• The Journal of Korean Society for Radiation Therapy
• /
• v.28 no.2
• /
• pp.149-159
• /
• 2016

Purpose : To evaluate the usefulness of the breathing exercise,we analyzed the change in the RPM signal and the diaphragm imagebefore 4D respiratory gated radiation therapy planning of lung cancer patients. Materials and Methods : The breathing training was enforced on 11 patients getting the 4D respiratory gated radiation therapy from April, 2016 until August. At the same time, RPM signal and diaphragm image was obtained respiration training total three steps in step 1 signal acquisition of free-breathing state, 2 steps respiratory signal acquisition through the guide of the respiratory signal, 3 steps, won the regular respiration signal to the description and repeat training. And then, acquired the minimum value, maximum value, average value, and a standard deviation of the inspiration and expiration in RPM signal and diaphragm image in each steps. Were normalized by the value of the step 1, to convert the 2,3 steps to the other distribution ratio (%), by evaluating the change in the interior of the respiratory motion of the patient, it was evaluated breathing exercise usefulness of each patient. Results : The mean value and the standard deviation of each step were obtained with the procedure 1 of the RPM signal and the diaphragm amplitude as a 100% reference. In the RPM signal, the amplitudes and standard deviations of four patients (36.4%, eleven) decreased by 18.1%, 27.6% on average in 3 steps, and 2 patients (18.2%, 11 people) had standard deviation, It decreased by an average of 36.5%. Meanwhile, the other four patients (36.4%, eleven) decreased by an average of only amplitude 13.1%. In Step 3, the amplitude of the diaphragm image decreased by 30% on average of 9 patients (81.8%, 11 people), and the average of 2 patients (18.2%, 11 people) increased by 7.3%. However, the amplitudes of RPM signals and diaphragm image in 3steps were reduced by 52.6% and 42.1% on average from all patients, respectively, compared to the 2 steps. Relationship between RPM signal and diaphragm image amplitude difference was consistent with patterns of movement 1, 2 and 3steps, respectively, except for No. 2 No. 10 patients. Conclusion : It is possible to induce an optimized respiratory cycle when respiratory training is done. By conducting respiratory training before treatment, it was possible to expect the effect of predicting the movement of the lung which could control the patient's respiration. Ultimately, it can be said that breathing exercises are useful because it is possible to minimize the systematic error of radiotherapy, expect more accurate treatment. In this study, it is limited to research analyzed based on data on respiratory training before treatment, and it will be necessary to verify with the actual CT plan and the data acquired during treatment in the future.

• Journal of Korean Ophthalmic Optics Society
• /
• v.13 no.4
• /
• pp.25-30
• /
• 2008

Purpose: In this paper, we have analyzed the problems of the Oh's report which is used to the basic data for supply and demand of medical technicians and studied a proposal for improvement to control system and supply and demand of korean optometrists. Methods: We have analyzed errors of Oh's report including supply and demand for medical technicians and management policy, expecting number for future optician, inaccurate estimation by limited data (employment rate, retirement rate, mortality rate) and an incorrect method of measurement for future supply and demand. Results: Oh's report showed the 18% error for estimation of supply which exclude the irregular entrance students. The estimation of supply was calculated by graduation rate 62.6% (college and University of Technology are 78.9% and 85.98% respectively), employment rate 65.8% (the average employment between 2002 and 2007 is 73.96%) and retirement rate is 2.3% (the retirement of pharmacists is 1.3%) but it showed the significant differences to objective data. For estimate the suitable ratio of optometrists to the population, the ratio use of medical facilities by an age group was used, and suggested spectacle wearers 1,280 persons (populations 2,928 persons) per optometrist but the different from reference of Germany (4,706 persons), America (1,789 persons) and Korea (1,825 persons/an optometrist) are applied to estimation on supply. This report applied the low employment rate and argued that maintain the present situation, but claimed that utilize unemployment persons. The above result has induced double weighting effect on estimation of supply. Conclusions: To solve the related problems of supply and demand, we have to make a search for exact data and optimum application model, have to take an example of nation similar job category as Germany and the research result of the job satisfaction into consideration. After we get the integrated research result, we must carried out the policy with fairness and balance for the estimation of supply and demand. Therefore exact research is required prior to beginning policy establishment, government and related group have to make a clear long-term plan and permanent organization for medical technician to establish supply and demand of medical technician.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.17 no.4
• /
• pp.193-204
• /
• 2022

