly accurate classifier can be built that can distinguish ten different activities by using only a single sensor data, i.e., the smartphone accelerometer data. The approach that we take to dealing with this ten-class problem is to use the ensemble of nested dichotomy (END) method that transforms a multi-class problem into multiple two-class problems. END builds a committee of binary classifiers in a nested fashion using a binary tree. At the root of the binary tree, the set of all the classes are split into two subsets of classes by using a binary classifier. At a child node of the tree, a subset of classes is again split into two smaller subsets by using another binary classifier. Continuing in this way, we can obtain a binary tree where each leaf node contains a single class. This binary tree can be viewed as a nested dichotomy that can make multi-class predictions. Depending on how a set of classes are split into two subsets at each node, the final tree that we obtain can be different. Since there can be some classes that are correlated, a particular tree may perform better than the others. However, we can hardly identify the best tree without deep domain knowledge. The END method copes with this problem by building multiple dichotomy trees randomly during learning, and then combining the predictions made by each tree during classification. The END method is generally known to perform well even when the base learner is unable to model complex decision boundaries As the base classifier at each node of the dichotomy, we have used another ensemble classifier called the random forest. A random forest is built by repeatedly generating a decision tree each time with a different random subset of features using a bootstrap sample. By combining bagging with random feature subset selection, a random forest enjoys the advantage of having more diverse ensemble members than a simple bagging. As an overall result, our ensemble of nested dichotomy can actually be seen as a committee of committees of decision trees that can deal with a multi-class problem with high accuracy. The ten classes of activities that we distinguish in this paper are 'Sitting', 'Standing', 'Walking', 'Running', 'Walking Uphill', 'Walking Downhill', 'Running Uphill', 'Running Downhill', 'Falling', and 'Hobbling'. The features used for classifying these activities include not only the magnitude of acceleration vector at each time point but also the maximum, the minimum, and the standard deviation of vector magnitude within a time window of the last 2 seconds, etc. For experiments to compare the performance of END with those of other methods, the accelerometer data has been collected at every 0.1 second for 2 minutes for each activity from 5 volunteers. Among these 5,900 ($=5{\times}(60{\times}2-2)/0.1$) data collected for each activity (the data for the first 2 seconds are trashed because they do not have time window data), 4,700 have been used for training and the rest for testing. Although 'Walking Uphill' is often confused with some other similar activities, END has been found to classify all of the ten activities with a fairly high accuracy of 98.4%. On the other hand, the accuracies achieved by a decision tree, a k-nearest neighbor, and a one-versus-rest support vector machine have been observed as 97.6%, 96.5%, and 97.6%, respectively.

• Journal of The Korean Society of Grassland and Forage Science
• /
• v.40 no.2
• /
• pp.98-105
• /
• 2020

This experiment was conducted to a comparison of the productivity according to variety and forage quality by plant parts of imported silage corn (Zea mays, L) in Pyeongchang. The corns evaluated in this experiment were 8 varieties (P1184, P1151, P1194, P1543, P1345, P1429, P1443, and P2105) introduced from the United States, Pioneer Hybrid Co. The harvested corn was divided into 5 plant parts (leaf, stem, cob, husk, and grain), and the ratio of each part was calculated using dry weight and the feed value was analyzed. The emergence rate of corn was generally good except for the P1151 and P2105 varieties. The average tasseling date was July 24^th and the silking date was July 27^th, but the P2105 variety was late to July 28^th and August 1^st, and the remaining varieties were similar. P1345 was the highest (289 and 123 cm), and P1151 varieties were the lowest (267 and 101 cm) in the plant and ear height. Disease resistance was low in P1184, P1443 and P1429, and P1197 and P1345 were high. In the case of stover, the dry matter (DM) content was the lowest at 19.6% in the P1151 and the highest at 24.9% in the P1429. DM content of ear was the highest in the P2105 (55.5%), and P1184 (54.2%) and P1345 (54.3%) were also significantly higher (p<0.05). The DM yield of stover of P2105, P1429 and P1194 varieties was significantly higher (p<0.05), and ear yield of P2105, P1345 and P1443 was higher. The proportions of each part of plants (leaf, stem, cob, husk, and grain) divided by 5 was high, with 50-60% of the ear(grain+cob) ratio. The ratio of husk and cob was roughly similar, and the leaf and stem part showed a ratio of about 20%. The crude protein (CP) content was highest in leaf, followed by grain. The CP content of the stem was the lowest, and the husk was not significantly different among the varieties (p>0.05). The acid detergent fiber (ADF) content was similar to the rest parts except grain, but the leaf part tended to be lower, and other parts except the stem and leaf showed no significant difference between varieties (p>0.05). There was no significant difference in NDF (neutral detergent fiber) content in husk, but there was a difference between varieties in other parts (p<0.05). In addition, there was a special difference by plant parts for each variety, P2015 on the stem, P1197 on the leaf, P1151 on the cob, P1197 on the husk, and P1197 on the grains with high NDF content. IVDMD (in vitro dry matter digestibility) was not significantly different between stems and grains, but there was a difference between varieties in cobs and husks. According to the results, DM yield of P2105 variety was the best in the experiment, and the ratio of grain was excellent in P1543 and P1345. In addition, it was found that the feed value was higher in the leaves and grains, and the leaf and stem had higher feed values than husk or cob.