This study analyzed the degree of influence of measurement and improvement of customer satisfaction level targeting kiosk users. In modern times, due to the development of technology and the improvement of the online environment, the probability that simple labor tasks will disappear after 10 years is close to 90%. Even in domestic research, it is predicted that 'simple labor jobs' will disappear due to the influence of advanced technology with a probability of about 36%. there is. In particular, as the demand for non-face-to-face services increases due to the Corona 19 virus, which is recently spreading globally, the trend of introducing kiosks has accelerated, and the global market will grow to 83.5 billion won in 2021, showing an average annual growth rate of 8.9%. there is. However, due to the unmanned nature of these kiosks, some consumers still have difficulties in using them, and consumers who are not familiar with the use of these technologies have a negative attitude towards service co-producers due to rejection of non-face-to-face services and anxiety about service errors. Lack of understanding leads to role conflicts between sales clerks and consumers, or inequality is being created in terms of service provision and generations accustomed to using technology. In addition, since kiosk is a representative technology-based self-service industry, if the user feels uncomfortable or requires additional labor, the overall service value decreases and the growth of the kiosk industry itself can be suppressed. It is important. Therefore, interviews were conducted on the main points of direct use with actual users centered on display color scheme, text size, device design, device size, internal UI (interface), amount of information, recognition sensor (barcode, NFC, etc.), Display brightness, self-event, and reaction speed items were extracted. Afterwards, using the questionnaire, the Kano model quality attribute classification of each expected evaluation item was carried out, and Timko's customer satisfaction coefficient, which can be calculated with accurate numerical values The PCSI Index analysis was additionally performed to determine the improvement priorities by finally classifying the improvement impact of the kiosk expected evaluation items through research. As a result, the impact of improvement appears in the order of internal UI (interface), text size, recognition sensor (barcode, NFC, etc.), reaction speed, self-event, display brightness, amount of information, device size, device design, and display color scheme. Through this, we intend to contribute to a comprehensive comparison of kiosk-based research in each field and to set the direction for improvement in the venture industry.

• Journal of Intelligence and Information Systems
• /
• v.23 no.2
• /
• pp.71-88
• /
• 2017

Language models were originally developed for speech recognition and language processing. Using a set of example sentences, a language model predicts the next word or character based on sequential input data. N-gram models have been widely used but this model cannot model the correlation between the input units efficiently since it is a probabilistic model which are based on the frequency of each unit in the training set. Recently, as the deep learning algorithm has been developed, a recurrent neural network (RNN) model and a long short-term memory (LSTM) model have been widely used for the neural language model (Ahn, 2016; Kim et al., 2016; Lee et al., 2016). These models can reflect dependency between the objects that are entered sequentially into the model (Gers and Schmidhuber, 2001; Mikolov et al., 2010; Sundermeyer et al., 2012). In order to learning the neural language model, texts need to be decomposed into words or morphemes. Since, however, a training set of sentences includes a huge number of words or morphemes in general, the size of dictionary is very large and so it increases model complexity. In addition, word-level or morpheme-level models are able to generate vocabularies only which are contained in the training set. Furthermore, with highly morphological languages such as Turkish, Hungarian, Russian, Finnish or Korean, morpheme analyzers have more chance to cause errors in decomposition process (Lankinen et al., 2016). Therefore, this paper proposes a phoneme-level language model for Korean language based on LSTM models. A phoneme such as a vowel or a consonant is the smallest unit that comprises Korean texts. We construct the language model using three or four LSTM layers. Each model was trained using Stochastic Gradient Algorithm and more advanced optimization algorithms such as Adagrad, RMSprop, Adadelta, Adam, Adamax, and Nadam. Simulation study was done with Old Testament texts using a deep learning package Keras based the Theano. After pre-processing the texts, the dataset included 74 of unique characters including vowels, consonants, and punctuation marks. Then we constructed an input vector with 20 consecutive characters and an output with a following 21st character. Finally, total 1,023,411 sets of input-output vectors were included in the dataset and we divided them into training, validation, testsets with proportion 70:15:15. All the simulation were conducted on a system equipped with an Intel Xeon CPU (16 cores) and a NVIDIA GeForce GTX 1080 GPU. We compared the loss function evaluated for the validation set, the perplexity evaluated for the test set, and the time to be taken for training each model. As a result, all the optimization algorithms but the stochastic gradient algorithm showed similar validation loss and perplexity, which are clearly superior to those of the stochastic gradient algorithm. The stochastic gradient algorithm took the longest time to be trained for both 3- and 4-LSTM models. On average, the 4-LSTM layer model took 69% longer training time than the 3-LSTM layer model. However, the validation loss and perplexity were not improved significantly or became even worse for specific conditions. On the other hand, when comparing the automatically generated sentences, the 4-LSTM layer model tended to generate the sentences which are closer to the natural language than the 3-LSTM model. Although there were slight differences in the completeness of the generated sentences between the models, the sentence generation performance was quite satisfactory in any simulation conditions: they generated only legitimate Korean letters and the use of postposition and the conjugation of verbs were almost perfect in the sense of grammar. The results of this study are expected to be widely used for the processing of Korean language in the field of language processing and speech recognition, which are the basis of artificial intelligence systems.