• Journal of the Korean Institute of Traditional Landscape Architecture
• /
• v.36 no.3
• /
• pp.115-127
• /
• 2018

This study was attempted to find out the value of cultural assets through the clear diagnosis and prescription of the dead and weakness factors of the Population of Retusa Fringe Trees in Pyeongji-ri, Jinan(Natural Monument No. 214), The results are as follows. First, Since the designation of 13 natural monuments in 1968, since 1973, many years have passed since then. In particular, despite the removal of some of the buried soil during the maintenance process, such as retreating from the fence of the primary school after 2010, Second, The first and third surviving tree of the designated trees also have many branches that are dead, the leaves are dull, and the amount of leaves is small. vitality of tree is 'extremely bad', and the first branch has already been faded by a large number of branches, and the amount of leaves is considerably low this year, so that only two flowers are bloomed. The second is also in a 'bad'state, with small leaves, low leaf density, and deformed water. The largest number 1 in the world is added to the concern that the s coverd oil is assumed to be paddy soils. Third, It is found that the composition ratio of silt is high because it is known as '[silty loam(SiL)]'. In addition, the pH of the northern soil at pH 1 was 6.6, which was significantly different from that of the other soil. In addition, the organic matter content was higher than the appropriate range, which is considered to reflect the result of continuous application for protection management. Fourth, It is considered that the root cause of failure and growth of Jinan pyeongji-ri Population of Retusa Fringe Trees group is chronic syndrome of serious menstrual deterioration due to covered soil. This can also be attributed to the newly planted succession and to some of the deaths. Fifthly, It is urgent to gradually remove the subsoil part, which is estimated to be the cause of the initial damage. Above all, it is almost impossible to remove the coverd soil after grasping the details of the soil, such as clayey soil, which is buried in the rootstock. After removal of the coverd soil, a pestle is installed to improve the respiration of the roots and the ground with Masato. And the dead 4th dead wood and the 5th and 6th dead wood are the best, and the lower layer vegetation is mown. The viable neck should be removed from the upper surface, and the bark defect should undergo surgery and induce the development of blindness by vestibule below the growth point. Sixth, The underground roots should be identified to prepare a method to improve the decompression of the root and the respiration of the soil. It is induced by the shortening of rotten roots by tracing the first half of the rootstock to induce the generation of new roots. Seventh, We try mulching to suppress weed occurrence, trampling pressure, and soil moisturizing effect. In addition, consideration should be given to the fertilization of the foliar fertilizer, the injection of the nutrients, and the soil management of the inorganic fertilizer for the continuous nutrition supply. Future monitoring and forecasting plans should be developed to check for changes continuously.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.14 no.4
• /
• pp.1-10
• /
• 2011

This study was conducted to compare the growth of Zoysia japonica depending on different soil treatments in Saemangeum sea dike, which is filled with dredged soil. Zoysia japonica was planted using sod-pitching method on the control plot. On plots which were treated with forest soil and soil improvement, Zoysia japonica seeds were sprayed mechanically. Sixteen months after planting, coverage rate, leaf length, leaf width, and root length were measured and analyzed. Also, three Zoysia japonica samples per plot were collected to analyze nutrient contents. Coverage rate was 100% in B treatment plot(dredged soil+$40kg/m^3$ soil improvement+forest soil), in C treatment plots (dredged soil+$60kg/m^3$ soil improvement+forest soil), and D treatment plots (dredged soil+$60kg/m^3$ soil improvement), while only 43% of the soil surface was covered with Zoysia japonica on control plots. The width of the leaf on C treatment plots (3.79mm) was the highest followed by D treatment (3.49mm), B treatment (2.40mm) and control plots (1.97mm). Leaf and root length of D treatment was 30.18cm and 13.18cm, which were highest among different treatments. The leaf length of D treatment was highest followed by C, B, and A treatments. The root length of D treatment was highest followed by C, A, and B treatments. The nitrogen and phosphate contents of the above ground part of Zoysia japonica were highest in C treatment, followed by D, B, and A treatments. The nitrogen and phosphate contents of the underground part of Zoysia japonica were highest in D treatment, followed by C, A, and B treatments. C and D treatments showed the best results in every aspect of grass growth. The results of this study could be used to identify the cost effective way to improve soil quality for soil surface fixation on reclaimed areas using grass species.

• Journal of Intelligence and Information Systems
• /
• v.26 no.2
• /
• pp.105-129
• /
• 2020

This study uses corporate data from 2012 to 2018 when K-IFRS was applied in earnest to predict default risks. The data used in the analysis totaled 10,545 rows, consisting of 160 columns including 38 in the statement of financial position, 26 in the statement of comprehensive income, 11 in the statement of cash flows, and 76 in the index of financial ratios. Unlike most previous prior studies used the default event as the basis for learning about default risk, this study calculated default risk using the market capitalization and stock price volatility of each company based on the Merton model. Through this, it was able to solve the problem of data imbalance due to the scarcity of default events, which had been pointed out as the limitation of the existing methodology, and the problem of reflecting the difference in default risk that exists within ordinary companies. Because learning was conducted only by using corporate information available to unlisted companies, default risks of unlisted companies without stock price information can be appropriately derived. Through this, it can provide stable default risk assessment services to unlisted companies that are difficult to determine proper default risk with traditional credit rating models such as small and medium-sized companies and startups. Although there has been an active study of predicting corporate default risks using machine learning recently, model bias issues exist because most studies are making predictions based on a single model. Stable and reliable valuation methodology is required for the calculation of default risk, given that the entity's default risk information is very widely utilized in the market and the sensitivity to the difference in default risk is high. Also, Strict standards are also required for methods of calculation. The credit rating method stipulated by the Financial Services Commission in the Financial Investment Regulations calls for the preparation of evaluation methods, including verification of the adequacy of evaluation methods, in consideration of past statistical data and experiences on credit ratings and changes in future market conditions. This study allowed the reduction of individual models' bias by utilizing stacking ensemble techniques that synthesize various machine learning models. This allows us to capture complex nonlinear relationships between default risk and various corporate information and maximize the advantages of machine learning-based default risk prediction models that take less time to calculate. To calculate forecasts by sub model to be used as input data for the Stacking Ensemble model, training data were divided into seven pieces, and sub-models were trained in a divided set to produce forecasts. To compare the predictive power of the Stacking Ensemble model, Random Forest, MLP, and CNN models were trained with full training data, then the predictive power of each model was verified on the test set. The analysis showed that the Stacking Ensemble model exceeded the predictive power of the Random Forest model, which had the best performance on a single model. Next, to check for statistically significant differences between the Stacking Ensemble model and the forecasts for each individual model, the Pair between the Stacking Ensemble model and each individual model was constructed. Because the results of the Shapiro-wilk normality test also showed that all Pair did not follow normality, Using the nonparametric method wilcoxon rank sum test, we checked whether the two model forecasts that make up the Pair showed statistically significant differences. The analysis showed that the forecasts of the Staging Ensemble model showed statistically significant differences from those of the MLP model and CNN model. In addition, this study can provide a methodology that allows existing credit rating agencies to apply machine learning-based bankruptcy risk prediction methodologies, given that traditional credit rating models can also be reflected as sub-models to calculate the final default probability. Also, the Stacking Ensemble techniques proposed in this study can help design to meet the requirements of the Financial Investment Business Regulations through the combination of various sub-models. We hope that this research will be used as a resource to increase practical use by overcoming and improving the limitations of existing machine learning-based models.

• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.16 no.4
• /
• pp.223-237
• /
• 2011

The intermittent freshwater discharge has an critical influence upon the biophysical environments and the ecosystems of the Yeongsan Estuary where the estuary dam altered the continuous mixing of saltwater and freshwater. Though freshwater discharge is controlled by human, the extreme events are mainly driven by the heavy rainfall in the river basin, and provide various impacts, depending on its magnitude and frequency. This research aims to evaluate the magnitude and frequency of extreme freshwater discharges, and to establish the magnitude-frequency relationships between basin-wide rainfall and freshwater inflow. Daily discharge and daily basin-averaged rainfall from Jan 1, 1997 to Aug 31, 2010 were used to determine the relations between discharge and rainfall. Consecutive daily discharges were grouped into independent events using well-defined event-separation algorithm. Partial duration series were extracted to obtain the proper probability distribution function for extreme discharges and corresponding rainfall events. Extreme discharge events over the threshold 133,656,000 $m^3$ count up to 46 for 13.7y years, following the Weibull distribution with k=1.4. The 3-day accumulated rain-falls which occurred one day before peak discharges (1day-before-3day -sum rainfall), are determined as a control variable for discharge, because their magnitude is best correlated with that of the extreme discharge events. The minimum value of the corresponding 1day-before-3day-sum rainfall, 50.98mm is initially set to a threshold for the selection of discharge-inducing rainfall cases. The number of 1day-before-3day-sum rainfall groups after selection, however, exceeds that of the extreme discharge events. The canonical discriminant analysis indicates that water level over target level (-1.35 m EL.) can be useful to divide the 1day-before-3day-sum rainfall groups into discharge-induced and non-discharge ones. It also shows that the newly-set threshold, 104mm, can just separate these two cases without errors. The magnitude-frequency relationships between rainfall and discharge are established with the newly-selected lday-before-3day-sum rainfalls: $D=1.111{\times}10^8+1.677{\times}10^6{\overline{r_{3day}}$, (${\overline{r_{3day}}{\geqq}104$, $R^2=0.459$), $T_d=1.326T^{0.683}_{r3}$, $T_d=0.117{\exp}[0.0155{\overline{r_{3day}}]$, where D is the quantity of discharge, ${\overline{r_{3day}}$ the 1day-before-3day-sum rainfall, $T_{r3}$ and $T_d$, are respectively return periods of 1day-before-3day-sum rainfall and freshwater discharge. These relations provide the framework to evaluate the effect of freshwater discharge on estuarine flow structure, water quality, responses of ecosystems from the perspective of magnitude and frequency.

• The Journal of Korean Academy of Prosthodontics
• /
• v.46 no.1
• /
• pp.65-82
• /
• 2008

Statement of problem : Today's market economy has been changed more and more to consumer concerned. It is owing to not only consumers ' rising standard of living and education, but also purchasers' easy accessibilities to products through various mass media. The consumer centered market system, where customer can choose items with diverse alternatives to satisfy their self esteem, is also applied to the field of medical business, and accelerated by an increasing income level of shoppers and introducing the whole nations' medical insurance system. Today, the medical industry has become competitive due to increasing number of medical institutions and medical personnel, and this offers wide choices to consumers in the medical market place. At this point of time, it is essential to survey on the primary factor of gratification for the patient in the Dental clinic, as well as on the problems and suggestions in medical service. Purpose : The analysis in this study shows essential factors and expected influential elements in satisfaction of the patient in the Dental Hopsital, and strategic suggestions for the provider of dental service, which can be of benefit to the prospective customer as well as can make improvement in the quality of dental treatment service. Material and method : This study had been researched by collecting and analyzing the organized questionnaires, which were filled in directly from 784 patients, who visit Dental Hospital, Yonsei University in Seoul, from January 23rd to April 15th. Result : It can be summarized like the followings. 1. The social and demographical peculiarities of respondents are as follows. Samples of gender and marital status are adequately extracted, but data on occupation and treatment are are under a bias toward students, undergraduates and graduate students, and orthodontics. 2. 74% of patients who answer the questionnaire were highly satisfied with the service of dental clinic in the section of overall satisfaction. 3. The survey result about specific service of dental treatment, within sections of independent variables, is like the followings; Patients are highly gratified with service system, kindness, explanation, explanation on expected waiting hours, reservation system, emergency measures, expert treatment, existence of knowledge of dentistry, size of hospital, disinfection, equipment and parking, but lowly satisfied with expense of treatment, preparatory hours for treatment, waiting hours, treatment hours and the period of subscription. 4. The correlation analysis showed that there is no significant linear relationship between the independent variables. 5. The probit regression analysis showed that 8 out of 34 independent variables explained the dependent variables at the level of 0.01. 6. It shows that 8 independent variables, which can affect customers 'satisfaction, are clearing up of inconvenience, service system, kindness, explanation, treatment hours per attendance, reservation system, existence of knowledge of dentistry, and contentment of equipment in the hospital. Conclusion : The consumer's satisfaction totally relies on subjective evaluations of customers. Providing appropriate service, which can meet the criteria for the customer who demands various wares, pursues luxury goods, and expects high quality of medical service, is essential to fulfill patients' satisfaction. Many medical institutions do their best to satisfy their customer, touch their consumer, and offer patience centered services, and it is also applied to the field of dentistry. Establishing brand new strategic managements and elevating the quality of dental service based on this survey are required to improve the satisfaction of patience in the Dental Hospital.

• Radiation Oncology Journal
• /
• v.15 no.4
• /
• pp.369-377
• /
• 1997

Purpose : This study was Performed to identify the histopathologic feature by the reevaluation of the Pathologic specimen of the cervical tumors and to evaluate the clinical findings and the treatment results of the patients with small cell carcinoma of the cervix treated by radiotherapy. Materials and Methods : 2890 patients with cervical carcinoma received radiotherapy at the Department of Radiation Oncology. Yonsei Cancer Center, Yonsei University College of Medicine between October 1981 and April 1995. Of the 2890 patients in this data base, sixty were found to have small cell carcinomas $(2.08\%)$. Among them thirty six patients were transferred from other hospitals. the biopsy specimens of those Patients were not available. So we could review the slides of the other twenty four patients who were diagnosed at our hospital. Twenty four patients with small cell carcinoma of the cervix were analyzed retrospectively based on the assessment of H & E staining and other four immunohistochemical stains for neuroendocrine differentiation (neuron specific enolase, chromogranin. synaptophysin and Grimelius stain). And we also evaluate the Patients and tumor characteristics. response to radiation. patterns of failures, 5 year overall and disease free survival rates. Results : Thirteen tumors were neuroendocrine carcinomas(13/24 = $54.2\%$) and eleven tumors were squamous carcinomas, small cell type (11/24 = $47.8\%$) based on the assessment of H & E staining and other four neuroendocrine marker studies. So we classified the Patients two groups as neuroendocrine carcinoma and small cell type of squamous carcinoma, Among the 13 neuroendocrine carcinomas, five were well to moderately differentiated tumors and the other eight were Poorly differentiated or undifferentiated ones. The median age was 54 years old (range 23-79 years). Eight Patients had FIGO stage IB disease, 12 had stage 11, 3 had stage III and one had stage IV disease, Pelvic lymph node metastases were found in five Patients $(20.8\%)$. three of them were diagnosed by surgical histologic examination and the other two were diagnosed by CT scan. There was no difference between two histopathologic groups in terms of patients and tumor characteristics. response to radiation. 5 year overall and disease free survival rates. However the distant metastases rate was higher in neuroendocrine carcinoma Patients (6/13:$46.2\%$) than in small cell type of squamous carcinoma Patients (2/11:$18.2\%$), but there was no statistically significant difference because of the small number of patients (P>0.05). Conclusion : More than half of the small cell carcinoma of the cervix patients were neuroendocrine carcinoma (13/24 : $54.1\%$) by reevaluation of the biopsy specimen of the cervical tumors. The tendency of distant metastases of the neurolndocrine carcinoma was greater than those of the small cell type of squamous carcinoma $(46.2\%\;vs.\;18.2\%)$. But there were no differences in the patients and tumor characteristics and other clinical treatment results in both groups. These data suggest that radical local treatment such as radiotherapy or radical surgery combined with combination systemic cytotoxic chemotherapy might provide these patients with the best chance for cure.